STEEL COMMITTEE OF CAUFORNIA

TECHNICAL INFORMATION & PRODUCT SERVICE

NOVEMBER 1990

Design of Small Base Plates for Wide Flange Columns*W. A. THORNTON

The 9th Edition• of the AISC Manual of Steel Construc-tion uses the Murray-Stockwell2 method for analysis ofsmall base plates, i.e., plates that are only slightly larger thanthe column depth d and width bf. It combines this methodwith the cantilever method of the 8th3 and earlier editionsfor large base plates. The Murray-Stockwell method assumesa bearing pressure of Ft,, the maximum permitted, over anH-shaped contact area under the column cross-sectionbetween the plate and the concrete. The cantilever method,on the other hand, assumes a uniform bearing pressure, fp< Fp, over the entire base plate surface of area BxN(Fig. 1). Thus, the two methods assume very different bear-ing pressure distributions and are difficult to combine intoa single method.

A solution to this dilemma is to return to the 8th Editionassumption of uniform pressure between the base plate andthe concrete. This assumption is conservative with respectto the base plate thickness determination because the truepressure distribution will be less near the plate edges andmore under the column cross-section, which cross-sectionalso provides support for the plate at its top surface. Sincethe plate is assumed more heavily loaded distant from its

A. Thornton, PhD, PE, is chief engineer, Cives Steel Com-pany, Roswefi, GA, and is chairman of AISC Committee onManual, Textbooks, and Codes.

supports than it will be, a plate thickness determined underthis load will be thicker than it needs to be.

To supplement the cantilever method for large base plates,which is actually a yield line method, it is consistent againto use yield line theory applied to the portion of the baseplate contained within the column depth and width. Hap-pily, exact solutions to this problem are available in the liter-ature.4 Consider Fig. 2, which shows a plate supported onthree edges and free on the fourth. The dimensions of theplate are taken as the column depth d and the half columnwidth bfi2, rather than the more correct d - 2tf and (bf -t,.)/2. This is done for simplicity and is conservative. If thethree supported edges are taken as completely fixed, i.e.,no displacement and no rotation about an axis parallel to eachedge, the required base plate thickness with a factor of safetyof 2 is

tp = o.t,j (1)

whereft, = uniform pressure between base plate and concrete

= P/BxN, ksiF.,. -- yield stress of base plate, ksi

G ,, f 3 G - l•--6-G- +I'•

where r/ = d/bf

Reproduced from AISC Engineering Journal, Volume 27, No. 3, 3rd Quarter 1990

dIOr

/

m

.95Or

m ,

NUnsuppo•ed

Edge

b

SuppoSedEdge

Fig. I. Column base plate geometry and symbols (from AISC'). Fig,. 2. Small base plate geometry and support conditions.

The expression for et given in Eq. 2 can be approximated by

et = 4 • (3)

with an error of -2.97 % (unconservative) to +6.00% (con-servative) in the range of ?7 from • to 3. Then, Eq. I becomeswith Eq. 3

1where • has been replaced by • with an error of 2%.

Combining Eq. 4 with the cantilever method for large baseplates, let

n'= (5)

and

I = max(m,n,n) (6)

where m and n are defined in Fig. 1. Then the required platethickness is

tr = 2 t J (7)

If the base plate is small with N • d, it may be unconser-vative to assume complete fixity of the base plate to the col-umn flanges. If the plate of Fig. 2 is completely fixed to thecolumn web along the side of length d but simply supported,i.e., no displacement but rotation unrestrained, along thesides of length bf/2, the required base plate thickness witha factor of safety of 2 is given by Eq. 1, with

•,2,•: + l,/--7•--lJ (8)

This expression for et can be approximated bY

et = '/2,J-• (9)

with an error of -0% (unconservative) and +t7.7% (con-servative) in the range of ,/from g to 3. In the more com-mon range of g _< ,/ < 2, the error is only +8.00% (con-servative). Using Eq. 9 in Eq. 1,

= 2(U•q•j) J (10)t,

Combining Eq. 10 with the cantilever method for large baseplates, let

n ' = i,• (11)

I = max(m,n,n') (12)

2

/Z_tp = 2 1 • (13)

The formulation for the two models just discussed can beseen to be exactly the same except for n'. Let the first for-mulation, for which n' = be referred to as Model 1and the second, with n' = 'Ax/db/be referred to as Model2. It will be instructive to see how these two models com-pare with a method suggested by Ahmed and Krepss andthe method of the AISC 8th Edition Manual. To this end,consider Table 1. The nine examples of this table show thatboth Models 1 and 2 produce plate thicknesses !ess than orequal to the method of the AISC 8th Edition. The methodof Ahmed and Kreps produces plate thicknesses betweenModels 1 and 2 for small base plates of square columns, buttends to produce plates too thick for nonsquare columns(T/ > 1), as seen from Examples 7, 8 and 9. In the case ofExamples 8 and 9, it produces plates thicker than the 8thEdition method.

Considering the results shown in Table 1, and recognizingthat Model 2 is clearly conservative while still producingplates thinner or at most as thick as the method of the AISC8th Edition Manual, it is recommended that Model 2, i.e.,

n' = I,•Xf'•f

I = max (m,n,n')

t, = 2l•,.

be used to replace the current AISC 9th Edition Manual baseplate design method for axial load.

The equivalent Load and Resistance Factor Design (LRFD)equation for base plate thickness is:

I,,[ 2Put, = •l 0.9F, BN (14)

whereP,, = total factored column load

NOTATION

The symbols used in this paper follow the usage of the AISCManual, 8th or 9th Edition.

REFERENCES

1. American Institute of Steel Construction, Manual of SteelConstruction, 9th Edition, 1989, pages 3-106 through3-110.

2. Murray, T. M., "Design of Lightly Loaded Column BasePlates;' AISC Engineering J., Volume 20, No. 4, 4thQuarter, 1983, pp. 143-152.

3. American Institute of Steel Construction, Manual of SteelConstruction. 8th Edition, 1980, pp. 3-99 through 3-102.

4. Park, R. and Gamble, W. L., Reinforced Concrete Slabs,Wiley, 1980, pp. 329-331.

5. Ahmed, S. and Kreps, R. R., "Inconsistencies in Col-umn Base Plate Design in the New AISC ASD (July 1989)Manual, AISC Engineering J., 3rd Quarter, 1990, pp.106-107.

6. DeWolf, J. T., and Ricker, D. T., Column Base Plates,AISC Steel Design Guide Series, No, 1, 1990, pp. 13-15.

7. Fling, R. S., "Design of Steel Bearing Plates/' AISCEngineering J., Volume 7, No. 2, 2nd Quarter, April 1970,pp. 37-40.

T a b l e 1.E x a m p l e s To C o m p a r e M e t h o d s (Fy = 36 ksi for al l c a s e s )

Data n'/tp(in.lin.)

Col. P d bt N B fp m n Mod. Mod. Ahmed & AISCExample Source Sect. (kips) (in.) (in.) (in.) (in.) (ksi) (in.) (in.) I 2 Kreps 8th Ed.

1. AISC Des. Guidea W 1 0 x l 0 0 200 11.10 10.34 11.5 11 1.58 .48 1.36 2.14 2.68 2.33 3.92.90 1.12 .98 1.64

2. Ahmed ,• Krepsb W12x106 331 12.89 12.22 14 13 1.82 .88 1.61 2.51 3.14 2.71 4.771.13 1.41 1.22 2.15

3. -- W12x106 300 12.89 12.22 14 13 1.65 .88 1.61 2.51 3.14 2.71 4.771.07 1.34 1.16 2.04

4. -- W12x106 300 12.89 12.22 16 16 1.17 1.88 3.11 2.51 3.14 2.71 4.771.12 1.13 1.12 1.72

5. AISC 8th Ed. W 1 0 x l 0 0 525 11.10 10.34 19 17 1.63 4.23 4.36 2.14 2.68 2.33 3.921.86 1.86 1.86 1.86

6. AISC8thEd. W12x106 600 12.89 12.22 18 16 2.08 2.88 3.11 2.51 3.14 2.71 4.771.50 1.51 1.50 2.29

7. Flingc 1 4 x 8 W F -- 14 8 -- -- .75 -- -- 2.12 2.65 2.94 3.68.61 .77 .85 1.06

8. -- W 2 4 x 6 8 · 450 23.73 8.965 24 9 2.08 -- -- 2.92 3.65 4.98 4.231.41 1.76 2.40 2.04

9. -- W36x160 1000 36.01 12.00 38 14 1.88 1.90 2.20 4.16 5.20 7.56 5.631.90 2.38 3.46 2.57

a. See Ref. 6b. See Ref. 5c. See Ref. 7, Fling gets tp = 0.711 in. for this example

I

3

Inconsistencies in Column Base Plate Designin the New AISC ASD Manual*SALAHUDDIN AHMED and ROBERT R. KREPS

The new AISC steel design manual (ninth edition)' sug-gests a new procedure for computing the thicknesses of col-umn base plates to rectify problems associated with the some-what conservative design approach adapted in its earlierversion. However, a close scrutiny of the suggested methodreveals that the new approach is sometimes overly conser-vative and even inconsistent.

Referring to Fig. 1 (pg. 3-106 of the AISC Manual),

P = Total column load, kipsAj = B x N = Area of plate, in.2

A, = Full cross-sectional area of concrete support, in.'Fh = Allowable bending stress in base plate, ksiv = Allowable bearing stress in support, ksi

fp = actual bearing pressure, ksif,' = Compressive strength of concrete, ksitp = Thickness of plate, in.

$alahuddin Ahmed, Ph.D., is structural engineer, LeonhardtKreps LeFevre, Toledo, Ohio.Robert R. Kreps, P.E., is principal, Leonhardt Kreps LeFevre,Toledo, Ohio.

Referring to page 3-108 of the Manual, the following proce-dure is followed to compute base plate size:

For a given P, f! and A2, minimum area of base plate iscomputed and reasonable values of B and N are selected.

Based on the column dimensions and selected B and N, quan-tities m and n are computed and the larger of the two con-trols. In the next step, the value of L (Fig. 2) is computedfrom the following expression, Fv = P/(2 + d + b -

2L)/L, which is quadratic in L. Solving for L,

L = [(d + b) _+ x/((d + /i)2 _ 4P/Fp)]/4

The Manual is silent as to which of the two solutions shouldbe used in further computation. However, a careful studyof the equation reveals that the smaller of the two L values

should be used. The required base plate thickness is thencomputed based on the larger of m and n calculated and thevalue of L, as described in the Manual.

The quantity L is computed based on an area with a pres-sure of Ft, and not fy. Thus it is not clear why fi, is used inthe expression t = Lx/(3fv/Fh) (pg. 3-107 of the Manual).

dllri

b

i i i l.8Ob.

4, '•'

N

, r D

r I

e L - - , T -

2L

I, I / l , - . , ×z [

Figure I Figure 2

· - Reproduced from AISC Engineering Journal, Volume 27, No. 3, 3rd Quarter 1990

4

Example

Let

P = 331 kipsColumn W12x106 (d = 12.89, b = 12.22)

f,'. = 3 ksiF,. = '36 ksiPier: 34 in. x 34 in.

1. A• = (1/(34 x 34))(331/(.35 x 3))" = 86

A• = 331/(.7 x 3) -- 158 Controls2. A = .5[.95 x 12.89 - .8 x 12.22] = 1.235

N = • + 1.235 = 13.8, use 14 in.

B = 158/14 = 11.3, use 13 in.A• = 14 x 13 = 182 in.-'

3. fv = 331/182 = 1.82 ksi4. m = [14 - 0.95 x 12.89]/2 = 0.88 in.

n = [13 - 0.80 x 12.22]/2 = 1.61 in. Controls

Fv = .35 x 3 x (34 x 34/182) = 2.65 ksi > 2.1 ksi,

use 2.1 ksiL =

= (25.11 - x/-•.036)/4 = 6.23 in.

tv = 1.61 x x/(1.82/(.25x36)) = 0.72 in.tt, = 6.23 x x/(3xl.82/27) = 2.80 in. Controls

It may be noted that the thickness of 2.80 in. is greaterthan what would be obtained according to the eighth edition

AISC Manual.Now if one repeated the same calculations with a load of

332 kips, L would become imaginary and as per the manualwould be ignored. As a result the required thickness wouldbe 0.72 in., less than that required for a lighter load.

Therefore the authors feel that the new way of computingL is basically inconsistent and likely to result in too thicka base plate when L controls and too thin a base plate whenL is imaginary and thus ignored.

SUGGESTED METHOD OF ANALYSIS

Let us assume that the pressure under the base plate is uni-form and is equal to P/Al. Let us also assume that the plate

is essentially fixed at the web and flanges of the column.Thus what we have here is a plate with one long and twoshort edges fixed and the fourth edge free with a uniform

load. One can go back to various moment coefficients avail-

able in the literature to compute maximum moment in theplate. Considering the width to length ratios of usual col-umn sections, the authors suggest a moment coefficient of

0.022 so that the maximum moment in the plate is 0.022 ×fp × d2 kip-in./in., where d is the depth of the column.

Sr•qa. = 0.022 x fp x d2/Fht = V(6S,e4a.) = V(O. 132fpd2/Fb) (1)

Therefore, to compute the base plate thickness,

a) Compute m and n as discussed in the Manual and select

a thickness based on the larger of the two.

b) Use the larger of the two thicknesses obtained in step

(a) and by Eq. 1.Applying this method to the example above,

a) Compute thickness to be 0.72 in. for the larger of mand n.

b) Use Eq. I for t = 4(0.132 × 1.82 × 12.892127)= 1.22 in. Controls

REFERENCES

1. American Institute of Steel Construction, Inc., Manualof Steel Construction, Allowable Stress Design, ninth edi-tion, Chicago, IL, July 1989.

2. Winter, Urquhart, O'Rourke, Niison, Design of ConcreteStructures, seventh edition, McGraw-Hill Book Company,New York.

This publication expresses the opinion of the author, and care has been taken to insurethat all data and information furnished are as accurate as possible. The author andpublisher cannot assume or accept any responsibility or liability for errors in the dataor information and 'in the use of such information.

The information contained herein is not intended to represent official attitudes, recom-mendations or policies of the Structural Steel Educational Council. The Council is notresponsible for any statements made or opinions expressed by contributors to thispublication.

5

STRUCTURAL STEEL EDUCATIONAL COUNCIL

TECHNICAL INFORMATION & PRODUCT SERVICE

JULY 1999

Practical Design and Detailing of Steel Column Base Plates

by

William C. Honeck Derek Westphal

Forell Elsesser Engineers, Inc.

No,

PRACTICAL DESIGN AND DETAILING OF STEEL COLUMN BASE PLATES

Table of Contents

Description Page No.

1.0

2.0

3.0

4.0

5.0

6.0

INTRODUCTION 2 1.1 Preface 2 1.2 Purpose 2 1.3 Organization 2

DESIGN GUIDELINES FOR MATERIALS AND FABRICATION 2.1 Materials

2.1.1 Anchor Bolts and Nuts 2.1.2 Plates

2.2 Base Plate Design for Fabrication 2.2.1 Material versus Labor 2.2.2 Welding 2.2.3 Base Plate Dimensions

DESIGN GUIDELINES RELATED TO ERECTION 3.1 Anchor Bolts

3.1.1 Anchor Bolt Position Mislocation 3.1.2 Rotated Anchor Bolt Patterns 3.1.3 Anchor Bolts Set Too Low or Too High 3.1.4 Columns Next to Walls

3.2 Washers 3.3 Base Plate Leveling

ENGINEERING GUIDELINES FOR DESIGN OF BASE PLATES 4.1 4.2 4.3 4.4 4.5 4.6 4.7

Design for Temporary Construction Loads Design for Gravity and Other Downward Loads Design for Gravity Loads in Combination with Uplift Loads Design for Gravity Loads in Combination with Shear Forces Design for Gravity Loads in Combination with Shear Forces and Moments Design for Moments due to Seismic Forces Architectural Issues

CONCLUSIONS

REFERENCES

3 3 3 3 4 4 4 5

8 8 9

10 10 12 13 14

15

15

PRACTICAL DESIGN AND DETAILING OF STEEL COLUMN BASE PLATES

1.0 INTRODUCTION

1.1 Preface

Steel column base plates are one of the most ~ndamental parts o f a steel structure, yet the design of base plates is commonly not given the attention that it should by engineers. This results in base plate details that are expensive, difficult to fabricate and may even contribute to the hazards of the steel erection process by not providing stability for erection loads applied to the column.

Base plates serve two basic fianctions:

1. They transfer column loads to the supporting member or foundation. These loads include axial due to gravity, moments, shears and sometimes axial due to uplift; 2. They allow the column to stand as a temporary vertical cantilever after the lifting line is released without having to guy off the column. The column and base plate must withstand temporary wind and erection loads safely.

Steel fabricators and erectors who are members of the Structural Steel Educational Council (SSEC) have commented that there are a variety of base plate designs and details from engineers. Some fabricators are critical of many of these designs because they are difficult to fabricate, or specify materials that are hard to obtain or that do not exist in the sizes specified. The designs often result in columns that are hard to erect or are unstable without guying the column. When anchor bolts are not properly set, expensive corrective work is required before the column can be erected, resulting in delays in the steel erection process. This publication of Steel Tips attempts to address these issues.

In order to understand better and respond to the fabrication and erection issues, a questionnaire was distributed to several SSEC member firms requesting their comments about problems

experienced in their shops during fabrication and in the field during steel erection. Specific issues included overly expensive designs and problems with obtaining the materials specified. Suggestions on how these designs could have been more economical were solicited. The questionnaire asked about ~teel erection problems experienced and requested suggestions to mitigate those problems. The responses received were very informative and many of the suggestions in the responses have been incorporated into this publication.

1.2 Purpose

The purpose of this issue of Steel Tips is to provide practical guidelines for engineers, fabricators and contractors regarding the design and detailing of steel column base plates. Guidance is provided toward resolving common design, fabrication and erection problems. Many of the topics discussed are simple to implement, yet are often overlooked.

Unfortunately the behavior of base plates in moment frames and braced flames subjected to earthquake forces is not fially understood. Research and code guidance are limited. The engineer is forced to use judgement in order to achieve a desired level of performance and it is hoped ,that this publication will initiate more research and development in the areas of base plate behavior and design guidelines for base plate assemblies that are subjected to high moments where some sort of yielding is necessary to achieve the desired performance.

1.3 Organization

The focus of this issue of Steel Tips is directed toward the practical aspects of the design and detailing of base plates particularly as they relate to economical fabrication and steel erection. Section 2.0 discusses fabrication issues. Section 3.0 discusses erection and anchor bolt placement

issues. Section 4.0 discusses the "issues" involved in the design of base plates, rather than providing "how to" design methods or guidelines, and lists the names of other authoritative publications where the reader can find design formulas and definitive procedures for design of base plates. Section 4.0 also discusses fixed and partially fixed column bases, for instance, moment frames which resist wind or earthquake forces.

2.0 D E S I G N G U I D E L I N E S F O R MATERIALS AND FABRICATION

Engineers have numerous types of steel to choose from when designing anchor bolts and base plate assemblies. However, materials are often specified that are not readily available or are not suitable for specific applications. Base plate details often are hard to fabricate, overly complicated, call for expensive welds and/or specify impossible welds. The following sections provide design guidelines for specifying suitable materials and suggestions for details to make fabrication easier and more economical.

2.1 Materials

According to the AISC Specification for Structural Steel Buildings Allowable Stress Design and Plastic Design (ASD Specifications), there are 16 ASTM designations specified for structural applications. For specific material properties, suitable applications and complete dimensional information, the reader should refer to the ASTM Specifications.

2.1.1 Anchor Bolts and Nuts

The most common and readily available anchor bolt materials are ASTM A36 and A307. Smaller bolts ge0erally are supplied in A307 and larger diameter in A36. The material properties for these relatively "low strength" bolts are very similar. These two grades are weldable and should be specified when possible.

When high-strength bolts are required, the materials typically available are A449, A354 and A193 type B7 (often referred to as "B7"). B7 bolts are the same material as AISI 4140 and can be substituted for A449 because A449 and B7 bolts both have material properties that are almost identical. A325 bolts only come in "headed" form, are limited to 1 1/2 inch diameter maximum and are limited in the lengths available. The properties and chemistry for A325 bolts are similar to A449 and B7. Generally, it is better to specify A449, A354 or B7 bolts when high-strength bolts are necessary. High-strength bolts come as plain bar stock and threads must be cut into both ends. Headed bolts fabricated from A325, A490 or A588 should not be specified since these are not readily available. All of these high strength materials are heat treated alloy steels and are therefore not suitable for welding. Before specifying a bolt material, contact local fabricators for information regarding material availability and review the ASTM standards for the grades being considered to determine their suitability.

It is important to specify the correct grade of nut that corresponds to the specified anchor bolt material. ASTM A563 specifies the various nut grades that are typically used in building construction and nuts suitable for use with the various grades of bolts (see Reference 4). The "Heavy Hex" nut style should be specified regardless of the nut grade that is selected. Footnote A below table X1.1 makes reference to ASTM A194 grade 2H as a substitute for A563 when certain sizes conforming to A563 are not available. A194 is a specification for pressure vessel and non-building uses, but the grades referenced in footnote A are suitable for use for anchor bolts in buildings.

2.1.2 Plates

The most common base plate materials are A36, A572 and A588. Fabricators responding to the questionnaire recommended that A36 material be specified if possible because it is the most readily available material. The table on the following page

contains material availability guidelines based on plate thickness.

Table 1 - Availability of Plate Material

Thickness (t) , , ,

t _< 4"

4" < t _< 6"

Plate Availability

A36 A572 Gr 42 or 50 A588 Gr 42 or 50

A36 A572 Gr 42 A588 Gr 42

t > 6" A36

2.2 Base Plate Design for Fabricat ion

Typically, except for very large columns with very heavy base plates, such as for high rise buildings, base plates are shop welded to the column. Unless the weld is a complete penetration, weld, the bottom end of the column needs to be cut square so that there will be full bearing where the column is in contact with the base plate. Some years ago, this was accomplished using milling machines in the shop. Today the cold sawing equipment used in most shops provides a column finished end with a maximum ANSI roughness height value of 500 which is satisfactory for contact bearing compression joints.

For very large columns, the base plate is erected first, using three leveling bolts around the perimeter of the base plate to level it, then the column is erected onto the base plate and connected using angles or other connection methods. The base plate is grouted before the column is erected. The mating surfaces should be prepared by milling or other means so that the column is in full contact with the base plate. Use of thick base plates can introduce welding problems .due to difficulty of meeting preheat requirements.

2.2.1 Material versus Labor

A common suggestion from steel fabricators for engineers to remember is that "material is cheap

relative to labor." If specifying thicker base plates will result in not having to add stiffener plates to the base plate, this will result in less labor to fabricate and will result in a more economic design. Adding stiffeners and other plates to a base plate assembly is labor intensive compared to using a thicker base plate that could eliminate the need for these additional stiffener plates.

2.2.2 Weld ing

The engineer should attempt to at least match the thickness of the base plate with the column flange thickness in order to prevent warping during welding, particularly if heavy welding, such as partial or complete penetration welds, is required to connect the column to the base plate. Thicker base plates without stiffeners are often more economical than using a thinner base plate with stiffeners. Stiffeners, if used, will have an impact on column finish dimensions. See Section 4.7 "Architectural Issues" for further discussion.

Another common suggestion from fabricators is to reduce weld sizes as much as possible (but account for minimum AWS weld sizes based on material thicknesses) and specify fillet welds in lieu of complete penetration welds where possible. Complete penetration welds require more labor due to the need to bevel the end of the column and fit up, and require extensive inspection. It is more economical to detail larger fillet welds, even if more weld metal is required for the fillet welds, as a substitute for partial penetration welds.

Fabricators have also pointed out that "all around" " welds should be avoided. Fillet welds that wrap around the flange toes (ends of column flanges) and the column web-to-flange fillets (the "k" region) can cause cracks due to high residual stresses in the welds. Such welds often require welding repair. Stop fillet welds 1/2 inch from these locations. See Figure 1 for clarification.

Welds should be detailed to account for clearances and access of welding equipment. Obviously the engineer should not show welds that are

impossible to access. For example, a common mistake is to specify "all around" welds at plate washers that are backed up against the column flange or web.

High strength bolts fabricated from high strength, heat treated steel (such as A354, A449 or B7) cannot be welded - not even tack welded - without adversely affecting the properties of these steels.

2.2.3 Base Plate Dimens ions

Where possible, the plate dimensions and bolt pattern of base plates should be symmetrical about both axes. This will preclude welding the base plate rotated 90 degrees from the correct orientation. Having a doubly symmetrical bolt pattern will also help avoid potential field problems (See Section 3.1.2).

The engineer should try to specify the same bolt hole diameter whenever possible to eliminate the need for multiple drill bit sizes. This also applies to any vent holes required to vent out air from under the larger base plates during the grouting operation.

Obviously the base plate dimensions should be sufficient to accommodate the column dimensions plus anchor bolt holes with sufficient dimensions to the column flanges and to the edge of the base plate. Also account for any square plate washers, if used. Several fabricators have stated that engineers sometimes erroneously assume their "typical" base plate detail will cover all conditions. Columns that are in different size groups require different base plate sizes. It is generally more economical to design a "typical" larger base plate to cover more than one column size in a column group (such as Wl0, W12, W14 groupings), than to design specific base plates for each column size. The fewer variations of base plates required will generally result in economy in fabrication even if more material is required. This is true because of the labor savings in shop drawing preparation and the different shop setups required for each variation in base plate configuration. It is also true that having fewer "different" anchor bolt

patterns will lead to less confiasion during anchor bolt placement. See Figure 1 on the following page for suggested details.

3.0 DESIGN GUIDELINES RELATED TO ERECTION

Anchor bolts and base plates should be designed and detailed to accommodate steel erection loads. Some simple, yet effecctive, attention to details and dimensions can go a long way in helping to prevent some common problems encountered during steel column erection. A previous edition of Steel Tips (Reference 7) contains usefial strategies for dealing with common field erection errors.

3.1 Anchor Bolts

Anchor bolt placement is obviously a difficult task but too often errors result due to poor quality control and quality assurance or lack of preparedness in the design. There are several ways to mislocate anchor bolts and typically one of the following will occur.

3.1.1 Anchor Bolt Posit ion Mis locat ion

Position mislocation is unfortunately a common problem. The horizontal location of the anchor bolts is often incorrect by as much as 1 to 2 inches. In some cases one of the anchor bolts is not in the correct location with respect to the remaining bolts and in other cases the entire layout is in the wrong location. There are several ways to avoid this problem during the design phase.

1. The best method for preventing anchor bolt mislocation is for the contractor to properly set and hold anchor bolts in the correct position for plan location and elevation. It is the contractors responsibility to set anchor bolts correctly within the tolerance given in the AISC Code of Standard Practice (Refer to Reference 3). A check by an independent surveyor will help locate misplaced bolts before steel is erected so that corrections can be made by the contractor before steel erection

! r~@@ V ~

_

WHEN REQUIRED @ <

GROUT Pi N

I / L ~, J ~

~ Tqq ~-,~ ~ ~ "~ ~L~ •:• • ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ , ~ ~ ~ - ~ . . . . . . ~

-

~ . f~C] ~ I<O>I ~ <C~>~ I~I L=VJ

~ v

SQUARE PLATE ~

_1. NO WELDTYP. ~ 2

GROUT HOLE IF REQUIRED ~ NO WELD TYP. ~

TOP WASHER PLATE rYP.

OVERSIZE HOLE ~

~ LEVELING NUT AND WASHER ~

ANCHOR BOLT ~ ~ Use square plate and hole pattem dimensions where possible to avoid problems assodated with mis-placed anchor bolts, rotated anchor

bolt patterns or plates that are accidentally rotated 90 degrees during fabrication.

~ Try to reduce numerous base plate variations by sizing typical plate based on the largest column in a size group (e.g. Wl0's, W12's or W14's). Reducing the number of variations will reduce the chance for error during erection and fabrication, and allow for simpler verification in the field. Provide maximum edge distance to bolt to allow base plate slotting if bolts are mislocated.

~ When additional bolts are required, add additional holes to make double symmetric bolt patterns. This is useful even if not all holes and bolts are needed. Four bolts is the suggested minimum for any base plate.

~:~ Anchor bolts should be at least 1" diameter. This is beneficial for erection safety and the anchor bolts are harder to accidentally bend in the field. Specify A307 or A36 material when possible. Both are easier to obtain and weldable.

~ Oversize holes in base plates should be used whenever possible to accommodate anchor bolt placement tolerances.

(~) Plate washers with field welds should be used in conjunction with oversize holes to resist nut pull-through and to transfer shear from the base plate to the anchor bolts. Special attention should be directed toward weld access. Plate washer should have hole 1/16" larger than bolt diameter. Welds may not be needed if the column is for "gravity only" and there are no shear forces at the base of the column.

~ Leveling nuts are recommended in lieu of leveling plates or shims for ease of construction, safety and efficiency.

~ The thickness of grout specified should accommodate the leveling nuts and be in proportion to the dimensions of the base plate (for example do not specify 3 inches of grout under a W6 column).

~ Specify an additional bolt extension above the top of the base plate to accommodate bolts that are set too low. Also specify extra threaded length to accommodate bolts set too high.

~ Specify fillet welds whenever possible. Partial penetration welds and complete penetration welds should only be specified when required.

~ Avoid specifying all-around welds. There should be no weld at the ends of the flanges and in the fillet (k region) of the column.

~ If a grout hole is needed, specify the same diameter as the anchor bolt holes to reduce the number of drill bit sizes required during fabrication.

FIGURE 1 - S U G G E S T E D BASE PLATE DETAILS

begins. This requirement should be included in the job specifications. In addition, the engineer should specify 1/8" sheet metal templates for every base plate. Typically contractors make one metal template and construct plywood copies of the template. This method of constructing templates and placing anchor bolts introduces several obvious opportunities for error.

Anchor bolts need to be rigidly held in position both top and bottom to prevent movement during concrete placement and to prevent the anchor bolts from tilting. Plates that connect the anchor bolts at the bottom should be considered, particularly for large anchor bolts.

2. Specify oversize bolt holes in the base plate with washer plates ("weld washers") that are field welded to the base plate (See Figure 1). The weld washer should have a standard hole (bolt diameter plus 1/16 inch). The AISC Code of Standard Practice allows the following oversized hole diameters,

Bolt Diameter i Oversize Hole Dia.

3/4" to 1" 5/16"

1" to 2" 1/2"

over 2" 1"

For larger bolts this may not be enough oversize allowance; a larger oversize of up to 2 inches would be better. Weld washers are necessary when using oversize holes to prevent nut pull- through and for shear transfer to the anchor bolt. The extra cost of the weld washers is small compared to the cost of making field corrections and erection delays due to misplaced anchor bolts.

3.1.2 Rotated Anchor Bolt Patterns

Anchor bolts with a non symmetrical pattern are sometimes turned 90 degrees from correct orientation. Detailing anchor bolt patterns with doubly symmetric patterns will prevent this problem. See Section 2.2.3.

3.1.3 Anchor Bolts Set Too Low or Too High

Specifying anchor bolts with extra bolt projection will help for anchor bolts set too low. The extra projection will also prevent the problem of nuts that do not have fiall thread engagement. If a fi~ll nut cannot be obtained, there are methods to extend the bolt length. Specifying A36 bolt material allows welding a stub onto the low bolt. Sometimes, the nut cavity above a low anchor bolt can be "filled out" with weld metal if weldable nut and bolt materials were specified.

Engineers should specify more of the bolt shank to be threaded than is actually needed. If the bolt is set high, the extra threads will allow the nut to be run down the bolt without requiring additional washers.

3.1.4 Columns Next to Walls

Another problem that frequently occurs is inaccessible anchor bolts due to a column located next to a wall. This occurs when the anchor bolts are located between the column flanges or at a comer where two walls intersect (See Figure 2). For these conditions, special base plate/anchor bolt patterns are necessary so that all anchor bolts are accessible. Refer to the ASD Manual, Connection Section, for assembly clearance requirements at nuts.

S WALL

¢

FIGURE 2 - INACCESSIBLE ANCHOR BOLT LOCATIONS

7

3.2 Washers

If high strength anchor bolts are "tensioned", hardened "cut" washers should be used in addition to any weld washer plates used. This will prevent the nuts from galling the weld washer or base plate. Normally, anchor bolts are not tensioned; nuts are usually tightened with a wrench using a "cheater".

3.3 Base Plate Leveling

Some erectors favor the use of leveling nuts instead of shim packs or leveling plates (See Figure 1), other erectors favor shim packs. Leveling nuts are easier to level and provide a more stable base for resisting erection loads than shim packs. Generally, leveling plates are reserved for special cases and should not be specified for typical use.

4.0 ENGINEERING GUIDELINES FOR DESIGN OF BASE PLATES

This section covers the engineering design of base plates. The focus of this section is not so much "how to" calculate base plates, but what the engineer needs to consider when engineering and detailing base plates. The reader is referred to other publications with formulas, design aids and procedures. See References 1, 2, 3, 5, and 6.

The base plate assembly must be designed to transfer all forces from the column to the supporting member whether it is a girder or a foundation. These forces include axial forces, shears and moments from the column. The magnitude and combinations of these loads will determine the design and details of the base plate. The simplest and most common condition encountered in practice is a column supporting gravity loads only. When there are high shear forces and moments, such as in a moment frame, the design becomes more complicated and the base plate and anchor bolts become heavier. Thd following subsections discuss the issues related to the various loads and combinations of loads.

4.1 Design for Temporary Construction Loads

The first fianction of a base plate is to temporarily support the column from overturning due to temporary wind, earthquake, and erection loads, and from the column getting bumped during erection until the,beams are attached to "tie in" the column. Therefore the base plates and anchor bolts need to be at least sufficient to resist the overturning moment and shear from these forces. Although erectors often check the column by assuming a one kip load applied horizontally at the top, this does not relieve the engineer from providing an adequate design.

If the anchor bolts and base plate are too small, f o r example, with only two anchor bolts or anchor

"bolts that are too close together, the base plate assembly may not be capable of resisting erection loading (See Figure 3).

1 ~ h

4

FIGURE 3 - UNSAFE ANCHOR BOLT CONFIGURATIONS

This can result in a dangerous condition during erection if the fabricator and erector have not checked the base plate assembly for erection loads and have not recognized that this condition exists. Either the base plate assembly must be enlarged by the fabricator during the shop drawing stage or the column will need to be guyed off or held with another lifting line and "tied in" before releasing the column. This process is expensive and it slows the erection progress. Engineers should recognize this and not undersize anchor bolts, make base plates too thin, or have anchor bolts too close together in the anchor bolt pattern. At least four

anchor bolts should be detailed and spread apart as much as possible. See Figure 1 for suggested details.

4.2 Design for Gravity and Other Downward Loads

The most common base plate condition is a base plate that transfers gravity loads to the supporting

member or foundation with relatively low shear forces and moments at the base of the column. These are "gravity only" columns that are not part o f moment frames or braced frames. The base plate must be large enough so that the area of the concrete beneath it is sufficient to support the loads. Usually these columns will transfer nominal shear and moments to the supporting member or foundation. Such forces are normally caused by story drift due to wind or earthquake loads.

The AISC Manual of Steel Construction (Reference 3) provides a two step procedure for the design of axially loaded base plates. First, the area of the plate is calculated based on the allowable beating stress defined by the following equations.

ASD: Fp ; 0.35 f/c A'-~ ~ 0.70 f/c

q A2 LRFD: % Pp = 0 . 8 5 tpc f'c A) ~'~ -< qL 1.7f'c A~

where, Fp = Allowable bearing stress (ksi) fo = Concrete compressive strength (ksi) A~ = Base plate area (in z) A2 = Area supporting base plate that is

geometrically similar (in z) q~o = 0.85 for compression Pp = Ultimate capacity of the concrete in

bearing

Based on this equation, the most efficient base plate area (A~) is at most one-fourth of the concrete support area (A2); or the concrete supporting area (A2) is ideally four times the base plate area (A 0.

II -]

El

0.80bf

d

0.80 D

~ 0.95 b ~

FIGURE 4 - CR IT ICAL B E N D I N G D I M E N S I O N S

Im I r

REBAR ~ DOWELS -~

• . ; STUDS OR LUGS

a) ANCHOR BOLTS WITH PLATE WASHERS

b) SHEAR KEY c) EMBEDDED SHEAR PLATE WITH WELDED SlOE PLATES

d) EMBEDDED SHEAR STRUT

FIGURE 5 - METHODS TO TRANSFER SHEAR FORCE FROM COLUMN TO FOUNDATION

The final step in determining the required base plate thickness is defined by calculating the flexural demand for a critical section of plate acting as a cantilever. For Allowable Stress Design (ASD) the elastic section modulus (S) is used; whereas for Load and Resistance Factor Design (LRFD) the plastic section modulus (Z) is used. The dimensions of the critical section are based on 0.95d and 0.8bf for wide flange sections, 0.80 times the outer diameter of pipes and 0.95 times the out-to-out dimension of tubes (See Figure 4). According to ASD, the allowable bending capacity is equal to 0.75Fy where Fy is the allowable yield strength of the steel. If LRFD is preferred, the design strength is equal to 0.90Fy. See References 1, 2, 3, 5 and 6 for useful design equations and design aids.

4.3 Design for Gravity Loads in Combination with Uplift Loads

When there are net uplift loads, such as can occur in the end columns of concentric or eccentric braced frames (CBF or EBF), the anchor bolts and base plates need to be checked and increased in size if necessary. When uplift loads are very high, it may be necessary to add stiffener plates welded to the column flanges and design longer

anchor bolts above the base plate to accommodate the stiffeners. However, it still may be more economical to use a thicker base plate than to add the stiffeners because of the high labor cost involved with the stiffeners. Anchor bolts need to be well embedded in the supporting foundation concrete to develop the tensile capacity of the anchor bolts, and to preclude anchor bolt pullout due to shear cone failure in the concrete. This detail becomes even more critical for braced frames or moment frames subjected to wind or earthquake forces where failure of the base plate assembly would cause overturning due to uplift resulting in loss of lateral resistance in the braced frame or moment frame. For earthquake loads, since actual loads are much higher than code design forces (which have been reduced to account for "ductility" in the braced frame), yielding should occur in the brace for a CBF or in the "link" beam for an EBF. The base plate assembly needs to be strong enough to ensure that yielding will occur in these other elements.

4.4 Design for Gravity Loads in Combination with Shear Forces

Taking section 4.3 a step further, i fa brace occurs at the base of a column, a high shear force is

10

introduced from the horizontal force component in the brace. This force must be resisted by the base plate assembly. There are various proposed methods to transfer this shear force:

1. Anchor Bolts (See Figure 5a);

. Shear Key - A steel shear key is welded to the bottom of the base plate to interlock with the concrete (See Figure 5b);

. Embedded Shear Plate - Shear plates are field welded to the sides of the base plate and to an embed plate that has welded shear studs or shear lugs to transfer shear forces to the concrete foundation (See Figure 5c);

. Embedded Shear Strut - A strut member with welded shear studs or shear lugs is connected to the base plate or a column gusset plate. The shear studs or shear lugs transfer the shear force into the slab concrete and then to the foundation through rebar dowels (See Figure 5d).The following is a discussion of the design issues pertaining to these methods of transferring shear at the base of a column.

1. Anchor bolts: When column shear forces are resisted by the anchor bolts, they must be checked for a combination of column shear, bending and tension. If oversize holes are used in the base plates for anchor bolt placement tolerance, welded washer plates must be added so that the base plate will not slip before engaging the anchor bolts. The washer plates are added to the top of the base plate and the additional bending in the anchor bolts must be accounted for due to the increased distance from the concrete to the washer plate. There is a practical limit to the amount of shear the anchor bolt/concrete interface can resist before the anchor bolts become very large. When shear fcrces are high, methods 2, 3 or 4 should be considered.

2. Shear Keys: Steel shear keys can be welded to the underside of the base plate to provide a shear interlock with the concrete foundation below. The bending and shear forces that the steel key imparts to the base plate must be accounted for. The use of such keys requires block-out voids to be formed in the top of the foundation to allow space for the keys and surrounding grout. Any rebars in the foundation under the base plate must be positioned vertically and/or horizontally to allow for the depression in the foundation concrete to accommodate the steel key. Shear keys are effective in transferring shear forces from a brace into the foundation, so that the anchor bolts only have to resist tension forces.

3. Side Plates; Another strategy would be to cast an embed plate into the top of the foundation. The embed plate would have shear studs or lugs welded to the bottom to transfer shear forces into the foundation. The embed plate would be larger than the base plate to accommodate setting tolerance and to accommodate side plates to transfer shear forces from the base plate to the embed plate and foundation. The column would be erected and leveled in the same manner as any conventional column. Loose plates would be added and field welded to the sides of the base plate and to the embed plate. Grouting between the base plate and embed plate would be the final step in the process. This detail is practical because it provides a template for the anchor bolts and allows for confinement of the grout.

4. Struts; When shear forces are high and shear keys or embedded plates are not practical for detailing reasons, steel struts can be added that are embedded into the slab concrete. The strut is welded or bolted to the base plate or to a stiffener or gusset plate welded to the base plate. The strut should have shear studs or lugs welded to it to transfer axial forces from the strut to the concrete slab. The slab adjacent to the strut should be doweled to the foundation to transfer forces from the slab to the foundation. Attention to construction details and sequencing is important so that the rebars around the strut do not interfere

11

with being able to position and connect the strut to the base plate.

4.5 Design for Gravity Loads in Combination with Shear Forces and Moments

When a base plate assembly must transfer column base moments to the foundation, the mechanism for resisting the moments is typically taken by the combination of the tensile capacity of the anchor bolts and the bearing capacity of the concrete or masonry. This forms a "couple" consisting of the tension force in the anchor bolts and the equivalent force at the centroid of the bearing area under the base plate. This is analogous to the internal forces to resist bending in a concrete "cracked" section. The other gravity, shear, and uplift forces acting in combination with the bending moment must also be added and accounted for.

Two methods are presented for consideration when designing base plates subject to bending moment. See References 1, 2, 3, 5 and 6 for additional information on how to calculate and design for this combination of loads.

The first method is based on the assumption that stresses caused by the moment are linear across the base plate length. The tensile force in the anchor bolt is dependent on the bearing area. An equation is provided in Reference 1 to calculate the effective length of bearing. (See Figure 6, Method 1). This may not be consistent with actual behavior since the assumption relies on linear deformation of the base plate.

The second method is based on the assumption that the resultant of the bearing length is directly beneath the column flange. The basis of this assumption is that the flange experiences a greater axial load compared to the web because of the higher combination of axial and flexural stresses, and the relative width of the column flange to the web (See Figure 6, Method 2). This method may produce inaccurate results as well since the bearing length may extend over to the anchor bolt in tension. More testing and research is required to confirm the validity of either method with actual

, ~ M

P+T /

METHOD 1

TENSION B FORCE IN BOLT

AXIAl. & FLEXURAL STRESS IN COLUMN

FORCE P I TRANSMITTED

~ ~ M TO BASE PLATE & ANCHOR BOLT

\

P÷T

METHOD 2

FIGURE 6 - COLUMN WITH AXIAL & FLEXURAL DEMAND

12

behavior. See References 1 and 5 for further explanation and useful equations.

When the column moments are known, the design is more straightforward than if the moments are more unpredictable, such as at the base of moment frame columns which resist earthquake forces where ductility becomes an important issue. See Section 4.6 for additional discussion of base plate assemblies that resist seismic forces.

4.6 Design for Moments due to Seismic Forces

Unfortunately, the behavior of base plates in moment frames and braced flames subjected to earthquake forces is not fully understood. Research and code guidance are limited. The engineer is forced to use judgment and the interpretation of the results from tests on assemblies with similar components in order to achieve a design that hopefully will have the desired level of performance.

Trying to fix, or partially fix, the base of moment frame or braced frame columns against rotation may be necessary to reduce the drift: in the story above the base plate location. Consider the following scenarios:

1. Continue the column into the foundation or into a basement level below;

2. Design a heavy base plate assembly strong enough to force a plastic hinge in the column. This is difficult to accomplish even for relatively small columns. The base plate and anchor bolts become very large and anchor bolt anchorage becomes difficult. The foundation must be capable of resisting the high moments from the column base assembly (See Figure 7a). Currently, research is ongoing at the University of Michigan by Professor Subhash Goel on base plate assemblies of this type;

3. If the steel frame is supported on spread footings with moment resisting grade beams between the footings or supported on a grade

beam grid system, partially fix the base of the column by designing the footing/grade beam system to form plastic hinges in the grade beams that behave in a ductile manner. Size the base plate assembly to develop the strength of the footing/grade beam considering the overstrength of the concrete sections.' Any plastic hinges should occur in the grade beams;

4. Design a partially fixed base plate assembly. This will help limit drift, and the base plate and anchor bolt dimensions will be more manageable than with a fixed base solution. Drift can be reduced dramatically because the column will be forced to bend in double curvature. The challenge is to design the base plate assembly to behave in a ductile manner. If partial fixity is lost during an earthquake due to the failure of the base plate or stretching or breakage of the anchor bolts, the drift of the first story will increase dramatically resulting in more damage and possible failure of the column. A failure of the second floor beams could also occur if they were not designed for the extra bending or are not ductile enough to accommodate the extra rotation. (See Figures 7b, 7c, and 7d);

5. "Pin" the base of the column by designing a base plate assembly that will have relatively little moment resistance, but will be ductile enough to accommodate the first story seismic drift:.

Some design issues relative to scenarios 4 and 5. will be discussed further. Scenarios 1, 2 and 3 are beyond the scope of this paper.

For all column to base plate welded connections, the same issues relative to beam to column connections in ductile moment frames should be considered to preclude a failure in or near the weld, particularly if a full plastic hinge in the column above the base plate is the desired design goal. The reader should refer to the documents and research currently being done by the SAC Joint Venture on moment frames (See Reference 8).

13

..co oM. j / \ ,.BAsE L

• • ~m m

~ " 1 ,.,. ,,: ...... : . ::.i.. : ~ :. ),:,: ~ (:..~.:i'~ :' ~! .::!! ~ LE~LING ~ ~ ":;~::~ ~ ~ - '~'

[ I i t Itk P TE /11 +

ANCHOR BOLT YIELD ZONE

a) HINGE IN COLUMN

b) BASE P~TE YIELDING

c) ANCHOR BOLT YIELDING BELOW BASE P~TE

FIGURE 7 - YIELDING MECHANISMS

~ - - - t - - ~ ] . ~ - ~ A N C H O R I I I I I I I I I 1 | BoLT ,' , , , , ', I.Y .%2

~ i:::: ~.~.~.. _~f.w~ ,:::

NO / I ' - LEVELING

6) ~NCNOR BOkT YEk~ING ~OV~ BASE P~TE

For.a partially fixed column base as described in scenario 4, there are two mechanisms to achieve the ductility in the base plate assembly:

a. Design the base plate to yield in bending by designing the anchor bolts to be strong enough to force plastic hinges in the base plate (See Figure 7b). The plate must be large enough, but not too thick so that a plastic hinge region can form between the column flanges and the anchor bolts without inducing a shear failure in the base plate. A leveling plate should be provided under the base plate to protect the grout while the base plate undergoes deformations during the cyclical bending excursions.

b. Design the base plate strong enough to force yielding in the anchor bolts (See Figures 7c and 7d). Nuts and washers must be used above and below the base plate (Figure 7c) or the top horizontal plate (Figure 7d) to force the anchor bolts to resist axial forces in both tension and compression so that there will be cyclic capacity and ductility in the assembly. There must be sufficient unbonded length in the bolts to allow for the required elongation without overstraining the bolts. The ultimate strength of the bolt must be high enough to preclude failure at the reduced

section in the threaded portion of the bolt before the bolt yields. This can become a problem when the ultimate strength of the bolt (F,) is too close to the yield strength (Fy). Some accommodation must be made so that the bolts will not buckle when they are in compression. De-bonding and buckling resistance can be provided by using pipe sleeves within the footing (Figure 7c). If the yielding portion of the anchor bolts is above the base plate, sleeves or "guides" can be provided to resist bolt buckling (Figure 7d).

Very few test results are available to validate the behavior of either mechanism described in a or b. More research and development into base plate behavior and design guidelines are needed for anchor bolt/base plate assemblies that are subjected to very high moments where yielding is necessary to achieve the desired performance.

4.7 Architectural Issues

Architectural issues should be considered when designing and detailing base plate assemblies. Anchor bolt assemblies need to fit within slab thicknesses. There needs to be sufficient distance between the top of foundation and the top of slab to accommodate grout, leveling nut and washer

14

plate below the base plate, the base plate, the washer plate, nut and bolt projection plus concrete cover above the top of bolt. Usually, this dimension is at least 12 inches. For projects with large columns and thick base plates, 12 inches is not enough. This is an important dimension to establish early because it affects the foundation depth.

Any stiffener plates added above the base plate must fit within the architectural finish around the column. If stiffeners are needed, the dimensions should be co-ordinated with the architect early in the design since it may be necessary to increase the finish dimensions, since this dimension will affect useable floor space. Also, the dimensions of any vertical stiffener plates should be checked to insure that the stiffener plates will not protrude above the slab outside of the column finish dimensions.

5.0 CONCLUSIONS

Base plates serve a critical role in transferring column loads to the foundation. This Steel Tips discusses design, fabrication and erection issues related to base plates and anchor bolts. Suggested details are presented and details to be avoided are shown. The engineer needs to be aware of materials available and should recognize that special attention to base plate and anchor bolt details can result in reduced costs during fabrication and erection. Base plate assemblies must be designed to accommodate temporary erection loads until the column is tied in with other structural members. Special attention by contractors when placing anchor bolts can reduce field problems and delays due to mislocated anchor bolts.

6.0 REFERENCES

1. Design of Welded Structures by Omer W. Blodgett, James F. Lincoln Arc Welding Foundation, 15 ~h Printing, 1996

2. AISC Manual of Steel Construction, LFRD Design, Volumes I and II, Second Edition, 1994

3. AISC Manual of Steel Construction, ASD Design, Ninth Edition, 1989

4. ASTM Standards in Building Codes, Volume 1, 35 'h Edition, 1998

5. Column Base Plates, Steel Design Guide Series 1 by John T. DeWolf and David T. Ricker, AISC 1990

6. AISC Engineering Journal "Beam-Column Base Plate Design - LRFD Method" by Richard M. Drake and Sharon J. Elkin, First Quarter 1999, Volume 36, Number 1.

7. Steel Tips, "Common Steel Erection Problems and Suggested Solutions" by James J. Putkey, SSEC publication, December 1993.

8. SAC Interim Guidelines: Evaluation, Repair, Modification and Design of Welded Steel Moment Frame Structures, FEMA 267, August 1995

More research and design guidelines are needed for base plate assemblies subjected to high bending moments, such as in moment frames subjected to earthquake forces. For partially fixed column base assemblies, mechanisms that must behave in a ductile manner are needed. Some alternative strategies and concepts are presented.

15

About the Authors

Bill Honeck, a Principal of Forell/Elsesser Engineers, Inc., has 34 years of experience in structural engineering, 9 of which were in structural steel design, fabrication and erection. This also included 2 years designing electric transmission towers and substation framing. In addition, several of his large-scale projects have been produced on fast-track.

From 1965 to 1974 Bill Honeck was project manager/engineer for Bethlehem Steel in the construction of numerous highrise and large scale structures. During these 9 years Bill Honeck had practical experience in structural steel highrise buildings and large bridges. His responsibilities included structural steel erection, cost estimating and erection engineering, coordinating jobs, scheduling, reviewing costs, and implementing savings where possible.

He worked in the field and office in connection with steel erection as a field engineer, and was in charge of erection engineering for the Western District from 1967 to 1974. He was also responsible for designing falsework and related structures, erection scheming, and checking structural integrity of steel framework for erection related loads.

Derek Westphal, a project engineer and analyst with Forell/Elsesser, began his career with the firm in early 1996. In his experience to date he has developed a strong background in the seismic retrofit of historic buildings as well as the new construction of office buildings, laboratory, and university facilities.

16

STRUCTURAL STEEL EDUCATIONAL COUNCIL

TECHNICAL INFORMATION & PRODUCT SERVICE

March 2001

Large Seismic SteelBeam-to-Column Connections

by

Egor P. Popov, ProfessorShakhzod M. Takhirov, Ph.D.

Pacific Earthquake Engineering Research Center (PEER)University of California, Berkeley

Professor Egor Paul Popov standing next to his last steel connection test specimen(February 2001).

Egor Paul Popov, Professor Emeritus of Civil and Environmental Engineering at theUniversity of California, Berkeley, passed away Thursday, April 19, 2001, after a briefillness.

Professor Popov began his engineering studies at the University of California, Berkeley,and continued with graduate work at the California Institute of Technology, theMassachusetts Institute of Technology, and Stanford University, where he obtained hisdoctorate degree in 1946. In his lengthy and illustrious career, he was called upon byNASA for his engineering expertise and played a key role in the structural analyses of theAlaskan Pipeline and the Oakland-San Francisco Bay Bridge. Popov joined theDepartment of Civil Engineering at UC Berkeley in 1946, and was active in teaching andresearch there for more than 50 years. His research interests covered a wide spectrum oftopics in earthquake engineering, including cyclic testing and modeling of structuralmembers; the development, research, and application of the eccentrically braced frame;research on the seismic resistance of steel connections and the development of improvedconnection details; and the development of friction devices to retrofit existing structures.He was appointed the first chairman of AISCD's Committee on Seismic Provisions forSteel Buildings and served in this position for several years. Elected to the NationalAcademy of Engineering in 1976, Professor Popov was honored in 1999 with theEarthquake Engineering Research Institute's highest honor, the George W. HousnerMedal.

Table of Contents

Abstract...........................................................................................ii1 Review of the Previous Research1.1 Introduction.....................................................................................11.2 Overview .........................................................................................11.3 Tension Tests on A490 1 ¼ Bolts............................................................11.4 Test Specimen Design and Detailing.......................................................22 Connection Design and Estimate Calculations2.1. Test specimens based on the proposed design...........................................22.2. Basic Parameters Used in the Connection Calculation................................22.2.1. Calculation of Plastic Hinge Location in the Beam.....................................32.2.2. Calculation of Probable Plastic Moment at the Hinges.................................32.2.3. Beam Shear Calculation....................................................................32.2.4. Calculation of the Moment at the Centerline of the Column..........................42.2.5. Check for Strong Column - Weak Beam Condition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .42.3. The Connection Details Calculations......................................................52.3.1. Calculation of the T-section Stem Thickness at the Weakest Section near theColumn Face. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .52.3.2. Calculation of the T-section Flange Size. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .62.3.3. Calculation of Bolt Size Between the T-section and the Column Flanges (Duringthe Design 1 ¼ inch High-Strength Bolts were Chosen)......................................72.3.4. Calculation of Weld Size to Beam Flanges for Both Specimens (3/4 inch FilletWeld Was Used). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .83 Experimental Program3.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .83.2 Test Specimens, Test Setup and Instrumentation.......................................93.2.1 Test Setup. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .93.2.2 Instrumentation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .95.2.3 Data Acquisition......................................................................................93.2.4 Loading History. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .103.2.5 Data Processing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .103.3 Test Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .113.3.1 Specimen 1.....................................................................................113.3.2 Specimen 2.....................................................................................124 Experimental Results and Conclusions4.1 Experimental results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .134.2 Conclusions: advantages and disadvantages of proposed connections............134.2.1 Advantages. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .134.2.2 Disadvantages. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .144.2.3 Future Research Directions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15

i

ABSTRACT

Two large bolted steel moment-resisting connections were studied by experiments. Theseconnections were single-sided beam-column assemblies that are representative of exteriorbeam-column connections, and they were composed of W36xl50 Grade 50 beams andW14x283 Grade 50 columns. T-sections were cut from W40x264 sections of Grade 50steel. The T-section stems were welded to the beams and pre-stressed by bolts to thebeam flanges in the shop. Final beam-to-column assembly required no additionalwelding: the T-section flanges were bolted to the column and the column shear tab wasbolted to the beam web. The specimens had two symmetrically located T-sections withthe difference in web geometry: the Specimen 1 had rectangular shape of stems, whereasthe Specimen 2 had U-shaped stems. During the cyclic testing the beam deformation wasminimal due to active participation of the T-section flanges: a separation between T-section flanges and the column flanges were observed. This separation was occurred duebending plastic deformation in the T-section flanges. This phenomenon allowed energydissipation and prevented the beam flanges and beam web from severe buckling.

ii

1 Review of the Previous Research

1.1 Introduction

The generally accepted detail of attaching steel beams to columns in seismic applicationsconsists of shear tabs attached to the column and direct welding of beam flanges with orwithout cover plates to column flanges. Numerous tests on this type of a connection weresupported by NSF with many specimens donated by the fabricators. The testing of suchspecimens was organized by SAC Joint Venture.

The moment capacity of such connections depends on cyclic endurance of flange weldsin both tension and compression. Under these conditions numerous tension weld failureswere observed both in the laboratory and the field. SAC has proposed six connections toavoid future weld failures. We propose another connection that avoids weld failures andit is field bolted and shop welded.

1.2 Overview

An attempt at the above approach on several end-plate connections was made by K.C.Tsai and E.P. Popov at University of California, Berkeley (1988 and 1990). An exampleof this kind of a connection is shown in Fig. 1-1. It is of interest to note that directwelding of a beam to a column stub shown Specimen 9 in Fig. 1-1 results in goodbehavior, but the erection is not generally practical. Specimen 10 in Fig. 1-1 with no ribsover the beam flanges did not give satisfactory results. Specimen 10R with ribs over thebeam flanges at the column stub behaved very well under cyclic loading as can be seen inFig. 1-2. Note the required large thickness of the end plate (a connection based on thedesign of Specimens 10 and 10R may require shims during assembling).

The above approach recently was also pursued by T.M. Murray and his associates in2000 at Virginia Polytechnic Institute (VPI) with good results. They achieved a numberof successful tests with W36xl50 beams. It appears that for larger or heavier beams theuse of ribs over beam flanges at columns would be required.

An extensive excellent study of bolted connections has been done at Georgia Institute ofTechnology by R. Leon and his associates in 2000. The work is very comprehensive, butis limited to small and medium size members.

The newly developed and tested connection at University of California, Berkeley issomewhat related to the end plate connection but is more versatile as it is more readilyadaptable to a larger range of heavier beams. The new connection depends on the use ofA490 1 ¼ bolts in tension in oversize round holes (as in the column flange and in the T-section flange as well) and shop fillet welds.

1.3 Tension Tests on A490 1 ¼ Bolts

l

In order to conduct ductility study on the A490 1 ¼ bolts two tests were performed. Forthe first test, a special device was built to test simultaneously the shank of the bolt and thethreaded part of the bolt below the nut. The actual failure occurred in the threaded region.The remarkable ductility of A490 bolt was clearly demonstrated and the load versuselongation diagram is presented on Fig. 1-3. Another experiment on a specimen ofconstant diameter machined from A490 bolt also showed excellent ductility. Figure 1-4shows stress versus strain diagram for this test.

1.4 Test Specimen Design and Detailing

One of newly developed connections using A490 1 ¼ bolts is shown in Fig. 1-5. Thedetails for the two specimens are shown in Fig. 1-6 and 1-7. In both cases the attachmentof a beams to a column is made using structural tees cut from W shapes (T-sections). Alarge choice of such sections is available. By rotating the beam all fillet welds can bedone in the shop in a down-hand position. Generous rounded fillets occur in all casesbetween a beam flange and the stem of the T-sections. Shop experience in fabricatingthese two specimens was very encouraging.

2 Connection Design and Estimate Calculations

2.1. Test specimens based on the proposed design.

The beams were fabricated from a W36xl50 section of A572-Gr.50 steel, the columnswere fabricated from a W14x283 section of A572-Gr.50 steel. The T-sections were madefrom W40x264 section of A572-Gr.50 steel. The geometrical properties of the beamsections used in the specimens are presented in Tables 2-1 though 2-3. They representwidely available data for standard rolled I-beams with W- shapes (see, for instance, AISC1995a).

The global dimensions and geometry of the specimens are shown in Fig. 1-5. Figures 1-6and 1-7 show the design details for the Specimen 1 and the Specimen 2. The materialproperties of the connections from mill certificate data are presented in Table 2-4. Thespecimens had two symmetrically located T-sections with the difference in the geometryof the stems: the Specimen 1 had rectangular shape of the stems, whereas the Specimen 2had U-shaped stems. The T-section stems were welded to the beams in the shop, and laterthe T-section flanges were bolted to the columns.

2.2. Basic Parameters Used in the Connection Calculation.

Nonlinear deformation of a building frame is typically accommodated through thedevelopment of inelastic flexural or shear stains within discrete regions of the structure.At large deformations these regions can develop into plastic hinges, which canaccommodate significant concentrated rotations without significant changes of the load,as shown in Fig. 2-1. This behavior is accompanied by significant energy dissipation,particularly by members involved in plastic deformation. The formation of plastic hingesin the beams is extremely desirable.

2

2.2.1. Calculation of Plastic Hinge Location in the Beam.

The probable location for the formation of the plastic hinges is a basic parameter for theconnection calculations. Figure 2-2 shows the suggested location of the plastic hinge. Thelocation depends on the type of connection and our design is close to a cover-plate type,therefore the plastic hinge can be developed at the following distance from the face of thecolumn (FEMA 267a):

Lh= dsts+ db/4,

Where:db is a depth of the beam, the value of the beam depth db is presented in the Table 2-1;dsts is a total depth of the T-section. This parameter is 28.75 inches for the Specimen 1and 20 inches for the Specimen 2, as it is shown in Table 2-3.

Therefore the plastic hinge is located at the following distance from the face of thecolumn:

Lh= 37.7 inches, for the Specimen 1 andLh= 29.0 inches, for the Specimen 2.

2.2.2. Calculation of Probable Plastic Moment at the Hinges.

The probable value of the plastic moment, Mpr, at the location of the plastic hinge shouldbe calculated from the equation, proposed in FEMA-267b:

Mpr=1.1ZbFyb.Where:

Fyb is the actual yield stress of the beam material, as identified from mill test reports inthe Table 2-4,Zb is the plastic modulus of the beam section determined from the Table 2-1.Therefore for the proposed design and for chosen material properties Zb = 581 inch3, andFyb = 56.6 ksi, the probable value of the plastic moment is as following:

Mpr= 36173 kip*inch.

2.2.3. Beam Shear Calculation.

The shear in the beam, at the location of the plastic hinge should be determined. Thelength of the arm at plastic hinge location is calculated from total beam length, Lb, minusthe distance of the hinge location, Lh. Therefore the shear at the plastic hinge Vp can bedetermined from the formula:

Vp=Mpr/(Lb-Lh).

3

Therefore the values of the shear at the location of the plastic hinge are

Vp= 375.7 kips, for the Specimen 1, andVp= 344.4 kips, for the Specimen 2.

2.2.4. Calculation of the Moment at the Centerline of the Column.

The moment at centerline of the column can be calculated as following, as it shown onFig. 2-3:

Mc = Mpr+ Vp (Lh + dc/2).

This expression is modified by using previous expression for the shear and the formulafor the moment calculation has the following form:

Mc = Vp (Lb + dc/2).

In our particular case the values of the moment at the centerline of the column are

Mc = 53485 kip*inch, for the Specimen 1, andMc = 49030 kip*inch, for the Specimen 2.

2.2.5. Check for Strong Column - Weak Beam Condition.

Buildings with the plastic hinges in the beams dissipate more energy than buildings withplastic hinges in the columns. Therefore, the connection in the building has to develop theplastic hinge in the beam rather than in the column. To determine if the desired strongcolumn - weak beam condition exists, the connection assembly should be checked for thefollowing condition (FEMA 267b):

Where:Zc - is the plastic modulus of the column section above and below the connection,Fyc - is the maximum specified yield stress for the column above and below,fa - is the axial stress in the column above and below,

- is the sum of the column moments at the top and the bottom of the panel zone,respectively, resulting from the development of the probable beam plastic moments, Mpr.

In our case the column moment at the top of the panel zone, Mct, and the moment at thebottom of panel zone, Mcb , are the same:

Mcb= Mct= (MC/2)(LC /2 - dc/2)/( Lc /2) .

For the proposed connection the last formula produces the following values for bothspecimens:

4

Σ ΣZ F f Mc yc a c( ) / . .− > 1 0

ΣMc

Mcb= Mct= 23351 kip*inch for the Specimen 1 andMcb= Mct= 21406 kip*inch for the Specimen 2.

The axial stress in the column is calculated from the shear load acting in the columndivided by the effective area of the column cross section:

fab = f ab = Vp/Ac.

In case of the proposed connection the axial stress is calculated as following:

fab = fab = 4.51 ksi for the Specimen 1 andfab = fab = 4.13 ksi for the Specimen 2.

Therefore the main criteria for the strong column - weak beam condition will be satisfiedfor both specimens, because of the following results:

2.3. The Connection Details Calculations.

2.3.1. Calculation of the T-section Stem Thickness at the Weakest Section near theColumn Face.

The weakest cross section near the column face is located at the K-line of the T-section.This section is at a distance, which includes the flange thickness of the T-section, tsts andradius of the fillet in the K-line. This value is equal to 3 inches. The location of thesection is presented on Fig. 2-4. The moment at this location is calculated as following:

Mws = Vp(Lb – 3).

The numeric values for the moment in case of the proposed connection are calculated asfollowing:

Mws = 49214 kip*inch for the Specimen 1 andMws = 45114 kip*inch for the Specimen 2.

The chosen cross section consists of the cross sections of T-section stems and the crosssection of the shear plate, as it is shown in Fig. 2-5. The out of scale picture of the chosencross section with the dimensions used during the calculation is presented on Fig.2-5.The connection design allows a plastic deformation in this cross section. Therefore thetotal moment in this cross section will consists of two moments. The fist moment, M1 , iscalculated with the plastic modulus of the shear plate, Z1, and has the followingexpression:

M1 =Fysp* Z1 ,

5

Σ ΣZ F f Mc yc a c( ) / . .− = >1 10 1 0

Σ ΣZ F f Mc yc a c( ) / . .− = >1 21 1 0

for the Specimen 1 andfor the Specimen 2.

where Z1 =(0.75*(24) 2) /4.

The corresponding numerical value (for Fysp = 56.6 ksi) is calculated as following and itis the same for both specimens:

M1 =6113 kip*inch.

The second moment, M2, is calculated with the plastic modulus of stems sections, Z2:

M2=Fyts*Z2,where Z2=2*18.4*16*0.96.

Therefore the numerical value of this moment (with Fyts = 64 ksi) can be calculated as

M2 =36176 kip*inch.

The total moment at this cross section calculated with the assumption of plasticdeformation of the section is:

Mdesigned =42289 kip*inch.

The difference between the designed moment, Mdesigned, and the expected one at thisweak section, Mws, is within 14 percent. This difference is acceptable, because the weakcross section has extremely small length (less then 3 percent of the beam depth) and it isfollowed by a very strong cross section. The strong cross section has very high value ofthe elastic section modulus and allows only elastic deformation up to the following valueof the moment:

Mstong =58415 kip*inch.

2.3.2. Calculation of the T-section Flange Size.

In order to find the thickness of the T-section's flange the flange was modeled as a fixed-end beam shown in Fig. 2-6. The concentrated load acts at the midpoint of the beam andrepresents the stem's puling force. The maximum force developed in the stem can becalculated as following:

P = Fyts Astem,

where the cross section area of the stem is Astem=16*0.96=15.36 inch2. Therefore thevalue of the load is as following:

P = 983 kips.

The statically indeterminate beam presented in Fig. 2-6, can be solved for the reactionforces and moments and they have the following expressions (AISC 1995a, page 4-195):

Ra= Rb =P/2,

6

Ma= Mb=PL/8.

Therefore the reaction moment applied at the bolt location has the following maximumvalue:

Ma= Mb=584 kip*inch

The connection design assumes that the plastic deformations can be developed near thebolt location. Therefore the required plastic modulus of the flange cross section is asfollowing:

Z required = Ma /Fyts ,or Zrequired = 9.1 inch3.

The chosen flange of the T-section has to have the plastic modulus not lower then therequired one, and it's value for the chosen flange thickness is:

Zdesigned = 16*(1.73) 2/4 =11.97 inch3.

The elastic section modulus for the rectangular cross section of the flange is 1.5 timesless then the plastic one and is equal to

Sdesigned = 7.98 inch3.

Therefore the design allows the flange yielding but without developing a plastic hingenear the bolt location.

2.3.3. Calculation of Bolt Size Between the T-section and the Column Flanges (Duringthe Design 1 ¼ inch High-Strength Bolts were Chosen).

As it was discussed before the model for the T-section flanges is shown in Fig. 2-6.According this model the total axial force acting in each row (of 4 bolts) is:

Ra=Rb=P/2=492kips.

The corresponding axial force acting in one bolt is:

Rbolt=123 kips.

For the chosen 1 ¼ inch high-strength bolts the design tensile strength is (AISC 1995b)

Rbolt(LFRD)=104 kips.

The conducted tests on the high-strength bolts (see section 1.3 of this report) show veryhigh ductility of the used bolts, the bolt specimen started to yield at 132 kips with theultimate tensile load of 150 kips (see results presented on Fig. 1-3). These results explain

7

the chose of the 1 ¼ inch high-strength bolts for the column and the T-sectionconnection.

2.3.4. Calculation of Weld Size to Beam Flanges for Both Specimens (3/4 inch FilletWeld Was Used).

The total length of the weld is different for the specimens, the length per the beam flangeis designed as following (including weld in U-shaped part for Specimen 2):

lweld - 2*23.5=47.0 inch for the Specimen 1 and.lweld = 2*14.75+2*9.25=48.0 inch for the Specimen 2.

The maximum force in the beam flange will be equal to the product of the flange crosssection area, Abflange, and the specified yield stress for the beam, Fyb :

Fmax = Abflange Fyb.

For the numerical parameters of the proposed connection, the value of this force is equalto

Fmax = 638 kips for both specimens.

The load per inch of the fillet weld is calculated as following:

Fper\inch =13.3 kips/inch for the Specimen 1 andFper\inch =13.6 kips/inch for the Specimen 2.

The required leg length is calculated the widely used expression

lleg = Fper\inch/(0.707*1.7*21 ksi).

The required leg length for the designed specimens is determined as following:

lleg =0.54 inch for the Specimen 1 andlleg =0.53 inch for the Specimen 2.

Based on the results of these calculations the 3/4 inch weld was used during themanufacturing the specimens.

3 Experimental Program3.1 Introduction

This section summarizes the results of cyclic testing of two full scale beam-columnbolted connection specimens. The specimens were designed by Prof. E.P. Popov andwere fabricated by Stoltz Metals, Inc. The tests were carried out in the Structural

8

Research Laboratory of the Pacific Earthquake Engineering Research Center, Universityof California at Berkeley.

3.2 Test Specimens, Test Setup and Instrumentation

3.2.1 Test Setup

The specimens were tested in the Structural Research Laboratory of PEER, UC Berkeley.The test setup was designed to accommodate specimens with columns in verticalposition, as shown in Fig. 3-1. The specimens were attached to horizontal and verticalframes. The horizontal steel frame was pre-stressed to the strong floor. The columns inthe test specimens were attached to the horizontal frame and the vertical reaction frameusing short segments of W 14x311 to achieve near pinned boundary conditions.

The load was applied to the cantilever beam end by a 400-kip hydraulic actuator, througha clevis bolted to the beam end plate. The testing setup had displacement capacity of±7.75 inches and load capacity of ±350-kip. No axial load was applied to the column.The test was conducted using the beam end displacement control. The beam end was at adistance of 134 in from the column face. To prevent out of plane movement of the beam,a vertical bracing system was provided near the beam end. The photograph in Fig. 3-2shows a view of a test in progress.

3.2.2 Instrumentation

Many sensors were used to monitor the response of the specimens during the test in orderto understand the specimen behavior. Figure 3-3 shows the location of displacementmeasuring instruments on the specimens. The imposed displacement at the end of thebeam was measured by LVDT (Linear Variable Differential Transformer). Thisdisplacement is denoted by , a load cell in-line with the actuator measured axial force P.The DCDT (which is a LVDT with built-in solid state oscillator and phase-sensitivedemodulator) displacement transducers were used to provide the remaining displacementmeasurements. The deformation of the beam panel zone was calculated from readings at

and DCDT locations. Global deflection shape of the column was measured bydisplacement transducers. The amplitude of gap opening between the T-section

flanges and the column flanges was measured by two displacement transducers and

Strain gages and rosettes were glued at critical locations to investigate local response.Figures 3-4 shows these locations on the Specimen 1, the strain measuringinstrumentation for the Specimen 2 was the same, excluding some minor changes in gagelocations. Thirty-eight channels of data were used during testing.

3.2.3 Data Acquisition

The test control and the data acquisition system were run by a PC Windows-based controland acquisition program called Automated Testing System (ATS) developed by SHRPEquipment Corporation of Walnut Creek, California. This Program is capable of signal

9

δ7

δ6

δ3δ2δ1

δ

δ8δ

-

.

generation, four-channel servo-actuator command, and sixteen-channel data acquisition.For the tests the ATS system was used to monitor and control the displacement and force-feedback signals.

Other data were monitored and recorded using an AutoNet data acquisition system with acapacity of 64 channels. Pacific Signal Conditioners were used to amplify the transducersand the strain gages signals and to remove frequencies above 100 Hz from the analogsignal.

3.2.4 Loading History

The testing program was based on the ATC-24 document "Guidelines for Cyclic SeismicTesting of Steel Structures". The specimens were tested under displacement control,following a loading history consisting of stepwise increasing deformation cycles. Atcertain stage of plastic deformation of the specimens a few cycles with small amplitudewere imposed. Each loading step was defined by the peak beam end displacement and bythe number of cycle. Table 3-1 presents the testing program for the Specimen 1 and theSpecimen 2.

3.2.5 Data Processing

The specimen behavior was characterized by the following parameters: applied load,beam end displacement, total plastic rotation of the connection, panel zone sheardeformation, column deformation, deformation in the T-section flange, and beamdeflection. A test specimen layout, the corresponding measurements, and the chosenpositive direction of applied load, and measured displacements are shown on Fig. 3-3.

Total displacement of the beam end is caused by rigid body motion of theconnection, the deformations of the beam itself, column, panel zone, and deformation inthe T-section flange. The rigid body motion was possible due small flexibility in thevertical reaction frame. This part of the displacement was not too large, but it could notbe neglected. Therefore the beam end relative displacement was calculated from thetotal one by subtracting of the rigid body displacement. As a result of the column andpanel zone deformations, the panel zone rotates trough an angle and changes its initialconfiguration. Four displacement measurements ( and ) were used to computethe connection rotation due column deflection and panel shear deformation . Thetotal beam rotation can be separated into four components: rotation due deformation ofthe beam itself , rotation caused by rigid connection rotation , the contribution fromthe panel zone , and the rotation due gap opening in the T-sections . These valueswere determined as follows:• Total beam end displacement:• Relative beam end displacement: The remainder of the

calculation was done using this value of the displacement; where H is a distance frompin to pin along the column, and L is the distance from the beam end to the center lineof the column

• Total rotation:

10

δtotal

δ1 δ2 δ4 δ5

δtotal

δ

, , ,

δ δ δ δ = - -total L H( ) / .6 3

θ δ= / L

γ

γ

θT

θb

θ

θc

θc

θc

• Connection rotation due column deflection: Where d is a distancebetween continuity plates

• Panel zone shear deformation: . Where a and b are thedimensions of the rectangular panel zone area (distance between targets in horizontaland vertical directions)

• Rotation due gap opening (and deformation) between T-section flange and columnflange:

• Total plastic rotation: Where M=PL0 (L0 is a distance from center-lineof the actuator to the face of the column) is the moment at the face of the column and

is the elastic stiffness determined from M versus curve. The unloading path ofone of the elastic cycles below the reverse point was used to estimate this stiffness, toavoid the influence of initial imperfections, clearances, hysteresis, etc.

A set of programs for the MATLAB 5.3 environment was created to process data and toplot results in accordance with the procedure described above.

3.3 Test Results

3.3.1 Specimen 1

Testing of the first specimen was conducted on July 30, 2000. The specimen sustained allloading steps up to and including the 5.69" beam tip displacement cycles withoutsignificant damage. Testing was stopped because the maximum load for the test setupwas reached. Photo of the specimen's close up side view after the test is presented on Fig.3-5. During the last set of the load reversals a slight buckling in the beam web andflanges was observed. The residual buckling in the beam flange and web is shown on Fig.3-6 and Fig. 3-7. During the test energy was dissipated by cyclic yielding of the flangesof the T-sections, the gap between the T-section and column flanges was opening andclosing periodically. The residual gap in the upper T-section is shown on Fig. 3-8.

Table 3-1 presents the loading protocol for the both specimens. The first row in the tablepresents total beam end displacement (in other words: beam end total verticaldeflection). The loading history for Specimen 1 is plotted in Fig. 3-9.

The plot of applied force versus beam tip displacement response is presented onFig. 3-10. The values of the displacement were obtained directly from the LVDT reading.The relative displacement ( ) was calculated from previous displacement by subtractingthe specimen's displacement as rigid body. Because of some small flexibility in thevertical reaction frame occurred this displacement could not be neglected. The plot ofapplied force versus relative beam tip displacement is presented on Fig. 3-11.

Based on the values of the relative beam tip displacement the total beam rotation iscalculated. The imposed moment versus beam total rotation is presented on Fig. 3-12.Figure 3-13 shows the applied moment versus the beam plastic rotation. The deformationof the column panel zone is presented on Fig. 3-14.

11

θ δ δc d= -( ) / .5 4

γ δ δ a b ab= - +( ) / ( )5 42 2 2

θ δ δT d= - /( ) .8 7

θ θ θpl M K= - / .

Kθ θ pl

δ total

δ total

δ

During the test the visible opening between the T-section flanges and the column flangeswas observed. The amplitude of the opening between flanges was measured in thefollowing way, the installed DCDT shows the relative displacement between targetslocated at the center plane of the column and the T-section flanges (see details in Fig.3-3). Therefore during mutual compression of two flanges this displacement is negative,whereas the tension in T-sections web increases this distance and it becomes positive.This relative displacement between the flanges is called as "gap opening" in the report.Figure 3-15a and 3-15b shows these values during the test. The gap opening between T-section flange and column flange for the upper T-section is presented on Fig. 3-15a. Thesame value for the lower T-section is presented on Fig. 3-15b. The beam rotation duethese openings in the T-sections is presented on Fig.3-16.

The imposed force versus beam rotation due panel zone rotation is presented on Fig. 3-17. The relative beam rotation calculated by subtracting rotation of the panel zone,rotation due gap opening in T-sections and the panel zone deformation is presented onFig.3-18.

3.3.2 Specimen 2

Testing of the second specimen was conducted on July 20, 2000. The specimen sustainedall loading steps up to the 5.69" beam tip displacement cycles and failed at the first rampof the last cycle. The fracture was caused by crack in the web of the lower T-section. Thecrack line started at the end of the weld and went through the hole for 1 inch bolt. Testingwas stopped after the finishing this cycle. Photo of the specimen's side view after thetesting is presented on Fig. 3-19.

During the test some energy was dissipated by cyclic yielding of the T-sections, the gapbetween the T-section and the column flanges was open and closed periodically. Theresidual gap in the top T-section is shown on Fig. 3-20.

At the end of the test a slight buckle in the beam web and flanges was observed. Theresidual buckling in the beam flanges is shown on Fig. 3-21 and Fig. 3-22.

The crack in the stem of the bottom T-section is shown on Fig. 3-23 and 3-24. Fig. 3-23presents the view of the location of this crack on the stem of the T-section. The locationwas close to the K-line of the T-section and it was parallel to it. The crack started fromthe end of the fillet weld, continues through the nearest hole for the 1 in bolt and ends atthe next bolt hole. The close view of the crack is presented on Fig. 3-24. The arrows tracethe crack line.

The loading protocol for the Specimen 2 is presented in Table 3-1. The loading history isplotted in Fig. 3-25.

The layout of the displacement measuring instrumentation was identical for bothspecimens given in Fig. 3-3. The displacement at the beam tip was measured by LVDT,whereas the remainder of displacement measurement was done using DCDT.

12

The plot of applied force versus beam tip total displacement is presented on Fig. 3-26.The values of the displacements were obtained directly from the LVDT reading. Therelative displacement was calculated from previous displacement by subtracting thespecimen's displacement as a rigid body. The flexibility of the reacting frame was takeninto account. The plot of applied force versus relative beam tip displacement is presentedon Fig. 3-27.

Based on the values of the relative beam tip displacement the total beam rotation iscalculated. The imposed moment versus the beam total rotation is presented on Fig. 3-28.Figure 3-29 shows the applied moment versus the beam plastic rotation. The deformationof the column panel zone is presented on Fig. 3-30.

During the test the visible opening between the T-section and column flanges wasobserved. The values of the gap opening were measured by DCDT. Figure 3-3la and 3-31b shows these values during the test. The gap opening between the T-section flangeand column flange for the top T-section is presented on Fig. 3-3la. The same data for thebottom T-section is presented on Fig. 2-31b. The beam rotation due these openings in theT-sections is presented on Fig. 3-32.

The imposed force versus beam rotation due panel zone rotation is presented on Fig. 3-33. The relative beam rotation calculated by subtracting rotation of the panel zone,rotation due gap opening in the T-sections and the panel zone deformation is presented onFig.3-34.

4 Experimental Results and Conclusions4.1 Experimental results

A brief summary of experimental results and key parameters characterizing theperformance of tested specimens is presented in Table 4-1. The beam end displacementcorresponds to the relative beam end displacement .

4.2 Conclusions: advantages and disadvantages of proposed connections

4.2.1 Advantages

The design and performance of the proposed beam-to-column connections shows thefollowing advantages:- all welding work can be done in a welding shop, in convenient welding positions

final assembling with bolts is relatively easy procedure and does not require arigorous quality assurance inspection (in order to achieve the required clamping forcebetween the column and the T-section flanges the widely available torque multiplierfrom WRIGHTTOOL: Model 9S393A was used; the device does not produce anynoise and has an accuracy of ±5%)

13

δ

-

after test disassembling of Specimen 2 shows that repairing and replacing beam withnew T-section is neither difficult nor expensivethe beam deformation is minimal due to active participation of the T-sections flangesand the column flanges during cyclic inputwith shims properly installed, the connection develops less residual straineliminating large quantities of field weld greatly helps the connecting work to keepup with the steel erection.

4.2.2 Disadvantages

The chosen design and the failure of Specimen 2 show the following disadvantages andsuggested improvements:

steel along the K-line of the T-section must be carefully selected1 inch bolts (as used in Specimen 2 to pre-stress the T-section web to beam flange)requires a greater distance between the bolt and the end of the fillet weld.Alternatively, it appears that the bolts can be omitted altogether

- steel material of 1 ¼ " bolts has to be high quality as used in the tested connections- connection based on the proposed design require shims for field assembly- beams with welded top and bottom T-sections require more shipping space during

transportation.

4.2.3 Future Research Directions

Based on the conducted tests and followed data analysis the following future research onthis type of connections is planned:- conduct 3D finite element analysis (FEA) of the connection to explore the possibility

of exchanging the existing 1 inch bolts to clamps and to investigate the decision toremove some or all of them

- conduct 3D FEA of the connection to evaluate the critical parameters at the column-tee joint, including the T-section size, bolt diameter, the clamping load variation andthe prying action

- fabricate and test new specimens with an improved design based on the theoreticalresearch and results of the previous tests.

14

-

-

--

--

References:

1. Tsai, K.C. and Popov, E.P. 1990. Cyclic behavior of end-plate moment connections.ASCE J. of Struct. Engineering, Vol.116, No.11.

2. Tsai, K.C. and Popov, E.P. 1988. Steel Beam-Column Joints in Moment ResistingFrames. Report No. UCB/EERC 88/19, Earthquake Engineering Research Center,University of California at Berkeley.

3. Murray, T.M. et al. 2000. Cyclic testing of bolted moment end plate connections.Struct. And Materials Lab., Virginia Polytechnic Institute and State University.

4. Leon, Roberto et al. 2000. Tests on bolted connections. School of Civil andEnvironmental Engineering, Report No. SEMM 00-02, Georgia Institute ofTechnology.

5. FEMA-267. 1995a. Interim guidelines: evaluation, repair, modification and design ofwelded steel moment frames. FEMA Report No. 267. Washington, D.C.: FederalEmergency Management Agency.

6. FEMA-267. 1995b. Interim guidelines: advisory No. 1. supplement to FEMA 267.FEMA Report No. 267. Washington, B.C.: Federal Emergency Management Agency.

7. AISC. 1995a. Manual of steel construction. Load & resistance factor design. Vol.1,Structural members, Specifications & codes, Second edition. Chicago: AmericanInstitute of Steel Construction, Inc.

8. AISC. 1995b. Manual of Steel Construction. Load & Resistance Factor Design.Vol.2, Connections, Second edition. Chicago: American Institute of SteelConstruction, Inc.

15

Table 2-1. Dimensions of the W36xl50 beam.

Flange width,bfb

[inch]

12

Flangethickness, tfb

[inch]

0.94

Webthickness,

twb

[inch]0.625

Depth, db[inch]

35.85

Sectionmodulus,

zb[inch3]

581

Area, Ab[inch2]

44.2

Moment ofinertia, I b[inch4]

9040

Length, Lb[inch]

134

Table 2-2. Dimensions of the W14x283 column.

Flange width,bf

[inch]

16.125

Flangethickness, tfc

[inch]

2.06

Webthickness,

tf[inch]1.29

Depth, dc[inch]

16.74

Sectionmodulus,

Zc[inch3]

542

Area, Ac[inch2]

83.3

Moment ofinertia, Ic[inch4]

3840

Length, Lc[inch]

136

Table 2-3. Dimensions of the T-sections cut fromW40x264.

Flange width,fts [inch]

12

Flange thickness,

[inch]1.73

Stem thickness,tsts

[inch]0.96

Total depth for Specimen 1(Specimen 2), dsts

[inch]28.75 (20)

Length, Lts[inch]

16

Table 2-4. Material Properties.

No

123

Part of Connection

BeamColumn

T-section

Yield Stress, Fy[ksi]56.65264

Ultimate Stress[ksi]74.46679

Section Size

W36xl50W14x283

WT40x264

Grade

Gr50Gr50Gr50

Table 3-1. Testing program for the both specimens.

Total beam enddisplacement[inch]No of cycles

0.36

6

0.53

6

0.71

6

1.07

6

1.42

4

2.14

2

0.53

2

2.85

2

4.27

3

5.69

6** Only 2 cycles at this level were performed for Specimen 2.

16

tfts

Table 4-1. Short summary of test results

Key parametersYield load [kips]Beam end displacement at the yield point [inch]Elastic stiffness of the connection [kips/inch]Maximum beam end displacementBeam end displacement at failure [inch]Maximum imposed load [kips]Maximum imposed moment at the column face [kips*inch]Maximum connection rotation [ % ]Maximum plastic connection rotation [ % ]Maximum rotation due gap opening [ % ]Maximum relative beam rotation itself [ % ]

Specimen 12301.21805.2

N/A345

486454

2.51.00.6

Specimen 22301.21785.23.5327

431644

3.30.71.5*

This value is high because it includes the beam rotation after the bottom beam flange failure

17

Figure 1-1. Design details of end-plate connections for Specimens 10 and 10R, and thatof direct welding to column, Specimen 9 (K.C. Tsai, E.P. Popov 1988, 1990).

Load

(ki

ps)

Beam Rotation (%)

Figure 1-2. Cantilever beam load versus beam rotation for Specimen 10R (K.C. Tsai, E.P.Popov 1988, 1990).

18

Figure 1-4. Stress versus strain for coupon test of A490 1 ¼ bolt material.

19

Figure 1-3. Load versus elongation for A490 1 ¼ bolt.

Figu

re 1

-5. G

loba

l dim

ensi

ons

and

geom

etry

of t

he te

sted

spe

cim

ens.

20

Figu

re 1

-6. D

esig

n de

tails

of S

peci

men

1.

21

Figu

re 1

-7. D

esig

n de

tails

of S

peci

men

2.

22

Figure 2-1. Desired plastic frame behavior with plastic hinges developed in beams.

Figure 2-2. Probable plastic hinge location.

23

Figure 2-3. Calculation of the moment at the centerline of the column.

Figure 2-4. The weakest cross-section of the beam near the column face.

24

Figure 2-5. Dimensions of the weakest cross section near the column face.

Figure 2-6. View of the column and T-section connection with the correspondingmechanical model.

25

Figure 3-2. View of a test in progress.

26

Figure 3-1. Test setup for both specimens.

Figure 3-3. Reference dimensions and measurements for the test specimens.

Figure 3-4. Strain gages and rosettes location for Specimen 1.

27

Figure 3-6. Residual beam flange buckling (after the test).

28

Figure 3-5. Specimen 1 after the test (side view).

Figure 3-7. Residual beam web buckling (after the test).

Figure 3-8. Residual gap opening in the upper T-section (after the test).

29

Figure 3-9. Loading history for Specimen 1.

Figure 3-10. Imposed load versus total beam end displacement for Specimen 1.

30

Tip

Dis

plac

emen

t [in

ch]

Time [sec]

Actu

ator

forc

e [k

ips]

Beam end total displacement [inch]

Actu

ator

forc

e [k

ips]

Beam end displacement [inch]

Figure 3-11. Imposed load versus beam end displacement for Specimen 1.

Mom

ent

[kip

s*in

ch]

Beam Rotation [%]

Figure 3-12. Moment versus beam total rotation for Specimen 1.

31

Mom

ent [

kips

*inch

]

Beam Plastic Rotation [%]

Figure 3-13. Moment versus beam plastic rotation for Specimen 1.

Actu

ator

forc

e [k

ips]

Panel zone shear deformation [%]

Figure 3-14. Imposed load versus deformation in panel zone for Specimen 1.

32

Actu

ator

forc

e [k

ips]

Relative displacement between flanges[inch]

Figure 2-15a. Relative displacement between column and top T-section flanges forSpecimen 1.

Actu

ator

forc

e [k

ips]

Relative displacement between flanges[inch]

Figure 3-15b. Relative displacement between column and bottom T-section flanges forSpecimen 1.

33

Actu

ator

forc

e [k

ips]

Rotation [%]

Figure 3-16. Imposed load versus beam rotation due gap opening in T-sections(Specimen 1).

Actu

ator

forc

e [k

ips]

Rotation [%]

Figure 3-17. Imposed load versus panel zone rotation for Specimen 1.

34

Actu

ator

forc

e [k

ips]

Relative beam rotation [%]

Figure 3-18. Imposed load versus relative beam rotation for Specimen 1.

35

Figure 3-19. Specimen 2 after the test (side view).

Figure 3-20. Specimen 2: Residual gap opening in top T-section.

36

Figure 3-21. Specimen 2: top beam flange buckling.

Figure 3-22. Specimen 2: bottom beam flange buckling.

37

Figure 3-23. Specimen 2: crack line location.

Figure 3-24. Specimen 2: close view of the crack line.

38

Figure 3-25. Loading history for Specimen 2.

Tip Displacement [inch]

Figure 3-26. Imposed load versus total beam end displacement (Specimen 2).

39

Forc

e [k

ips]

Tip

Disp

lace

men

t [in

ch]

Time [sec]

Forc

e [k

ips]

Tip Displacement [inch]

Figure 3-27. Imposed load versus beam end displacement (Specimen 2).

Beam Rotation [%]

Figure 3-28. Moment versus beam total rotation (Specimen 2).

Mom

ent

[kip

s*in

ch]

40

Mom

ent

[kip

s*in

ch]

Beam Plastic Rotation [%]

Figure 3-29. Moment versus beam plastic rotation (Specimen 2).

Forc

e [k

ips]

Panel Zone Shear Deformation [%]

Figure 3-30. Imposed load versus column panel zone deformation (Specimen 2).

41

Forc

e [k

ips]

Gap Opening [inch]

Figure 3-3la. Relative displacement between column and top T-section flanges(Specimen 2).

Forc

e [k

ips]

Gap Opening [inch]

Figure 3-31b. Relative displacement between column and bottom T-section flanges(Specimen 2).

42

Forc

e [k

ips]

Rotation [%]

Figure 3-32. Imposed load versus beam rotation due gap opening in T-sections(Specimen 2).

Forc

e [k

ips]

Rotation [%]

Figure 3-33. Imposed load versus panel zone rotation for Specimen 2.

43

Forc

e [k

ips]

Rotation [%]

Figure 3-34. Imposed load versus relative beam rotation for Specimen 2.

44

STEEL COMMITTEE OF CALIFORNIA

TECHNICAL INFORMATION & PRODUCT SERVICE

DECEMBER, 1990

Design of single PlateShear ConnectS, ohs.

by

Abolhassan Astaneh

Steven M. Call

Kurt M. McMullin

with discussion by

Ralph M. Richard

DISCUSSION*Design of Single Plate Shear ConnectionsPaper by ABOLHASSAN ASTANEH, STEVEN M. CALL and KURT M. McMULLIN(lst Quarter, 1989) See page 7 of This Publication.

Discussion by Ralph M. Richard

The paper develops a design procedure for single plate shearconnections based upon the results of a shear-rotation de-vice (shown in Fig. 4 of the original paper). The claim ismade that in previous studies "...the shear connectors havebeen subjected to moment and rotation or only direct shear

without rotation." This is not true.This writer developed a design procedure for single plates

based upon stub beam tests and full scale beam tests thatincluded realistic connection shears.• Shown in Figs. 13 and14 of this writer's paper• are moment-rotation curves whichshow the effect of shear and given on page 45 of that paperis the analytical moment-rotation curve which indeed includesthe effect of shear. It was found, however, that for practi-cally all single plate designs the ratio, e/h, (eccentricitydivided by bolt pattern depth), was 0.5 or greater and asshown in Fig. 13, the moment-rotation relationship is not sig-nificantly affected by the connection shear. The reason forthis is that the maximum moment in single plate shear con-nections occurs at about 1.5 times the service load. This isshown for a three and a five bolt connection in Figs. 1 and2, respectively, of this discussion paper and is in agreement

Ralph M. Richard is professor, Department of Civil Engineer-ing and Engineering Mechanics, University of Arizona, Tuc-son, Arizona.

with Astaneh's observation that "...based on observationsmade during the tests, it appears that shear' tabs go throughthree distinctive phases of behavior. At the very early stages,a shear tab acts as a short cantilever beam with moment be-ing dominant. As load increases, the shear tab acts as a deep

shear beam with the shear yielding effect dominant." HadAstaneh performed a full scale test, he would have observedthat the shear tab does not begin the shear yielding phaseof action before application of 1.5 times service load. Thislinear connection action is shown in the shear-rotation plotsof Fig. 9 in Astaneh's paper. Moreover, consider Astaneh'sDesign Example 1. His design procedure results in a 21 in.x 1/2 in. x 41/2 in. plate with a shear of 102 kips serviceload. At 1.5 times service load, the shear stress in this 3 in.long and 21 in. deep cantilever beam is approximately 15

ksi which is less than the shear yield stress of 21.6 ksi forA36 steel. In his Design Example 2, he uses a 12 in. x IAin. × 41/2 in. plate with a service shear load of 33 kips. Theshear stress in this plate at 1.5 times service load is 16.5 ksiwhich again is well below the yield stress of 21.6 ksi for A36

steel.The research at the University of Arizona, based upon stub

beam tests, full scale beam tests, and inelastic finite elementanalyses that used experimentally determined bolt-deforma-tion results, found thaf the maximum connection moment

00

Moment

· Uniform •' •.•.- IO

" x ./' . .... •,וs 'X,span L=20 '•x

' • / C ......... .3,3/4 A325 '•x/ • Plate : 5/t6 x

' " l " l h l l l l l l l l h I H H ' ' ' I ' ' H ' ''1 'J'Lcltlll' '[ " ' ' ' ' ' ' ' l ' ' " ' ' ' H I ' " ' ' H ' 0

lO 20 30 40 50 60 70 80Uniform Load. If, Kips

Fig. 1. Single plate moments and eccentricities.

ca

c

400

3'50

300

3,90

190

100

50

O

•'m• rT?T?T?• • r •n r• •TrrlTrT•r' [ ......... I ......... f ......... I'" crrrr•

· 't

•, Loed Uniform ' •

Ir' Benm w 18 X ,'1,5 '•· ,• Span L=20

· w Connection 5 . 3/4' A325• . ' Plate 5/16'

· . , w.,.•. 46 I Kips

d

uLt u •r[ uu•axt tl•u&xu• h i i ,, ,, i, I . . . . . . . , h . , , , ,

lo 30 30 40 50 60 70 80 90Unilorm Iond. W, Kips

Fig. 2. Single plate moments and eccentricities.

3O

20 >•

tO

5

•o°

'k Reproduced from AISC Engineering Journal, Volume 27, No. 3, 3rd Quarter 1990

occurred near or above 1.5 times working load as shown inFigs. I and 2 of this paper. The structural engineering pro-fession requires that structural elements (connections, beam,etc.) must be designed to have the strength to resist the max-imum value of the envelope of forces the element is subjectedduring loading. For the single plate shear connection, themaximum value of the moment the weld is subjected is atabout 1.5 times the service load. Beam end rotations at theseloads are of the order of 0.006 to 0.014 radians which arewell below the 0.030 test values used by Astaneh. For uni-formly loaded beams, it is noted that in Design Example 1,the end rotation of this beam is 0.0055 radians at serviceload and for Design Example 2 it is 0.0046 radians. How-ever, Astaneh's recommended test and design procedure

which is based upon shear yielding of the plate, used rota-tions four to six times these values.

Because of the significant difference in the design eccen-tricities recommended by Astaneh and those of this writerfor the design of the single plate welds, this writer stronglyrecommends that a minimum of three full scale tests withbeams subjected to a factored uniform load of 1.5 times theservice load be performed by an independent laboratory toevaluate the moment generated by the single plate shear con-nection before this design procedure is recommended to thestructural engineering profession. This writer has found thatthese connections generate significantly larger moments thandouble framing angles subjected to the same beam shear.2

Because the bolts of the single plate are in single shear,whereas these are in double shear for double framing an-gles, the single plate is twice as deep and therefore muchstiffer.

REFERENCES

1. Richard, R. M., P. E. Gillett, J. D. Kreigh, and B. A.Lewis, "The Analysis and Design of Single Plate Fram-ing Connections," AISC Engineering Journal, Vol. 17(No. 2): 38-52.

2. Richard, R. M., W. K. Hsia, and M. Chmielowiec, "Mo-ment Rotation Curves for Double Framing Angles," In-ternational Journal of Computers and Structures, Vol. 30(No. 3): 485-94.

Addendum/Closure by A. Astaneh, S. C. Call and K. M.McMullinThe discussion by Professor Richard mainly compares theresearch methodologies and design procedures developed byresearchers at the University of Arizona (UA Method) tothose developed by Authors at the University of Californiaat Berkeley (UCB Method). The UCB Method has formedthe basis of the methods currently in the 9th Edition of theAISC Manual. u In order to make the closure of discussionuseful to the readers, the authors have responded to the state-ments made in the above discussion and have provided a brief

comparison of the UA and UCB design methods in the fol-lowing sections.

RESEARCH METHODOLOGY

In the paper, it was indicated that "... in the past in mostcases, the shear connections have been subjected to momentand rotation or only shear without rotation instead of a realis-tic combination of shear and rotation." This statement is par-ticularly true with experiments conducted at the Universityof Arizona. Figure I (reproduced from Fig. 3 of the paper)shows representative shear-rotation relationship that existedin the connections tested by Professor Richard and his re-search associates (Lines OA and OB). Also shown in the fig-ure are actual shear-rotation relationship in a shear connec-tion (Line OCD) and shear-rotation relationship that existedin UCB tests (Line OCE).

In the stub (cantilever) tests conducted at UA, the con-nections were primarily subjected to rotations with very smallshear applied to the connection. The shear-rotation relation-ship for these tests is represented in Fig. I by the line OA.By comparing this shear-rotation line to the actual shear ro-tation line (Line OCD), it is clear that the connections instub beam tests were subjected to unrealistically large rota-tions with very small shear forces applied to the connection.Since shear forces generated in stub (cantilever) beam tests

are small compared to actual shear forces in shear connec-tions, failure modes are very unrealistic, therefore, unrealistictests should not be used to develop design procedures forshear connections.

From published data apparently a total of four tests havebeen conducted using the test set-up shown in Fig. 2. Simi-lar test set-ups have been used in the past by severalresearchers to apply large shear forces to the connection.However, if the beam shown in Fig. 2 is not loaded to fail-ure, the amount of rotation that will be developed in the con-

Q_

:fO

0

zz00

Z0

bO

ACTUAL SHEAR-ROTATIOb! CURVE D G4•.CAc2u AL >- •,• ' • '•100

0.02 0.04 0.06

ROTATION OF BEAM END, rod.

Fig. 1. Shear-rotation relationship in UA and UCB tests.

2

nection will be very small and will be limited to elastic endrotations which are very small compared to realistic rota-tions that will be imposed on the connection at the time ofbeam collapse.

In the full-scale tests conducted at the University of Ari-

zona, the amount of maximum shear applied to the connec-tions is unexplainably very low. A representative of the shear-rotation relationship applied to the connections in UA full-scale tests is shown in Fig. 1 as Line OB. Due to applicationof very low shear to the connection in these full-scale tests,no realistic failure mode has been observed or reported andapparently only some minor yielding of bolt holes and defor-mation of bolts have been observed.

It is unfortunate that full-scale tests conducted at the iJni-versity of Arizona have not been loaded to failure. Appar-ently, the loading was not even enough to cause significantyielding in the connections. If the tests were destructive,several failure modes observed by us as well as by otherresearchers •-6'•° might have been observed and invaluabledata on strength of connection could be obtained. The rea-son for stopping the loading at such a low level apparentlywas a decision to load the specimens up to 1.5 times yieldcapacity of the beam. From published information, it is not

clear why strength of the connections were studied under suchan arbitrary and unrealistically low load level. Therefore,in our view, full-scale tests conducted at the University ofArizona were incomplete and have not provided informa-tion regarding strength and failure modes of the connections.

The details of full-scale tests conducted at the Universityof Arizona and the results are not published. However, frompublished data, it appears that the objective of full-scale testsat the University of Arizona may have been to study move-ment of point of inflection of the beam and moment-rotationbehavior. Since these full-scale tests have been non-destructive and no connection failure modes have been ob-served, it is not clear how the information obtained fromloading of specimens in elastic range could be used to de-velop design procedures concerning failure modes and thecorresponding shear strength capacities.

The inelastic finite element program used in UA studiesis an analysis program and could only provide useful infor-

{top and bottom)

W Section

Load Cell I ,

•/2 • •/2

i

Fig. 2. Test setup used in UA tests (Ref 8).

mation on the state of the strain and or stress. The programis not capable of predicting failure modes and strengths suchas weld fracture, bolt fracture, fracture of net section or frac-

ture of the edge distance. Apparently, the finite element pro-gram is used to simulate moment-rotation response. Again,similar to full-scale tests, in the finite element analyses themaximum load was about 1.5 times service load of the beams.

As far as behavior of the connection is concerned, the max-imum load of 1.5 times service load of the beam used in UAtests and finite element analyses is very small. For exam-ple, the connection studied in Fig. 2 of the Discussion isloaded up to about 50 kips shear force (100 kips total beamload) whereas according to information obtained from ourdestructive tests of similar connections and by using wellestablished design concepts, the shear capacity of the con-nection is about 130 kips (260 kips total beam load). Itappears that the University of Arizona studies were limitedto the initial stage of loading where beam and connectionare almost elastic. Then the results of these studies are appliedto full range of loading up to the failure. Since the problemis highly nonlinear, the validity of this extrapolation isquestionable.

To remove the above difficulties, the authors have devel-

oped and used a test set-up that has enabled them to applyrealistic combinations of shear and rotation to the connec-

tion until the connection fails. The shear-rotation relation-ship used by the authors is shown in Fig. 1 as Line OCE.The details of test set-up as well as authors' methodologyare given in several references (1 to 6) and are not repeatedhere. The experimental work has resulted in establishingrealistic failure modes and corresponding design proceduresas reported in the paper.

COMPARISON OF UCB DESIGN PROCEDURESWITH UA PROCEDURES

The destructive tests conducted by a number of researchersincluding the authors have indicated that single plate shearconnections have six failure modes as follows:

a) shear yielding of plateb) bearing failure of bolt holes

c) failure of edge distanced) shear fracture of net sectione) bolt failuref) weld failure

The following sections provide a discussion of each fail-ure mode and corresponding design equations in UCBMethod and UA Method. In summarizing UA Method, theauthors have used the available published information. TM

a. Shear Yielding of Plate

In UCB method, this failure mode, which is very ductile anddesirable, is intentionally made to be the governing failure

mode.

3

The equation to be used to calculate the ultimate shearstrength of connection for this failure mode is:

R, = (L)(t)(0.6F,.) (1)

In UA method, this failure mode is not recognized.

b. Bearing Failure

In the UCB studies, t 6 bearing failure was observed in somespecimens. In the corresponding design procedures bearingfailure mode is recognized and equations that already existin the AISC Specification u are used to predict bearing fail-ure capacity of the connections.

In UA method, this failure mode is not considered. UsingUA method, since there is no lower limit on the thicknessof the shear tab, it is quite possible that designer unknow-

ingly can use a thin plate with relatively large diameter boltand cause bearing failure to be governing without ever notic-ing it.

The UCB design procedures as well as UA method recog-nize the beneficial effects of limited bearing yielding at thebolt holes. As a result both methods have an upper limit ofthickness of plate relative to the bolt diameter. In UCBmethod the limit is dr,/2 + h6 inch and in UA method thelimit is d•,/2. The limited bearing yielding provides rota-tional ductility and causes release of moment in the con-nection.

c. Shear Fracture of Net Area

I n U C B method this failure mode is fully recognized andthe following design equation is recommended to be usedto predict ultimate shear capacity of the net area:

R,,/, = [L - N(V2)(dl, + •6)l(t)(0.6F,) (2)

In a conservative approach, Eq. 3 which reflects the phi-losophy used in the AISC Specification u for shear failureof net area can be used.

R,a, = [L - N(dh + •6)](t)(0.6F,) (3)

The UA method apparently does not consider this failuremode. Again, similar to bearing failure mode, it is possiblethat by using thin plates, net section failure can govern with-out the knowledge of the designer.

d. Edge Distance Failure

As a result of experiments conducted by the authors at UCB,it was realized that due to dominance of shear, the verticaledge distance below the lowest bolt is the most critical edgedistance and should not be less than 1.5db nor 1.5 in. InUCB design method, it is recommended that this limitationbe applied to all edge distances (see Fig. 3a).

In UA method, it is recommended that horizontal edge dis-tance should not be less than 2do (see Fig. 3b). Apparentlythis recommendation is derived from results of cantilever+•o* .... h•ra la,•,•mc or• cnlaiopt•cl tn l•rof• rntation• and small

shear forces. In our tests, the horizontal edge distances didnot show signs of being critical whereas vertical edge dis-tances particularly the lower vertical edge distance provedto be very important and critical.

e. Failure of Bolts

In UCB method, bolts are designed for the combined effectsof direct shear and bending moment along the bolt line. Ourtests indicated that as beam is loaded, connections yield andbending moment in the connection continuously is releasedto the midspan of the beam. As a result, point of inflection

of the beam continuously moves toward the connection andis stabilized at a distance of el, from the bolt line. The valueof et, can be obtained from the following equation.

el, = (n - a - 1)(1.0) (4)

Therefore, in UCB method, bolts are designed to resistcombined effects of shear reaction of the beam and a mo-ment equal to reaction multiplied by et,.

In UA method, bolts are designed for direct shear only.This implies that bolt line is the location of point of inflec-tion of the beam where moment is zero and only shear ex-ists. Our experiments, as well as other tests conducted in

Canada, •° have clearly indicated that some moment de-velops along the bolt line.

Figure 4 shows variation of shear force and bending mo-ment in a typical shear tab connection. The connection usedto plot the curves is the same used in Fig. 2 of the Discus-sion. Figure 4 shows an experimental curve, UA finite ele-ment results and design equations according to UCB and UAmethods. It should be mentioned that test results shown inFig. 4 are plotted using test results for exactly similar speci-men but with Y8 in. thick plate rather than •6 in. The testresults for 3/8 in. plate are multiplied by Y6 to adapt them toh6 in. plate and then are plotted in Fig. 4.

It is not known why UA's design method neglects the mo-ment that exists along the bolt line. Even the finite element

Critical

)

Critical

Critical ical

Fie. 3. Ed. ee distance requirements in UCB and UA methods.

4

analysis given by Professor Richard in Figs. 1 and 2 of theDiscussion shows that considerable moment is present alongthe bolt line. In our view, based on seven tests conductedso far by us and several other tests by other researchers onthe shear tabs, neglecting moment along the bolt line is notjustifiable and can result in unconservatively overestimat-ing shear capacity of the bolts.

f. Weld Failure

In UCB method welds are designed for the combined effectsof direct shear and a moment due to the eccentricity of thereaction from the weld line, e,.. The eccentricity e,, is givenby the following equation. The equation is based on resultsof tests.

e,. = n(1.0) (5)

In UA method welds are designed for combined effectsof shear and moment, however, the moment that is estab-lished for design of the welds is unrealistically very large.

Figure 5 shows shear and moment variation along the weldline for the same shear tab shown in Fig. 2 of the Discus-sion. Similar to bolt design, the figure shows test results,UA finite element analysis (adapted from Fig. 2 of the Dis-cussion) as well as design equations according to bothmethods. The plots clearly shows that if one follows UAmethod in design of welds, the design point will be some-where in the vicinity of point A where moment is much largerand shear force is smaller than the realistic values that actu-ally occur in the connection (test curve).

The reason UA method results in using very large and un-realistic moment in design of welds is the use of large ec-centricity. Notice that in Fig. 5, slope of lines drawn fromthe origin (such as aA and OB) represent values of constant

eccentricity. In the Discussion Professor Richard indicatesthat connection should be designed for maximum possiblevalues of shear and moment. This statement is correct, butin UA's method rather than designing connection for maxi-mum combination of shear and moment, the connection isdesigned for shear corresponding to 1.5 times service loadof the beam and an eccentricity of shear that exists at thepoint of 1.5 times service load of the beam. What this actu-ally means is that as beam is loaded, eccentricity moves to-ward the support and when shear force exceeds a value cor-

responding to 1.5 times service load of the beam, theeccentricity remains constant. This is shown in Fig. 5 byLine CA. This is not realistic. As Fig. 5 indicates in actualloading shown by test curve, after onset of the bolt slip andyielding in the connection (Point D), eccentricity decreasescontinuously and stabilizes at much smaller value than theeccentricity corresponding to Point C. This can easily be seen

by comparing slope of Line CA (e,,. = 13 in.) and Line EB(e, = 5 in.).

In summary, tests conducted at the University of Arizonawere not destructive and thus cannot be used to establish fail-ure modes and design procedures. And, furthermore, the cor-responding design procedure considers only bolt failure andweld failure which are only two of the six failure modes thatactually should be considered. In addition, the design equa-tions suggested for the bolt failure appear to be unconserva-rive whereas equations proposed for weld design are ba•edon unrealistically large moment and a small shear.

The design procedures proposed by the authors are onlya step in direction of improving the design methods by us-ing more realistic test results and failure modes. Much workneeds to be done in this area particularly with respect to cy-clic behavior of these connections.

700

i 600

5OO

400

Eo 300

.__ 2oo_J

0 100CD

0

iI '•

[ •- eb=O

I ' , " , " T . • i F - / • , : , I ' '20 40 60 80 100 200

....... UA's Analysis

•- ,; ; • UA's Design Line

. . . . UCB Test Result

UCB's Design Line

I I I I i I I

300

Uniform Load, W , Kips

400

1500

1400 --

1300' i 1200

1300

1000

900

800

700

•_ 600

500

400•2 300-• 200

100

. . . . . . . UA's Analysis

- - - - - UA's Design Une

Ultimate Strer•gth / . . . . UCB Test Result

of P/ot•.•.. • / • - - UCB's Design Line

Y i e ' d S t r e n g t h / • • m

I I I I I I I I I I' I I • I I • I I I'20 40 60 80 100 200 300

Uniform Load, W , Kips

4OO

Fig. 4. Variation of shear and moment along the bolt line. Fig. 5. Variation of shear and moment along the weld line.

5

ACKNOWLEDGMENTS

The opinions expressed in this closure are those of the authorsand do not necessarily reflect the views of the AISC or theUniversity of California at Berkeley. The words "UCBmethod" and "UA method" are only used to refer to the

methods developed by the authors and by the researchersat the University of Arizona respectively.

sign of Single Plate Framing Connections with A307Bolts," AISC Engineering Journal, Vol. 19 (No. 4).

10. Wyss, U., "Single Plate Connections for Steel Beams,"Thesis presented to University of British Columbia,April 1967.

11. American Institute of Steel Construction, Manual ofSteel Construction, 9th ed., Chicago: AISC, 1989.

NOTATION

a Distance between bolt line and weld line, in.dt, Diameter of bolts, in.et, Eccentricity of beam reaction from bolt line, in.e,, Eccentricity of beam reaction from weld line, in.L Length of shear tab, in.M•, Moment along bolt line, kip-in.n Number of bolts.R, Reaction of the beam causing yielding of shear tab, kips.R,, Reaction of the beam causing fracture of net section,

kips.t Thickness of shear tab, in.W Total load carried by the beam, W = 2R, kips.

REFERENCES

1. Astaneh, A., "Experimental Investigation of Tee-Framing Connection," Progress Report, submitted toAmerican Institute of Steel Construction, April 1987.

2. Astaneh, A., "Demand and Supply of Ductility in SteelShear Connections," Review in Journal of Steel Con-

' struction Research, March 1988.3. Astaneh, A., K. M. McMullin, and S. M. Call, "De-

sign of Single Plate Framing Connections," Report No.UCB/SEMM-88/12, Department of Civil Engineering,University of California-Berkeley, July 1988.

4. Astaneh, A., K. M, McMullin, and S. M. Call, "De-sign of Single Plate Shear Connections," AISC Engi-neering Journal, Vol. 26 (No. 1).

5. Astaneh, A. and M. Nader, "Behavior and Design ofSteel Tee Framing Connections," Report No. UCB/

SEMM-88/ll, Department of Civil Engineering, Uni-versity of California-Berkeley, July 1988.

6. McMullin, K. M., and A. Astaneh, '•Analytical and Ex-perimental Investigations of Double-Angle Connec-tions;' Report No. UCB/SEMM-88/14, Department ofCivil Engineering, University of California-Berkeley,August 1988.

7. Richard, R. M., "Single Plate Framing ConnectionDesigns," Steel Tips, Steel Committee of California,December 1986.

8. Richard, R. M., P, E. Gillett, J. D. Kriegh, and B. A.Lewis, "The Analysis and Design of Single Plate Fram-ing Connections," AISC Engineering Journal, Vol. 17(No. 2).

9. Richard, R. M., J. D. Kriegh, and D. E. Hormby, "De-

AISC Commentary on Design of Shear Tabs

AISI and AISC sponsored research on single shear plate con-nections (shear tabs) at the University of Arizona in the late1970s and early 1980s. At the request of the AISC Commit-tee on Manuals and Textbooks and the ASCE Committee onSteel Building Structures, additional research was conductedat the University of California-Berkeley in 1988-89. In eachcase, the project scope and limit state criterion was suggestedby AISC and followed by the researchers. Because the limitstate was different in the two cases, the design procedureresulting from each research effort is different. This is evi-dent by the two preceding discussions in this issue of theEngineering Journal, AISC assumes responsibility for thesechanges in the context of a natural evolution of research andimproved understanding of shear tab behavior.

In the University of Arizona case, AISC directed the limitstate to be a maximum connection rotation in this initial re-search on shear tab connections. Because AISC did not re-quest tests to destruction, none were made. On this basis,tests and analytical studies were made and a design proce-dure appearing in several AISC publications was developed.

In the recent University of California-Berkeley case, thelimit state was changed to ultimate load, to be determinedby testing to destruction. Based on this work and previousresearch, a different design procedure was then developedby applying a conservative factor of safety.

The AISC Committee on Research and the AISC Com-mittee on Manuals and Textbooks determined that the ulti-mate load criterion given to the University of California-Berkeley was more realistic and better represented the be-havior traditionally assumed for steel connections. The ASCECommittee on Steel Building Structures concurred in thisjudgment.

AISC feels that both shear tab design procedures includean adequate factor of safety and either can be safely used.Because of the simpler nature of the new University ofCalifornia-Berkeley method, and because its strength limitstates are considered to be more complete and realistic, thatmethod was adopted for inclusion in the Ninth Edition ofthe Manual of Steel Construction. Additional research on thismethod to expand its applicability to other detailing condi-tions is in progress.

AISC expresses its appreciation to both Professor Richardand Professor Astaneh for their contributions to the solu-tion of this vexing design problem.

6

Design of Single Plate Shear Connections*ABOLHASSAN ASTANEH, STEVEN M. CALL AND KURT M. McMULLIN

INTRODUCTION

Single plate shear connections, often referred to as sheartabs, have gained considerable popularity in recent yearsdue to their efficiency and ease of fabrication. Shear tabconnections are primarily used to transfer beam end reac-tions to the supporting elements. The connection consistsof a plate welded to a support at one edge and bolted toa beam web. Figure 1 shows typical applications of singleplate shear connections. This paper presents the summaryof a research project on the behavior and design of singleplate shear connections. Based on experimental and aha-lyrical studies, a new design procedure is developed andpresented.

The AISC-ASD •5 as well as AISC-LRFD •6 specifica-tions have the following provisions with regard to shearconnections:

"Except as otherwise indicated by the de-signer, connections of beams, girders, ortrusses shall be designed as flexible, and mayordinarily be proportioned for the reactionshears only.

"Flexible beam connections shall accommo-date end rotations of unrestrained (simple)beams. To accomplish this, inelastic action inthe connection is permitted."

Steel shear connections not only should have sufficientstrength to transfer the end shear reaction of the beam butaccording to above provisions, the connections should alsohave enough rotation capacity (ductility) to accommodatethe end rotation demand of a simply supported beam. Inaddition, the connection should be sufficiently flexible sothat beam end moments become negligible. Thus, like anyshear connection, single plate shear connections should bedesigned to satisfy the dual criteria of shear strength androtational flexibility and ductility.

Shear-Rotation Relationship in a Shear Connection

To investigate the behavior and strength of a shear con-nection, it is necessary that realistic shear forces and theircorresponding rotations be applied to the connection. In

Abolhassan Astaneh is assistant professor, University of Califor-nia, Berkeley.

Steven M. Carl is graduate research assistant, University ofCafifornia, Berkeley.

Kurt M. McMullin was graduate research assistant, University ofCafifornia, Berkeley.

an earlier research project? the shear-rotation relation-ship for the end supports of simply supported beams wasstudied. A computer program was developed• and used tosimulate increased monotonic uniform loading of thebeams supported by simple connections until the beamscollapsed. 1,2

The studies indicated that the relationship between theend shear and end rotation is relatively stable and dependsprimarily on the shape factor Zx/Sx of the cross section,L/d of the beam and the grade of steel used. Figure 2shows a series of curves representing shear forces and cor-responding rotations that will exist at the ends of simplysupported beams. The curves correspond to beams of A36steel having cross sections from W16 to W33 and L/d ra-tios of 4 to 38. Also shown in Fig. 2 is a tri-linear curve"abcd" suggested to be a realistic representative of theshear-rotation curves. The tri-linear curve "abcd" is pro-posed to be used as a standard load path in studies of shearconnections. Curve "abcd" is used instead of the moreconservative curve "aef" because it is felt that curve"abcd" represents a more realistic maximum span-to-depth ratio for most steel structures. For special cases ofvery large span-to-depth ratio or high strength steels, therotational demand may be greater than that of curve"abcd". For such cases special care must be taken to as-sure the rotational ductility demand of the beam is sup-plied by the connection.

CONCRETE SUPPORT COLUMN

(o) (b)

(d) (e)

BEAM BEAM

(c)

Fig. 1. Typical Single Plate Shear Connections

'k Reproduced from AISC Engineering Journal, Volume 26, No. 1, 1st Quarter 1989

7

The shear-rotation curves plotted in Fig. 2 are estab-lished based on the assumption of elastic-perfectly-plasticbending moment capacity for the beam. To include the ef-fect of strain hardening, the segment "cd" in curve "abcd"is included.

The behavior of shear connections has been studied inthe past by several investigators.83°42 However, in mostcases, the shear connections have been subjected to mo-ment and rotation or only direct shear without rotation in-stead of a realistic combination of shear and rotation. Fig-ure 3 shows the shear rotation relationships that existedin several studies including this research project.

EXPERIMENTAL RESEARCH

In order to identify limit states of strength and to verifythe validity of the design procedures that were developedand proposed, five full scale beam-to-column connectionassemblies were tested. A summary of the experimentalstudies follows. More detailed information on the researchproject can be found in References 3 and 6.

Test Set-up

The test set-up shown in Fig. 4 was used to apply shear-rotation relationship of curve "abcd" in Fig. 2 to the speci-mens.

The main components of the test set-up were a com-puter based data acquisition and processing system, twoactuators R and S and support blocks. Actuator S, whichwas close to the connection, was force controlled and pro-vided the bulk of the shear force in the connection. Actua-tor R, which was displacement controlled, provided andcontrolled the beam end rotation.

Test Load Path

The proposed standard shear-rotation relationshipshown as curve "abcd" in Fig. 2 was applied to the connec-

tions in all of the test specimens. To establish the curve,coupon tests of the plate material were conducted priorto connection tests and the yield point and ultimatestrength of the plate material were obtained. The shearyield capacity of the single plate in each test specimen wascalculated by multiplying the von Miess criterion of shearyield stress, 1/X/3Fi, by the shear area of the plate. Theshear yield capacity of the plate, denoted as Ry, was takenas equal to the shear at point "c" of curve "abcd" in Fig.2. Thus the shear yield capacity of the shear tab was as-sumed to occur when the moment at midspan was equalto flip. As a result, a corresponding Mp can be calculatedfor each connection to be equal to RyL/4. The end rota-tion of the beam when midspan moment reached Mp wasset equal to 0.03 radians.

To establish point "b" in curve "abcd", the shear at thispoint was set equal to 4My/L and the rotation was setequal to 0.02 radian. This implies that when beammidspan moment reaches My, the end rotation will beequal to 0.02 radian. The value of My, the end rotationwill be equal to 0.02 radian. The value of My for each spec-imen was calculated by dividing Mp by the shape factor.A shape factor of 1.12 was used in all specimens.

Segment "cd" in Fig. 2 corresponds to strain hardeningof the beam and the increased moment at beam midspanwhich results in increased shear at the beam ends. To es-tablish "cd", it was assumed that when the midspan mo-ment reaches a value of (F,/Fy)Mp, the beam end rotationwill be equal to 0.1 radian.

In summary, load path "abcd" in Fig. 2 reflects the be-havior of the beam and its effect on connection shear androtation. Segment "ab" corresponds to the elastic behav-ior of beam. At point "b", midspan moment of the beamreaches My and the beam softens. Segment "bc" corre-sponds to inelastic behavior of the beam. At point "c", themidspan moment reaches Mp. Segment "cd" representsextra beam capacity that can develop due to beam strainhardening.

!6

!41 dLO=4 C

' • 'f(: /,',,. •//',//•, ( / / o Legend:-I/J w3o - -

06 _...•__•_d•'.•/,'/g'Y'• ?' w2r,,, . , , ,, , ¥ , , . ,. L . o . 3 8 W24 - - -

• ( / W•8 ....02

-•1• a Vy=4SxFy L

oo !Ir I I II t I

O0 001 002 003 004 005 006 007

END ROTATION, r a d .

Fig. 2. Shear-Rotation Relationship for Ends of Simple Beams

g

Z

9F-O

Zz 10000ZO

Ud

fao

F/g. 3.

,•-• Ref 8

s•EAR

11 ROI 12 .•[ . . . . . . . . . . .J ,4,

0.0 002 0 04 006

ROTATION OF BEAM END, r a d .

Shear-Rotation Relationship used in Several Studies

8

Table 1.Properties of Test Specimens

TEST TEST NO. OF DIA. OF TYPE OF PLATE EDGE ACTUAL BEAM PLATEGROUP NO. BOLTS BOLTS BOLTS* DIMENSIONS DISTANCE W E L D MATERIAL MATERIAL

SIZE

in. in. x in. x in. in. in.

ONE 1 7 3/4 A325-N 21 x 3/8 x 4-Y4 1-V2 1/4 A36 A36

2 5 % A325-N 15 x % x 4-Y4 1-V2 '/4 A36 A363 3 % A325-N 9 x % x 4-Y4 1-V2 % A36 A36

TWO 4 5 3/4 A490-N 14-V3 x % x 3-% 1-Va %2 Gr. 50 A365 3 3/4 A490-N 8-% x 3/8 x 3-% 1-V8 %2 Gr, 50 A36

*All bolts were tightened to 70% of proof load. In all specimens diameter of bolt hole was •6 inch larger than nominal diameter of bolt. indicates thatin al l specimens threads were included in shear plane.

.1:Size of all welds was specified as • inch.

Test Specimens

Each test specimen consisted of a wide flange beambolted to a single plate shear connection which was weldedto a column flange as shown in Fig. lb. The properties ofthe test specimens were selected in consultation with aprofessional advisory panel. These properties are given inTable 1. The bolt holes in all specimens were standardround punched holes. All bolts were tightened to 70% of

proof load using turn-of-the-nut m e t h o d ? '•4 All sheartabs were cut from a single piece of steel. The yield stressand ultimate strength for material of shear tabs were 35.5ksi and 61 ksi respectively. The condition of faying sur-faces was clean mill scale. The electrodes were equivalentof E7018.

The bolt spacing in all specimens was 3 in. The edge dis-tance in the horizontal as well as vertical direction forspecimens 1,2 and 3 was 1-1/2 in. (two times diameter ofbolt) and for specimens 4 and 5 was l-lA in. (1.5 times di-ameter of bolts).

CONTROLCOMPUTER GRAPHICS ANALOG

i•'•,• • PLOTTERACTUATOR ACTUATOR PRINTERCONTROLLER CONTROLLER['"'"'"3

FEEDBACK

TO INSTRUMENTAION

^OTUATOR' :V !7

SINGLE PLATE 4•1•:-i::! :: :-:-ii:.:]SHEAR CONNECTION / •:i-i!i;!i :¥.•- ]

COLUMN / li:.i-:.!:! i<: • : ]REACTION BLOCK j

Fig, 4. Test Set-up Used in Experiments

Behavior of Test Specimens

The experiments were conducted in two groups as indi-cated in Table 1. The main differences of specimens inthese two groups were the type of bolt (A325 or A490),material of beam (A36 or grade 50) and edge distance (2dbor 1.5db). The behavior of specimens in the two groupsis summarized in the following sections.

Behavior of Specimens 1,2 and 3 (Group One)

Specimens 1,2 and 3 showed very similar behaviorthroughout the loading. The most important observationwas the significant inelastic shear deformations that tookplace in all three specimens as shown in Fig. 5.

All test specimens failed due to sudden shear fractureof the bolts connecting the single plate to the beam webas shown in Fig. 6a. The examination of bolts after failureindicated that the A325 bolts in these specimens had de-veloped significant permanent deformations prior to frac-ture as indicated in Fig. 6b. In these three specimens thewelds did not show any sign of yielding other than in speci-men 3 which showed minor yielding at the top and bottomof welds prior to fracture of bolts.

A study of the bolt holes after the completion of tests1,2 and 3 indicated that permanent bearing deformationshad taken place in the plate as well as in the beam web.The magnitude of the deformations in the plate and beam

9

bolt holes were almost equal but in opposite directions.The deformations of the plate bolt holes, drawn to scaleare shown in Fig. 7. The arrows indicate the direction ofthe movement of the bolts which is expected to be approx-imately the direction of the applied force due to shear andmoment. It is interesting to note that nearly vertical orien-tations of arrows indicate the presence of a large verticalshear accompanied by a relatively small moment in theconnections.

Behavior of Specimens 4 and 5 (Group Two)

The behavior of specimens 4 and 5 was similar to theprevious three tests. However, shear yielding of the platewas more apparent. Specimen 4 failed due to shear frac-ture of bolts in a manner similar to previous tests shownin Fig. 6a. In addition, minor yielding was observed on theweld lines of this specimen. Specimen 5 failed by almostsimultaneous fracture of weld lines and bolts as shown inFig. 8. It appears that at the time of failure, weld linesstarted to fracture first while bolts were on the verge offracture. When sudden fracture of welds occured the re-sulting shock caused fracture of the bolts which appearedto be almost simultaneous with weld fracture. Bolts inspecimens 4 and 5 were A490 bolts. An examination of thebolts after fracture showed less permanent deformationsin these bolts than the A325 bolts used in previous threetests (see Fig. 6b).

Study of bolt holes in the shear tabs of specimens 4 and5 indicated that significantly larger bolt hole deformationshad occured in these two specimens compared to speci-mens 1,2 and 3. However, the bolt holes in the beam webin specimens 4 and 5 had only minor permanent deforma-tions.

In summary, based on observations made during the

OiIIq)

O

<Y

43+

TEST ONE TEST TWO TEST THREE

Fig. 5. Plate Shear Deformations in Specimens 1,2,3

tests, it appears that shear tabs go through three distinc-tive phases of behavior. At the very early stages, a sheartab acts as a short cantilever beam with moment beingdominant. As load increases, the shear tab acts as a deepshear beam with the shear yielding effect dominant (as inspecimens i through 4). If bolts and welds do not fail dur-ing the shear phase, because of large deformations, theshear tab acts similarly to the diagonal member of a trussand carries the applied shear by a combination of shearand diagonal tension effects (as in specimen 5),

Experimental Data

The results of experiments at the time of failure aresummarized in Table 2.

DISCUSSION OF EXPERIMENTAL RESULTS

Shear Yielding of Single Plate

The yielding of the single plate was primarily due toshear stresses and was quite ductile. It was evident thatconsiderable shear yielding occurred in the plate betweenthe bolt line and weld line. The shear yielding was almostuniformly distributed throughout the depth of the plate asmeasured by strain gages that were attached to theplates. 3'0 Therefore, in the proposed design procedure dis-cussed later, the shear capacity of plate is calculated bymultiplying gross area of plate by uniformly distributedshear stresses.

In specimen 3, at later stages of loading and after signifi-cant shear yielding, the bottom portion of the shear tabshowed signs of minor local buckling as shown in Fig. 6a.This local buckling was attributed primarily to loss of stiff-ness of plate material due to shear yielding. Until this phe-

A325 Bolts(15 Bolts Tested)

Fig. 6.

A490 Bolts(8 Bolts Testedl

(a) (b)

Typical Bolt Failure of Test Specimens

10

Table 2.Results of shear strength Tests

Specimen Observed Connection ResponseFailure Mode $ $ Maximum

Test Test No. of Shear Shear Beam End Momentat Moment at Moment atGroup No. Bolts Displacement Force Rotation BoltLine WeldLine Weld Line

in. kips rad. kip in. kip in.

(1 ) (2) (3) (4) (5) (6) (7) (8) (9) (1O)

1 7 Bolts Fractured 0.27 160 0.026 306 745 1028

One 2 5 Bolts Fractured 0.34 137 0.054 314 691 734

3 3 Bolts Fractured 0.46 94 0.056 20 279 350

Two 4 5 Bolts Fractured 0.35 130 0.053 273 631 686

5 3 Welds and Bolts Fractured 0.52 79 0,061 -47 170 237

* In some cases like these, moment decreased as shear and rotation increased.•Positive moments cause top of connection to be in tension.

nomenon is studied thoroughly, it is suggested that localbuckling be avoided. To prevent local buckling, it is rec-ommended that the distance between the bolt line and theweld line be less than 1/2 of the plate length.

Fracture of Net Area of Plate

In the single plate specimens that were tested, the netarea of the plate did not fracture. Only specimen 5 showedsigns of approaching fracture of net section. Nevertheless,this failure mode has been observed in similar cases in sev-eral experiments on tee framing connections. 4'5 The stemin a tee framing connection behaves similarly to a sheartab. The formula currently used in calculating net area inshear fracture is: ]5

Ans = Avg-n(db +l/16)tp (1)

The studies of tee connections indicated that the shearfracture occurred consistently by fracture of net sectionalong the edge of the bolt hole and not along thecenterline of bolts. It was suggested that4,s the net area ef-fective in shear be equal to the average of net area alongthe bolt centerline and the gross area. Using the suggestedmethod to calculate net area in shear, the effective netarea in shear can be written as:

Anse = Avg-(n/2)(db +V]o)tp (2)

Shear-Rotation Behavior

Figure 9 shows the actual shear-rotation relationshipthat was recorded during each test. It is observed that therotational ductility of the connections increased as the

Fig. 7. Plate Bolt Hole Deformations after Tests Fig. 8. Failure of Welds and Bolts in Specimen 5

11

number of bolts decreased. The rotational ductility of theconnection in specimen ! with 7 bolts was 0.026 radianswhich was about half the rotational ductility of the connec-tions in specimens 2, 3, 4 and 5 with three or five bolts,all of which were able to reach rotations in excess of 0.05radians.

Movement o f Point of Inflection

Figure 10 shows movement of point of inflection of thebeam toward the support as the shear force was increased.Even under relatively small load, in all specimens, thepoint of inflection moved toward the support and re-mained almost stationary for the remainder of each test.

Using experimental data, the following empirical equa-tion was developed to define the location of the point ofinflection for test specimens.

e = (n-I)(1.0), in. (3)

where n is the number of bolts used in the connection, ande is the distance of point of inflection from the support(i.e. from the weld line).

It is important to realize that in the experiments re-ported here, the columns were fixed to supports and rigidbody rotation of the connections was prevented. If due toframe action or other causes, the support to which a sheartab is connected rotates, due to rigid body rotation, thelocation of point of inflection may be affected. However,the concurrent values of shear and moment acting on theshear tab at any given time cannot exceed the values ob-tained from plasticity conditions (interaction curves) ofplate for shear and moment.

Behavior and Design of Bolts

In all specimens, an examination of bolts and bolt holesafter failure indicated that bolt shanks had experiencedconsiderable shear deformations before failure.

Studies on the behavior of single bolts in shear• haveindicated that for A325 bolts and A36 plate, if the thick-ness of the plate is not greater than 1/2 times the diameterof the bolt, considerable but tolerable bolt hole deforma-tions will take place. The limited bolt hole deformationsare desirable since they increase rotational flexibility andductility of the connections. In studies of tee connec-tions4'5 in three specimens, V2 in. thick tee stems wereused with 7/8 in. diameter bolts. The behavior of these teespecimens indicated that even when thickness of stem wasequal to db + V•6 in., desirable bearing deformations tookplace in the bolt holes. Therefore, based on these studies,and to obtain flexible and ductile single plate connections,the thickness of the plate is recommended to be less thanor equal to 1/2 of the bolt diameter plus 1/16 in.

An examination of the deformations of bolts and boltholes at the completion of the tests indicated that the boltswere primarily subjected to direct shear accompanied bya small moment (see arrows in Fig. 6a).

As Fig. 10 indicates, the point of inflection for test spec-imens was almost stationary, fluctuating between an ec-centricity of n and n - 1 in. At the time of failure of thebolts in all specimens, the location of the point of inflec-tion was close to n - 1 in. Therefore, it is recommendedthat bolts be designed for combined effects of direct shearand a moment equal to the shear multiplied by the eccen-tricity of the bolt line from point of inflection given by:

eb = ( n - 1 ) ( 1 . 0 ) - a (4)where,

a = distance between the bolt line and weld line,eb = distance from the point of inflection to the bolt

line.

Behavior and Design of Welds

Table 2 gives values of shear and moment at failure foreach test. The fillet welds mainly experienced a direct

200-

180;

1601

03 1401Cl.."• 1201a/ ',<• 100.LU"1- 80.

60

40

20-'

0'

TEST 1 0 V •f (7-Bolt)

/ :::':4 ...... . . . . . . . . .

/ / / •_---Tost3

· % " ; " ' " " . . . . . . . ,°.. . . . . . . ( 3 B o l t )

0.02 004 0 06

ROTATION AT BOLT LINE, rad.

g

of<I:W

OO

200-1 7" ,• . . . . Test 1 (7 A325 Bolts)

I -- - - - Test 2 (5 A325 Bolts)I I · I - - - - - Test 3 {3•A325 Bolts)

150 1 q" J(I Test4(5.A490Bolts)

lOO. Li(',

0 . . . . . . . . . . . . . . . . . •,, , • - • - • ' ; -

fewf

v

Point ofInflection

I •-"mllMomenl

10 20 30

DISTANCE OF PT. OF INFL. FROM WELD, in.

4O 5O

Fig. 9. Shear-Rotation Curves for Test Spectmens Fig. 10. Movement of Point of Inflection

12

shear accompanied by a relatively small moment. Thestrain measurements adjacent to the welds also supportedthis conclusion. 3'6 Therefore, fillet welds are recom-mended to be designed for the combined effects of shearand a small bending moment.

The main goal of the proposed design procedure is toensure yielding of shear tab prior to failure of welds. Inorder to achieve this goal the welds should be designed tobe stronger than the plate. Thus, the design shear forceacting on the welds is recommended to be equal to theshear capacity of the plate and not the applied shear force.Therefore, the maximum shear force acting on the weldis equal to 1/V'-3 FyLptp. In Allowable Stress Design, thedesign shear force for welds is equal to 0.40FyLvtp. Themoment acting on the weld is equal to shear force multi-plied by the eccentricity of the point of inflection from theweld line. To be conservative, it is recommended that theeccentricity of the point of inflection from the weld linebe equal to n inches,

ew = (n)(1.0) (5)

Since the design of welds in the proposed method is acapacity design, it is not necessary to use welds that canresist forces much greater than the plate capacity. As partof phase two of this investigation, a study was conductedto establish minimum and maximum weld requirements todevelop the strength of single plate. The study indicatedthat for A36 plate and E70 electrodes the weld size neednot be more than 0.75tv and should not be greater thantv . The upper limit of tv on the weld size was imposed toprevent excessive welding of the plate which will be costlyand might cause heat damage to the plate without achiev-ing extra strength in the connection.

Moment-Rotation Curves

Moment-rotation curves for the test specimens areshown in Fig. 11. Moments and rotations were measured

1200-6

lO00-Z.-I

800-,,-ILLI

600-

I--

l,,-, 400-ZLIJ

200-0

00

Specimens with A325 Bolts

Specimens with A490 Bolts

Test 1

(7-Bolt) . • Test 2

. . . . . . . %' Test 4(5-Bolt)

Test 3(3-Bolt)

/ ¥ / • . . . . . . . . . . . . . . . . . . . . ' - Test 5- - - (3-Bolt)

' o . b 2 ' & 4 ' 0.•6 '

ROTATION OF BEAM END, rad.

Fig. 11. Moment-Rotation Curves for Test Specimens

along the bolt line. As the plots indicate, connections withfewer bolts developed smaller moments and exhibitedlarger rotational ductility. During the elastic range of be-havior, moment increased with shear. As the load in-creased, due to connection deformations, rotational stiff-ness and bending moment decreased and then graduallyincreased at a much smaller rate. The decrease is attrib-uted to slips and inelastic deformations in the connectionsand the increase is attributed to strain hardening.

PROPOSED DESIGN PROCEDURE

The following design procedure is based on the analysesof the experimental results and the information availableon the actual behavior of shear connections?6'9

General Requirements

The single plate framing connections covered by theseprocedures consist of a plate bolted to a beam web andwelded to a support on one edge of plate.

In design of a single plate framing connection, the fol-lowing requirements should be satisfied:

1. The cortnection has only one vertical row.of boltsand the number of bolts is not less than 2 or morethan 7.

2. Bolt spacing is equal to 3 in.3. Edge distances are equal to or greater than 1.5db.

The vertical edge distance for the lowest bolt is pre-ferred not to be less than 1.5 in.

4. The distance from bolt line to weld line is equal to3 in.

5. Material of the shear plate is A36 steel to facilitateyielding.

6. Welds are fillet welds with E70xx or E60xx elec-trodes.

7. Thickness of the single plate should be !ess than orequal to db/2 + 1,/16.

8. The ratio of Lp/a of the plate should be greater thanor equal to 2 to prevent local buckling of plate.

9. ASTM A325 and A490 bolts may be used. Fullytightened as well as snug tight bolts are permitted.The procedure is not applicable to oversized or longslotted bolt holes. Standard or short-slotted punchedor drilled holes are permitted.

Consideration of Limit States in Design

The following limit states are associated with the singleplate framing connections.

1. Shear failure of bolts.2. Yielding of gross area of plate.3. Fracture of net area of plate.4. Fracture of welds5. Bearing failure of beam web or plate.

13

Shear Failure of Bolts

Bolts are designed for the combined effects of directshear and a moment due to the eccentricity eb of the reac-tion from the bolt line. The eccentricity eb for single plateconnections covered by these procedures can be assumedto be equal to 3 in., which is the distance from bolt lineto weld line. The value is conservative when the singleplate is welded to a rigid support. The value is more realis-tic when the supporting member is a relatively flexible ele-ment.

More realistic values for eb can be calculated from thefollowing equations:

if single plate is welded to a rotationally rigid element, ebis obtained from:

eb = (n- 1)(1.O)-a (6)

if single plate is welded to a rotationally flexible element,eb is larger value obtained from:

eb = Max(n- 1)(1.0) - a (7a)

a (7b)

where,n = number of boltsa = distance from bolt line to weld line, in.eb = eccentricity, in.

By using methods outlined in Reference 7 includingusing Tables X of the AISC-ASD Manual •3 the bolts aredesigned for the combined effects of shear R, and momentequal to R%.

Yielding of Gross Area of Plate

The equation defining this limit state in allowable stressdesign (ASD) format is:

fry < F•y (8)where,

fry = R / Avg (9)

Fry = 0.40 Fy (10)

Avg = Lp tp (11)

Fracture of Net Area of Plate

The equation defining this limit state in allowable stressdesign (ASD) format is:

fy. -< F•u (12)where,

fuu = R /Ans (13)

Fvu = 0.30/:•, (14)

Ans= [Lp -- n(db + 1/16)]tp (15)

If the beam is coped, the block shear failure of the beamweb also should be considered as discussed in the AISC-ASD Specification. •5

Weld Failure

The welds connecting the plate to the support are de-signed for the combined effects of direct shear and a mo-ment due to the eccentricity of the reaction from the weldline, ew. The eccentricity ew is equal to the larger valueobtained from:

ew = Max(n)(1.0) (16a)

a (16b)

where,n = number of boltsew = eccentricity, in.a = distance from bolt line to weld line, in.

By using methods outlined in Reference 7 includingusing Tables XIX of the AISC-ASD Manual, 13 the filletwelds are designed for the combined effects of shear equalto R and moment equal to Re,,.

Bearing Failure of Plate or Beam Web

To avoid reaching this limit state, it is recommendedthat the established rule of horizontal and vertical edgedistances equaling at least 1.5 the bolt diameter be fol-lowed. The bolt spacings should satisfy requirements ofthe AISC-ASD Specification. •5 The bearing strength ofconnection can be calculated using the provisions of theAISC-ASD Specification. •5

Summary of Design Procedure

The following steps are recommended to be taken in de-sign of single plate framing connections:

1. Calculate number of bolts required to resist combinedeffects of shear R, and moment R% using Table X of theAISC-ASD Manual. 13

If the single plate is welded to a rotationally rigid sup-port eb is the value obtained from Eq. 6.

If the single plate is welded to a rotationally flexible ele-ment, eb is the value obtained from Eq. 7:

2. Calculate required gross area of plate:

Avg • R / 0.40Fy (17)

Use A36 steel and select a plate satisfying the followingrequirements:

a. lh and l,, • 1.5db.b. Lp--> 2ac. t, -< db/2 + V•6d. ti, --> Ava/Lpe. Bolt spacing =3 in.

(18)(19)(20)(21)

14

3. Check effective net section:Calculate allowable shear strength of the effective net

area:

Rns=[tp-n(db+Vl6)](¥)(o.3eu) (22)

and satisfy that R.• -> R.

4. Calculate actual allowable shear yield strength of theselected plate:

Ro = Lptp (0.40Fy) (23)

Design fillet welds for the combined effects of shear Roand moment Roew using Table XIX of the AISC Manual?ew is given in Eq. 16 as:

ew = Max ]

i

(n)(1.0) (16a)

I a (16b)

The weld is designed for a capacity of Ro, and not forthe applied R, to ensure that the plate yields before thewelds. However, for A36 steel and E70 electrodes theweld size need not be larger than 3/4 of the plate thickness.

5. Check bearing capacity of bolt group:

(n)(t)(db)(1.2Fu) > R (24)

If the bolts are expected to resist a moment (as they nor-mally would), this calculation should reflect the reducedstrength as determined by Table X of the AISC Manual•3as demonstrated in the following examples.6. If the beam is coped, the possibility of block shear fail-ure should be investigated.

Application to Design Problems

The following examples show how the design procedurecan be implemented into the design of steel structures.

Design Example 1

Given:Beam:Beam Material:Support:Reaction:Bolts:Bolt Spacing:Welds:

W27 x 114, t• = 0.570 in.A36 steelColumn flange (Assumed rigid)102 kips (Service Load)7/8 in. dia. A490-N (snug tight)3 in.E70XX fillet welds

Design a single plate framing connection to transfer thebeam reaction to supporting column.

Solution:

1. Calculate number of bolts:Shear = R = 102 kipsLet us assume M = 0, (will be checked later)n = R/rv= 102/16.8 = 6.1

Try 7 boltsThe distance between the bolt line and the weld linea is selected equal to 3 in.Check moment:e0 = ( n - 1 ) l . 0 - a = 7 - 1 - 3 = 3.0 in.Moment = 3 x 102 = 306 kip-in.

Using Table X of the AISC•ASD Manual•3 with eccentric-ity of 3 in., a value of 6.06 is obtained for effective numberof bolts (7 bolts are only as effective as 6.06 bolts).

Therefore,

Rbolt = 6.06 x 16.8 = 101.8 -• 102 kips O.K.

Use: Seven • in. dia. A490-N bolts.

2. Calculate required gross area of the plate:

Avg = R / 0.40FyAvg = 102/(0.40 x 36) = 7.08 in.2

Use A36 steel and select a plate satisfying the followingrequirements:

a. lh and l• > 1.5dblh = lv = 1.5(7/8) = 1.32 in.W = a +Ih = 3 + 1.32= 4.32; useW=

4V2 in.b. Lp/a -> 2.0

Lp = 2 x 1.32 + 6 x 3.0 = 20.6 in.; use Lp =21 in.

Check:Lp/a= 21/3 = 7 > 2 O.K.c. tv < db/2 + 1/16

tp -- (7/8)/2 + 1/16 = V2 in.

d. tv = Aug/L.tv = 7.08/21 = 0.337 in.Try PL 21 x 3/8 x 4-1/2

3. Calculate allowable shear strength of the net area:Rns = [Lp-n(db + IA6)](tp)(O.3Fu)R•, = [21-7(% + 1/16)](¥8)(0.3 × 58) = 94 < 102

kips N.G.Try V2 in. thick plate:R•s = [21-7(7/8 + 1A6)](I/2)(0.3 x 58) = 125 > 102

kips. O.K.

Use: PL 21x¼x4%, A36 Steel.

4. Calculate the actual allowable yield strength of the se-lected plate:

Ro = Lptp (0.40Fy)Ro = 21 x 0.5 x 0.40 x 36 = 151 kips

Design fillet welds for the combined effects of shearand moment:

Shear = Ro = 151 kips

[ n(1.0) = 7(1.0) = 7 in.ew Max

I a = 3in.

Therefore, ew = 7.0 in.Moment = Roe• = 151 x 7 = 1057 kip-in.Using Table XIX AISC Manual•3

1 5

a = 7/21 = 0.333C1 = 1.0C = 1.07Dl6 = Ro/CC/Lp = 151/(1.0 x 1.07 x 21). = 6.72

Since weld size need not be greater than 0.75tp,

Use: % in. E70 Fillet Welds.

5. Check bearing capacity:For plate:

rv = drip (1.2Fu) = .875 x .5 x 1.2 x 58 = 30.45

Rbrg = 6.06(30.45) = 184.5 kips > 102 kips. O.K.Since the beam web is thicker than the plate, the webwill not fail.

6. Beam is not coped, therefore, there is no need for con-sideration of block shear failure.

Design Example 2

Given:Beam:Beam Material:Support:Reaction:Bolts:Bolt Spacing:Welds:

W16x31, tw = 0.275A572 Gr. 50 steelCondition of support is unknown33 kips (Service Load)3/4 in. dia. A325-N or A490 (snug tight)3 in.E70XX fillet welds

Design a single plate shear connection to transfer thebeam reaction to the support.

Solution:

1. Calculate number of bolts:Shear = 33 kipsLet us assume M = 0, (will be checked later)Try A325-N bolts with 9.3 kips/bolt shear capacity:n = R/r, = 33/9.3 = 3.5Try 4 bolts.The distance between bolt line and weld line a is

selected equal to 3 in.Check moment:Since condition of support is not known, the sup-

port is conservatively assumed to be flexible forbolt design. Therefore eb is equal to 3 in.

Moment = 3 x 33 -- 99.0 kip-in.Interpolating from Table X13, C • 2.81Rat/= 2.81 x 9.3 = 26.1 kips<33 N.G.Which indicates 4 A325 bolts are not enough. Let

us try 4 A490-N bolts:Ra//= 2.81 x 12.4 = 34.8 kips>33 O.K.

Use: Four % in. dia. A490-N bolts.

2. Calculate required gross area of plate:Avg = R / 0.40FyAvg = 33/(0.40 x 36) = 2.29 in.2

Use A36 steel and select a plate satisfying the following

requirements:

a. lh and Iv -> 1.5db.Ih = Iv = 1.5(3/4) = 1.125 in.W= a + lh = 3 + 1.125 = 4.125 in.Use: W = 41/2 in.

b. Lp/a -> 2.0Lp = 3 + 3 x 3 = 12 in.Check: Lp/a = 12/3 = 4 > 2 O.K.

c. tp -< db/2 + 1/16tp < (3/4)/2 + 1/16 = ?/16 in.

d. tp = A vg/mptl, = 2.29/12 = 0.19 in.

Use: PL 12x¼x41/2, A36 Steel.

3. Calculate allowable shear strength of the net area:R,s = [Lp-n(db + 1A6)](tp)(O.3F•)R,s = [12 - 4(3/4 + V16)](I/4)(0.3 x 58) = 38.1 kipsR,, -> R is satisfied.

4. Calculate actual allowable yield strength of the selectedplate:

Ro = Lpt, (0.40Fy)Ro = 12 x 0.25 x 0.40 x 36 = 43.2 kips

Design fillet welds for the combined effects of shearand moment:

Shear = Ro = 43.2 kips

(n)(1.0) = 4(1.0) = 4 in.ew = Max

a = 3.0

Therefore, ew = 4.0 in.Moment = Roe,, = 43.2 x 4 = 172.8 kip-in.Using Table XIX AISC Manualt3a -- 4/12 = 0.33

C1 -= 1.0C = 1.07D/6 = Ro/CC/Lp = 43.2/(1.0 x 1.07 x 12) = 3.36

Since weld size need not be greater than 0.75tp,

Use: •6 in. ET0 Fillet Welds.

5. Check bearing capacity.For plate:

nddp (1.2F,) = 2.81 x .75 x .25 x 1.2 x 58= 36.7 kips > 33 kips.

and for beam:nddw(1.2F•) = 2.81 x .75 x .27 x 1.2 x 65

= 44.4 kips > 33 kips.

6. Beam is not coped, therefore, no need for considera-tion of block shear failure.

CONCLUSIONS

Based on the studies reported here, the following con-clusions were reached:1. The experimental studies of single plate connections in-

16

dicated that considerable shear and bearing yieldingoccurred in the plate prior to the failure. The yieldingcaused reduction of the rotational stiffness which inturn caused i•elease of the end moments to midspan ofthe beam.

2. The limit states associated with single plate connectionsare:

a. Plate yielding.b. Fracture of net section of plate.c. Bolt fracture.d. Weld fracture.e. Bearing failure of bolt holes.

3. A new design procedure for single plate shear connec-tions is developed and recommended. The procedureis based on a concept that emphasizes facilitating shearand bearing yielding of the plate to reduce rotationalstiffness of the connection.

4. To avoid bearing fracture, the horizontal and verticaledge distance of the bolt holes are recommended to beat least 1.5 times diameter of the bolt. The study re-ported here indicated that vertical edge distance, par-ticularly below the bottom bolt is the most critical edgedistance.

5. Single plate connections that were tested were veryductile and tolerated rotations from 0.026 to 0.061 radi-ans at the point of maximum shear. Rotational flexibil-ity and ductility decreased with increase in number ofbolts.

ACKNOWLEDGMENTS

The project was supported by the Department of CivilEngineering, the University of California, Berkeley andthe American Institute of Steel Construction, Inc. Thesupport and constructive comments provided by R. O.Disque, N. Iwankiw and Dr. W. A. Thornton are sin-cerely appreciated. Single plates used in the test speci-mens were fabricated and supplied by the Cives SteelCompany. The assistance of R. Stephen, laboratory man-ager, in conducting the experiments was essential and isappreciated.

REFERENCES

1. Astaneh, A., "Experimental Investigation of Tee-Framing Connection", Progress Report submitted to

American Institute of Steel Construction, April 1987.2. Astaneh, A., "Demand and Supply of Ductility in

Steel Shear Connections", Journal of Steel Construc-tion Research, 1989.

3. Astaneh, A., K. M. McMullin, and S. M. Call, "De-sign of Single Plate Framing Connections," ReportNo. UCB/SEMM-88/12, Department of Civil Engi-neering, University Of California, Berkeley, July,1988.

4. Astaneh, A., and M. Nader, "Design of Tee FramingShear Connections," Engineering Journal, AmericanInstitute of Steel Construction, First Quarter, 1989.

5. Astaneh, A., and M. Nader, "Behavior and Design ofSteel Tee Framing Connections," Report No. UCB/SEMM-88/ll, Department of Civil Engineering, Uni-versity of California, Berkeley, July, 1988.

6. Call, S. M., and A. Astaneh, "Behavior of SinglePlate Shear Connections with A325 and A490 Bolts",Report No. UCB/SEMM-89/04, Department of CivilEngineering, University of California, Berkeley,April 1989.

7. Iwankiw, N. R., "Design for Eccent'ric and InclinedLoads on Bolts and Weld Groups," Engineering Jour-nal, American Institute of Steel Construction, 4thQuarter, 1987.

8. Lipson, S. L., "Single-Angle Welded-Bolted. Connec-tions,'' Journal of the Structural Division, March,1977.

9. McMullin, K. M., and A. Astaneh, "Analytical andExperimental Investigations of Double-Angle Con-nections'', Report No. UCB/SEMM-88/14, Depart-ment of Civil Engineering, University of California,Berkeley, August, 1988.

10. Patrick, M., I. R. Thomas, and I. D. Bennetts, "Test-ing of the Web Side Plate Connection," AustralianWelding Research, December, 1986.

11. Richard, R. M., P. E. Gillett, J. D. Kriegh, and B.A. Lewis, "The Analysis and Design of Single PlateFraming Connections," Engineering Journal, Ameri-can Institute of Steel Construction, 2nd Quarter,1980.

12. White, R. N., "Framing Connections for Square andRectangular Structural Tubing, Engineering Journal,American Institute of Steel Construction, July, 1965.

13. American Institute of Steel Construction, Manual ofSteel Construction, 8th Edition, Chicago, 1980.

14. American Institute of Steel Construction, Manual ofSteel Construction. LRFD, 1st Edition, Chicago,1986.

15. American Institute of Steel Construction, Inc., Speci-fication for the Design, Fabrication and Erection ofStructural Steel for Buildings, Chicago, November 13,1978.

16. American Institute of Steel Construction, Inc., Loadand Resistance Factor Design Specification for Struc-tural Steel Buildings, Chicago, September 1, 1986.

17

Arts

An. se

Ay sCc,D•6

VU

L

RRboltR.•RoRy

NOMENCLATURE

Net area in shear, in.2Effective net area of plate in shear, in.2Gross area of plate in shear, in2.Coefficient in the AISC Manual Tables X and XIXCoefficient in the AISC Manual Table XIXNumber of sixteenth of an inch in fillet weld sizeSpecified minimum tensile strength of steel, ksiAllowable shear stress for plate in yielding =0.40Fy, ksiAllowable ultimate shear strength = 0.30Fu, ksiSpecified yield stress of steel, ksiLength of span, in.Length of plate, in.Plastic moment capacity of cross section = ZxFyYield moment of beam cross section, kip-in.Reaction of the beam due to service load, kipsAllowable shear capacity of bolt groupAllowable shear fracture capacity of the net sectionAllowable shear yield strength of plate, kipsReaction corresponding to plastic collapse of beam,kips

$x Section modulus in.3V Shear force, kipsW Width of plate, in.Zx Plastic section modulus, in.3a Coefficient in the AISC Manual Table XIXa Distance between bolt line and weld line, in.d Depth of beam, in.db Diameter of bolt, in.e Eccentricity of point of inflection from the supporteb Eccentricity of beam reaction from bolt line, in.ew Eccentricity of beam reaction from weld line, in.fry Computed shear stress in plate gross area, ksifvu Computed shear stress in plate effective net area,

ksiIn Horizontal edge distance of bolts, in.lv Vertical edge distance of bolts, in.n Number of boltsrv Allowable shear strength of one bolt, kipstp Thickness of plate, in.tw Thickness of beam web, in.

This publication expresses the opinion of the author, and care has been taken to insurethat all data and information furnished are as accurate as possible. The author andpublisher cannot assume or accept any responsibility or liability for errors in the dataor information and in the use of such information.

The information contained herein is not intended to represent official attitudes, recom-mendations or policies of the Structural Steel Educational Council. The Council is notresponsible for any statements made or opinions expressed by contributors to thispublication.

18

STRUCTURAL STEEL EDUCATIONAL COUNCIL

TECHNICAL INFORMATION & PRODUCT SERVICE

OCTO BER 1999

Welded Moment Frame Connections

With Minimal Residual Stress

By

A l v a r o L. Co l l i n and

James J. P u t k e y

Acknowledgments

The Authors wish to thank the following persons for their review and comments on the content of this Steel TIPS:

Pat Hassett, Hassett Engineering Bill Honeck, Forell/Elsesser Engineers, Inc. Dave McEuen, California Erectors, Bay Area, Inc. Larry McLean, McLean Steel Members of the Structural Steel Educational Council

In Memoriam

Alvaro L. Collin

The members of the Structural Steel Educational Council dedicate this Steel TIPS to the memory ofAl Collin. Mr. Collin died on April 26, 1999. He had a long and distinguished career in the structural steel fabrication and erection, especially in the welding of structural steel. Council members and the rest of the structural steel industry will miss AI.

Disclaimer. The information presented in this publication has been prepared in accordance with recognized engineering principles and is for general information only. While it is believed to be accurate, this information should not be used or relied upon for any specific application without competent professional examination and verification of its accuracy, suitability, and applicability by a licensed professional engineer, designer, or architect. The publication of the material contained herein is not intended as a representation or warranty on the part of the Structural Steel Educational Council or of any other person named herein that this information is suitable for any general or particular use or of freedom from infringement of any patent or patents. Anyone making use of this information assumes all liability arising from such use.

Caution must be exercised when relying upon other specifications and codes developed by other bodies and incorporated by reference herein since such material may be modified or amended from time to time subsequent to the printing of this publication. The Structural Steel Educational Council and the authors bear no responsibility for such material other than to refer to it and incorporate it by reference at the time of the initial printing of this publication.

W E L D E D M O M E N T F R A M E C O N N E C T I O N S WITH M I N I M A L R E S I D U A L STRESS

By Alvaro L. Collin and James J. Putkey

C O N T E N T S

ACKNOWLEDGMENTS

TABLE OF CONTENTS

1. INTRODUCTION / Page 1

2. TERMS / Page 2

3. THE PROBLEM / Page 3

4. THE SUGGESTED SOLUTION / Page 9

5. ADVANTAGES AND DISADVANTAGES / Page 14

6. CONCLUSION / Page 16

7. REFERENCES / Page 17

EXHIBITS / Page 18

ABOUT THE AUTHORS / Page 22

1. INTRODUCTION

PURPOSE

The purpose of this Steel TIPS is to address the problem of residual stress in welded moment frame connections with heavy steel members. First, the authors extensively review the residual stress problem. Second, they suggest a solution to the problem.

The Problem. Present design and welded construction practice produces residual stress by:

Joint and connection restraint. Member restraint resulting from two floor erection and welding sequence.

Residual stress results when restraint does not allow welds to shrink.

Suggested Connection. The suggested welded moment frame connection uses a design concept that:

Reduces residual stress in joints and connections to a minimum. Eliminates residual stress from member restraint. Locates the plastic hinge outside the connection.

ORGANIZATION

To accomplish the purpose, the authors present the subject matter of this Steel TIPS as follows:

Terms used in the suggested solution and problem. The problem of residual stress. The suggested solution. Advantages and disadvantages of the suggested solution. A Conclusion.

Welded Moment Frame Connections with Minimal Residual Stress, A.L. Collin and J.J. Putkey, Steel TIPS, October 1999 1

2. ERMS The various Sections use the terms listed below. We took terms marked with an asterisk from the AISC Engineering Journal. [1]

Res idua l S t r e s s

.g Restra/nt ¢'s"

Component Restraint.* Restraint existing because of rigidity of various elements of a joint or connection.

Connection.* Complete assembly consisting of the various joints making up the total unit.

Joint.* A single element of a connection.

Member Restraint.* Restraint in closure member where inherent rigidity requires weld shrinkage to be absorbed by the parent metal.

Residual Stress. Stress remaining in connection or member after completing the connection's welds.

Restraint.* Resistance of the joint or connection to weld shrinkage strains.

Shrinkage. Contraction in the size of a weld joint as it cools.

Strain.* Deformation due to changes in applied forces.

Stress.* Force per unit of area.

Thermal Upsetting. Permanent expansion of weld perpendicular to face of a weld when restraint prohibits a joint from expanding.

Our, We. Pronouns referencing the authors of this Steel TIPS.

*American Institute of Steel Construction, Inc. Reprinted with permission. All rights reserved.

Welded Moment Frame Connections with Minimal Residual Stress, A.L. Collin and J.J. Putkey, Steel TIPS, October 1999 2

3. THE P R O B L E M

This section addresses the causes and locations of residual stress that occur in currently used welded connections.

C A U S E S OF R E S I D U A L S T R E S S

Residual stress results when restraint does not allow large welds in heavy members to shrink. Two conditions cause residual stress:

• Thermal upsetting • Shrinkage

Various parts of this Section 3 set forth restraint conditions and residual stress levels caused by thermal upsetting and shrinkage. The residual stress levels generally correspond to the restraint levels noted in AWS Welding Code, Annex XI. [2]

Thermal Upsetting. As noted in Section 2, thermal upsetting results from permanent expansion of weld metal perpendicular to the face of a weld when fit-up and weld sequence restrain the joint from expanding.

Within Joint. A joint expands from heat input when it is welded. The hot weld will try to push the members apart. However, as the bottom weld layers cool they restrain the members. Heat input in the larger volume of the top weld layers causes thermal upsetting. Low residual stress results from this internal restraint.

By External Restraint. When fit-up and weld sequence restrain the joint from expanding, the weld can only expand perpendicular to the weld surface. Medium or high residual stress results from this perpendicular expansion.

Shrinkage. Weld shrinkage causes most residual stress. The stress level depends on the physical fit-up of the joint.

Within Joint. Fit-up will allow the members of a joint to move. However, as the bottom weld layers in the joint cool, they restrain the members while welding the top layers. Low residual stress results from this internal restraint.

By External Restraint. Fit-up and weld sequence restrain the members of a joint and prevent them from moving. Medium, high, or very high residual stress results from this external restraint. The residual stress level depends on the weld size and restraint level.

Welded Moment Frame Connections with Minimal Residual Stress, A.L. Collin and J.J. Putkey, Steel TIPS, October 1999 3

RESTRAINT AND RESIDUAL STRESS IN JOINTS

This discussion is limited to joints with one plate welded to another plate, beam, or column; shop attached; and with or without external restraint. Plates considered include:

• Continuity Plates • Cover Plates • Shear Plates

Continuity Plates. Continuity plates usually attach to the column web and flanges with groove welds. See Figure 3-1. Medium or high residual stress may occur when welding the plate to the column flanges.

To Column Web. Make these welds first. Restraint: Within joint. Thermal Upsetting and Shrinkage: Low residual stress.

To Column Flanges. Make these welds second. Restraint: External, from flanges. Thermal Upsetting and Shrinkage: Medium or high residual stress depending on the flange thickness.

F 1

Figure 3-1

Cover Plates. Depending on the type of connection, cover plates attach to the column flange with groove welds, or attach to the beam flange or flanges with fillet welds.

To Column Flange. Only bottom plate welded. See Figure 3-2. Restraint: Within joint. Thermal Upsetting and Shrinkage: Low residual stress.

To Beam Flange(s). Cover plates fabricated wider than beam flanges and attached with fillet welds. See Figures 3-3 and 3-4.

Restraint: Within joint. Thermal Upsetting and Shrinkage: Very low residual stress.

!

Figures 3-2, 3-3, and 3-4

Welded Moment Frame Connections with Minimal Residual Stress, A.L. Collin and J.J. Putkey, Steel TIPS, October 1999 4

Shear Plates. Shear plates usually attach to the column flanges with complete penetration groove welds as shown in Figure 3-5. Engineers design the shear plates for either a bolted or a welded connection to the beam web.

To Column Flange. Shop attached. Restraint: Within joint. Thermal Upsetting and Shrinkage: Low residual stress.

Figure 3-5

RESTRAINT AND RESIDUAL STRESS IN CONNECTIONS

Connections consist of field welded or bolted joints acting together. Bolts and fit-up material in the connection usually extend some level of restraint to the joints. The bolts and fit-up material carry erection loads, plumb-up the steel frame, and set root openings for welded joints. Fit -up material may include drift pins, wedges, and small welded plates. For this discussion, we assume the bolts, fit-up material, and weld sequence externally restrain the connection joints while welding.

Connections currently used include: beam flanges welded directly to the column, cover plates on the beam welded to the column, a combination of cover plates and beam flanges welded to the column, and the beam web connection to the shear plate. Residual stress in these connections varies, as discussed in the following groups:

Beam Flanges Beam Cover Plates Beam Cover Plates and Flanges Webs of Beams

Beam Flanges. Beam Section connection. See Figure 3-6. These welds are similar to the welds commonly made prior to the Northridge earthquake.

Top Flange, Erectors usually make this weld first.

Restraint: External, from bolts and fit-up Thermal Upsetting and Shrinkage: Medium to high residual stress.

Beam flanges still weld directly to the column flange when using the Reduced

Bottom Flange. Welded after top flange. Restraint: External from bolts, fit-up, and completed top flange weld. Thermal Upsetting and Shrinkage: High residual stress because of added top flange restraint.

Figure 3-6

Welded Moment Frame Connections with Minimal Residual Stress, A.L. Collin and J.J. Putkey, Steel TIPS, October 1999 5

Beam Cover Plates. The Forell/EIsesser cover plate design looks similar to our suggested connection. See Figure 3-7. This design uses a bottom cover plate shop welded to the column flange and a top cover plate shop fillet welded to the beam top flange. Field crews first groove weld the top cover plate to the column flange, and then fillet weld the bottom flange cover plate to the beam bottom flange. ~,-

Top Flange. Erectors make this weld first. ~ Restraint:External. ~ _ _ _ ~__~_ ~ Thermal Upsetting and Shrinkage: Probably high residual stress, but ~ depends on external restraints, level. ~ ~ ~ ~ . ~ t ~ - ~ 3 S~DE~

Bottom Flange. Welded after top flange. Restraint: Within fillet weld. [[ [[~, ~ Thermal Upsetting and Shrinkage: Low.

Figure 3-7

Beam Cover Plates and Flanges. Another connection uses top and bottom cover plates shop fillet welded to the beam flanges. Field crews then weld these cover plates and the beam flanges to the column flange. See Figure 3-8. Web bolts to the shop welded shear plate restrain both top and bottom joints.

Top Flange. Assume erectors weld first. Restraint: External. Thermal Upsetting and Shrinkage: Medium residual stress.

Bottom Flange. Welded after top flange. Restraint: External from web bolts and completed top flange weld. Thermal Upsetting and Shrinkage: High residual stress because of added top flange restraint.

We do not recommend the combined flange and cover plate joint because of the discontinuity caused by the interface between the beam flange and the cover plate.

I • I I I I I I A

Figure 3-8

Webs of Beams. Beam webs connect to the column with the following options:

.

2. 3.

Bolts to the shear plate. Fillet welds to the shear plate. A groove weld direct to the column flange.

In option 2, low residual stress occurs. In options 1 and 3, tightened bolts or already completed flange welds externally restrain the weld joint, and high residual stress may occur.

!

Welded Moment Frame Connect ions with Minimal Residual Stress, A.L. Collin and J.J. Putkey, Steel TIPS, October 1999 6

MEMBER RESTRAINT ~ •

Figure 3-9 shows an example of member restraint when welding / /

beams between rigid braced ~ bents. As stated in Figure 3-9, / / member restraint causes high residual stress because, "Closing welds for members / / between rigid assemblies are "~,,,~ subject to high restraint." v,. x

Another similar example of member restraint, and the main focus of this Steel TIPS, occurs because of the erection method used to construct a typical tier building. This type of member restraint has been a problem

r i

L Fig. 19. Closing welds .for members between rigid assemblies are

subject to high restraint.

Figure 3-9* AISC Engineering Journal, 1973 [1]

since the advent of heavy welded moment connections--whether recognized or not.

Erect ion Method, A typical tier building usually has two floors of steel per tier, but sometimes has three floors per tier. Column splices usually occur approximately 4 ft. above every second or third floor. If a tier building had only one floor per tier, the problem would not exist. For discussion purposes, this TIPS assumes two floors of steel per tier.

Two Floor Erection. Figure 3-10 illustrates a simplified example of two floors per tier. Starting on the work floor, covered solidly for safety purposes, field crews erect the columns, then the mid floor beams, and then the top floor beams.

Top Floor

Mid Floor

Column Splice

Work Floor --~

rl

Figure 3-10

* American Institute of Steel Construction, Inc. Reprinted with permission. All rights reserved.

Welded Moment Frame Connections with Minimal Residual Stress, A.L. Collin and J.J. Putkey, Steel TIPS, October 1999 7

Weld Sequence. After crews erect and plumb-up the steel, they then weld it. Welders can weld the top floor first and then the mid floor, or they can weld the mid floor first and then the top floor. Either way, the problem of member restraint occurs. For illustration purposes we show the top floor welded first. See Figure 3-1 1 below. Columns

Step 1

Top

Mid

Work

Floor - - 1

Floor '

_Z

I leaned out

Assume beams detailed longer or root openings made larger to allow for weld shrinkage [5] p4-11

- - B e a m s detailed correct length

- - C o l u m n splice

Beams welded

Step 2

Member Restraint

Step 3

- ~k,__ Closing [ ~ welds

Beams welded, weld shrinkage moves columns in

Beams must hove correct root opening

.~]~-- Column splice Column splice

-- Beams welded

/

H Columns provide

J external restraint

~ _ Closing ? welds

_ I1

Beams welded

_ High residual ~ stress from

- member restraint

- - C o l u m n splice --Beams welded

Figure 3-11

Welded Moment Frame Connections with Minimal Residual Stress, A.L. Collin and J.J. Putkey, Steel TIPS, October 1999 8

4. THE SUGGESTED SOLUTION

This section describes and details the welded moment connection suggested by the authors.

USE JOINT DETAIL WITH MINIMAL RESTRAINT

The suggested solution avoids residual stresses caused by large groove welds not allowed to shrink because of external restraint. See Section 3, The Problem. Suggested readings on restraint and weld shrinkage include the following references:

AISC Engineering Journal, 1973. [1] AWS Structural Welding Code. [2] AISC Manual of Steel Construction, Allowable Stress Design. [3] AISC Manual of Steel Construction, Load and Resistance Factor Design. [4] AISC Manual, Volume II Connections. [5] Preece and Collin, Steel TIPS, 1991. [6]

Joint Detail. Figure 4-1, Connection Detail, on page 10, shows an elevation view and plan sections of the suggested solution.

Restraint Condit ions. Please note the beam flanges and web do not directly weld to the column flange. Instead, the beam flanges fillet weld to cover plates already welded to the column flange. This procedure results in minimal residual stress because:

Cover plates and the web shear plate groove weld to the column flange without external restraint and without member restraint with resulting low residual stress. Fillet welds from the beam flanges to the cover plates contribute very low residual stress. The connection from the beam web to the shear plate contributes low or no residual stress. On the bolted web connection, the bolts obviously contribute no residual stress. If the shear plate is fillet welded to the beam web, those fillet welds contribute very low residual stress.

Continuity plates on the column contribute medium to high residual stresses from both thermal upsetting and shrinkage due to flange restraint. However, these stresses are the same as for other welded connections.

Welded Moment Frame Connections with Minimal Residual Stress, A.L. Collin and J.J. Putkey, Steel TIPS, October 1999 9

ALTERNATE TOP COVER PLATE CONFIGURATION

L [

FOR HORIZONTAL FILLET WELD 7

1' L ~ _ ~ __ 7 ~- - - COVER PLATE, T & B

PLAN

CONTINUITY PLATE

COLUMN I

ERECTION BOLTS ~

C.JoP. REMOVE BACK-UP BAR (STEEL OR CERAMIC) AND RUN-OFF TABS SEE NOTE 1

~-- COVER PLATE ~ / / - - BEAM

~ ~,~ SIDES

SHEAR PLATE

WELD AFTER I GROOVE WELDING TOP COVER / PLATE TO / COLUMN ~

. "-,.,.......---.---- COVER PLATE

C.J.P. REMOVE BACK-UP BAR AND RUN-OFF TABS SEE NOTE 1

NOTE I. GRIND SMOOTH TO REMOVE STRESS RISERS AND LAYER OF MARTENSITE FROM BURNING OPERATION

ELEVATION

CONNECTION DETAIL

Figure 4-1 Detail of Suggested Solution

Welded Moment Frame Connections with Minimal Residual Stress, A.L. Collin and d.d. Putkey, Steel TIPS, October 1999 1 0

H I S T O R Y OF S U G G E S T E D S O L U T I O N

Our suggested solution is not new. Engineers have described it in various welding publications and used it on past projects, but not for relocating the plastic hinge. The following publications and personal experience discuss and illustrate the suggested connection.

The James F. Lincoln Arc Welding Foundation

The Book, "Design of Welded Structures" by Blodgett [7], presents designs similar to the suggested solution. Among other things, Blodgett presents the loose top cover plate design for:

Proper fit up and correct root opening. Reduced beam preparation.

See Figures 4-2 and 4-3 for the connections presented.

Figure 4-2* Figure 4-3*

Personal Experience

In 1972 Mr. Putkey served as project manager for steel erection on a Federal Office Building in Seattle, Washington. This building is 38 stories high with the interior moment frame connections designed very similarly to the suggested solution. See Figure 4-4. The shop fabricated the loose top cover plate narrower than the beam flange so the field welders could make the fillet weld in a horizontal position.

I

I

I I

Figure 4-4

*Reprinted with permission from Lincoln Electric Co.

Welded Moment Frame Connections with Minimal Residual Stress, A.L. Collin and J.J. Putkey, Steel TIPS, October 1999 1 1

A I S C Journa l

The article, "Commentary on Highly Restrained Welded Connections" in the AISC Engineering Journal [1], comments on occurrences of lamellar tearing in highly restrained joints. This article presents recommendations to minimize or avoid lamellar tearing. (Note: Engineers found lamellar tearing was not a problem in the SAC program. We refer to the Commentary to address restraint.) Recommendation No. 2 states, "Design connections to minimize accumulation and concentration of s t ra ins resul t ing from metal concentrated in localized areas."* The recommendation includes a Fig. 24 showing a moment connection detail, " . . . designed to allow dissipation of

~ m !

~ . . . . . . ,,,,1.

I /

Fig. 24. Moment connection using flange connection plates. To avoid high shrinkage strains, beam seat is shop welded to column; top plate weld to column is made infield, followed by fillet welds of top and bottom

plates to beam flanges.

Figure 4-5*

shrinkage strains from full penetration flange welds."* See Figure 4-5. This connection is very similar to our suggested solution. The detail solves the problem of connection and member restraint. See Section 3.

The article also mentions that, "Closing welds for members between rigid assemblies are subject to high restraint."* The suggested solutiqlt avoids this restraint condition. (See two floor erection procedure in Section 3).

AISC Manua l

A loose top cover plate connection is illustrated in Chapter 4, Moment Connections, AISC Manual, Volume II Connections [5]. See Figure 4-5, Part II on page 4-10, and Figure 4-7 on page 4-16. The details shown in these figures are similar to the details in the above Lincoln Arc Welding Foundation book and the AISC Journal.

S A C Repor t , , , , •

"Steel Moment Frame Connection, Advisory No. 3", by the SAC Joint Venture [8], contains loose top cover plate connection design details on pages C - 33 and D - 154. Rutherford & Chekene Consulting Engineers present a design detail on page C - 33 that is similar to our suggested solution. See Exhibit 1. Mr. Collin presents design details of our suggested solution on page D - 154. See Exhibit 2.

Please note the Rutherford & Chekene design represents specific cover plate thicknesses and sizes for specific column and beam sizes.

*American Institute of Steel Construction, Inc. Reprinted with permission. All rights reserved.

Welded Moment Frame Connections with Minimal Residual Stress, A.L. Collin and J.J. Putkey, Steel TIPS, October 1999 12

TESTING

Codes require testing of beam to column moment frame connections unless prior testing of the joints and connection meet specified requirements.

Tests By Forell/Elsesser Engineers, Inc.

In August 1995, Forell/Elsesser Engineers, Inc. published a pamphlet titled, "Steel Connection Update: Successful Test of Welded Steel Beam - Column Moment Connection." [9] This pamphlet discussed successful tests conducted on a cover plate connection similar to our suggested solution. See Figure 3-7 and Exhibit 3 for the design details of the Forell/Elsesser connection. See Exhibit 4 for the plastic rotation results of one test specimen. The Department of Energy funded the testing.

Note: The cover plate thickness and size in the Forell/Elsesser connection detail shown in Exhibit 3 are sensitive to the beam and column sizes used. Forell/Elsesser "tuned" the plate thicknesses and dimensions to those sizes. The connection detail shown in the 1995 publication [9] was for specific column and beam sizes on a specific project. Do not apply to other member sizes without closely matching the original sizes tested. Different sizes would require additional testing.

Tests By SAC

The University of California, Berkeley, Pacific Earthquake Engineering Research Center is testing various moment connections for SAC. One connection has top and bottom cover plates with both plates fillet welded to the beam in the shop and groove welded to the column in the field. The end result is similar to our suggested solution.

Welded Moment Frame Connections with Minimal Residual Stress, A.L. Collin and J.J. Putkey, Steel TIPS, October 1999 13

5. ADVANTAGES AND DISADVANTAGES

The following advantages and disadvantages result from comparing the suggested connection presented in Section 4 to other connections in current use.

ADVANTAGES

Principal Advantage

Minimal Residual Stress. Our suggested connection avoids direct beam flange to column flange welds and restrained cover plate to column flange welds. It eliminates medium or high residual stress that occurs when welding these joints to a column flange because connection restraint or member restraint is not present. Instead, low residual stress occurs in the top and bottom cover plate joints to the column flange because erectors do not restrain these plates during welding.

Other Advantages

Correct Fit-up. The loose top cover plate assures fit-up with the correct root opening. Erectors lay the top plate on the top flange of the beam, position it for correct root opening, and loosely clamp its end to prevent lifting.

No Change in Column Spacing Because of Weld Shrinkage. No groove weld shrinkage on beam to column connections allows detailing the beams for exact column spacing. Detailers need not contend with shrinkage allowance.

Plumbing-up. Plumbing-up is easier and more exact because of:

Beams detailed for positive connections and to exact column spacings. No requirement to plan and allow for weld shrinkage. No need to work around a groove welded beam joint.

Welding Sequence. With no change in column spacing because of weld shrinkage, beam to column welding can start in any part of the building. When weld shrinkage occurs, erectors usually start welding in the center of a building.

Welded Moment Frame Connections with Minimal Residual Stress, A.L. Collin and J.J. Putkey, Steel TIPS, October 1999 14

Beam Fabr icat ion. Beam ends require no preparation for field welding; they require only hole punching in the beam web for erection bolts. Further, the positive web connection eliminates the need for an access hole to place the back-up bar for the top cover plate weld.

Bol t /Weld Sequence. Other connections with welded flanges and bolted webs still have the problem of when to tighten the boltsmbefore or after the flange welds. [10] The suggested connection eliminates that problem.

DISADVANTAGES

Loose Top Cover Plates. The loose plates require more shipping pieces and more pieces to handle in the field.

Cover Plates. Cover plate design may require additional steel and additional field welding.

Welding Position. Loose top plates the same size as the bottom plates require welding fillet welds in an overhead position. Narrower and thicker top plates change the welding position to the horizontal position, but may increase the quantity of weld metal for flat groove welding.

Shipping. Shop welded bottom cover plates make stacking, handling, and shipping columns more difficult than other types of connections.

Welded Moment Frame Connections with Minimal Residual Stress, A.L. Collin and J.J. Putkey, Steel TIPS, October 1999 1 5

6. CONCLUSION Opt o s! Options! Options!

LEVEL OF RESIDUALSTRESS

Current field welding practices produce residual stress in welded moment frames. Some connections produce more residual stress than others either by design, welding practices, or welding sequence. The residual stress level is unknown, but the stress is present.

The Northridge earthquake caused many welded connection failures, with most failures related to the bottom flange joint. We conclude residual stress may have contributed to the bottom flange joint failure because of a welding sequence of top flange first and bottom flange second. This sequence sets up high residual stress in the bottom flange.

A CONCLUSION TO CONSIDER

While developing this Steel TIPS, we kept arriving at the same conclusion, "The structural steel industry should change its current practice of making welded moment frame connections from a connection that produces high residual stress to a connection that produces minimal residual stress."

Welded Moment Frame Connections with Minimal Residual Stress, A.L. Collin and J.J. Putkey, Steel TIPS, October 1999 16

7. REFERENCES

.

.

.

.

.

.

.

.

.

10.

"Commentary on Highly Restrained Welded Connections," AISC Engineering Journal, AISC, Chicago, Third Quarter, 1973, p. 61-73.

Structural Welding Code-Steel D1.1-98, AWS, Miami, 1998, Annex XI, p. 266 and 271; C426.

Manual of Steel Construction: Allowable Stress Design, 9 th ed, AISC, Chicago, 1989, p. 1-6, 4-152.

Manual of Steel Construction: Load and Resistance Factor Design for Structural Steel Buildings, 2nd Edition, AISC, Chicago, 1994, p.1-5,1-6, 5-177.

Manual of Steel Construction: Volume II Connections, ASD/LRFD, First ed., AISC, Chicago, 1992, p.2-19 to 22, p. 4-10 and 11, p. 4-16.

Preece, F. Robert and Collin, Alvaro L., "Structural Steel Construction in the '90s," Steel TIPS, Structural Steel Education Council, Walnut Creek, September 1991, p. 14.

Blodgett, Omer W., "Design of Welded Structures," The James F. Lincoln Arc Welding Foundation, Cleveland, June 1966, p.5.1-8; p.5.7-1 and 2.

"Steel Moment Frame Advisory No. 3," SAC Joint Venture, Sacramento, 1995, p. C-33 to 36, p, D-154.

"Steel Construction Update: Successful Test of Welded Steel Beam - Column Moment Connection," Forell/Elsesser Engineers, Inc., Structural Engineers, San Francisco, August 1995.

Putkey, James J., "Common Steel Erection Problems and Suggested Solutions", Steel TIPS, Structural Steel Educational Council, Moraga, December 1993, p. 33.

Welded Moment Frame Connections with Minimal Residual Stress, A.L. Collin and J.J. Putkey, Steel TIPS, October 1999 17

RI.ITH~RFORD & GH~.KE,N~ I O~II~EUI~TIN~ = ~ , ~ , ~ = i I ~ ~

m ~ + ~ ~ ~ ~ ~ " - ~ ~ ~ " ~ ~ ~ ' ~ ~ ~ ~ , ~ o ~ m ~ ~ + ~ ~ . .. , m~ ~ m : ( ~ 4 + ~ z t ~=:(m~S+~SS+ ~ ~ ~++ ~+ ~ _ ~ ~

- ~ ~. ++ +f~ - + .. ~ - + • . " I t . ~ + ~ + m + t . " / o ~ ~ I ~ ~ - - I~ . . . . . . ~ . ~ ' ~ ~ I i ~ . u , ~ ~ ~4

" r ~ ; ' - - " i ~ ~ ' ~ ~ ~ ' ~ " ~, ~ ~ ] ~ ~ + ~ ~ 9 ~ ~ ~ m b m ~ ~ .

+ ~ ' ~ ~ . ~ ~X~ ~ = . . . . ~ k ~ ~ + - - + . , . + + ~ ~wt+

~ + m + ~ + , . + ~ ~ ~ , p '+

$ ~ ~ ~ I I

t I ~ ~ ' > ' ~ ' i i ~ ~ P

+ .i-

~-t,ml=

,i~i~>:,-- ,F-+-i l I I I ----+}...'i e ' - - "--+

'~ , ~-l.-----~-m t.~ +' .; ~ ~'~ ~:~-'

"7 - ~'~ &~+~ ~

~I~'4 ++++++ I~I ~- ~ ~4 ~¢~.- -r~,

m.+-,,l,,

' - ' rY P. 5 , , ~ , , \ II \

" , 1 " ~ . ~ ' ~ \ \ ~/ /~,.,__~ ~ , ~ i..I~'P¢,,.,~ ~ $ m~ " ~ ' ~ " - - ' ~ ' ~ - ~ c ~ ..... .~'t ,+" ~; \ ~" " "'~" ~ ~

, , t . . ~ ~ ~ ~ . - _ _

. . . . . I..~'~ ~ ~ + ~ ~ I ~ ~ P ~ I

- _ ~ ~ ' ~ ~ . ~ ~

C-33

EXHIBIT 1

Welded Moment Frame Connections with Minimal Residual Stress, A.L. Collin and J.J. Putkey, Steel TIPS, October 1999 18

Ref 5.4k

. _ ~ A C J O H . d ' 7 - V , ~ ' t G T~i~x~ - W o ~ ~ o ~ ~ T I M E ~ o . ~ ~ ' " ~ o ~ M ~ ~I~ T/~ ~ ~ f ~ ~ ~ ~ ~o~M~T/o~

..~_c_~,__,~_~ ". B " - ,~_~.M M <a~_c.z~-.p__~_~_~.. -_C#_<~,~.~z,e.zr,~'~__z:~__,5, = D/2 ,~,,'I~.

l- T o p C o ~, c r -~ ~ /o ~

L.~ F/a ld ,~..,l a ld '~'/~ _ _ ~o~.~,~,~ ! - -~ ~ M . Llo r ' ~ 5

/ - do ~o~a,6o/c~ oor ~Tel~i z i o ~ x - WouF£~9. .

~ ~ ~ I ~ o ~ t ~ i ~ ~ l " A " ~ ' , ~ - -,P-I : " ! ~ r ~ / / ~ , -/c> , ~ / o , ~ n .

~ , ~ . = / ~ ~ $ ~ ~,- ¢ ~ # t ~ 1 ~ x , ~ ~ I -~ . - - V ~ ~ P/o/~ I ~ t e e = / ~ ~. A r ~ o F

I ~ ~ ~km ~ d d ~ ~ / . I ~ ~ ~ s ~m. . - ~ { ~ ~ , ~ - ~ ~ # ~ ~ ~,i ~.- ~ ~ ~ ~

" ' I. ' . . " ~ I ~ l B ~ ~ n ~ o ~ o y ~ " _I I . ~ • . 1. . ~ t . . ~ . . o ~ z . . . . . . '

. . . . . . . . . . . . . ~X~.~.~, ; ~ l ~ d

. ~ ~ ~ V T ~ . n 7 g ? ~ T -

,,.,,, ~_-._ ;; . . . . . . ~ Z . . ~ ~. ~ ~ i d y

_ ~ ..... ~ k ~ . ~ l ~ . . ~ ~ L ( x T ~ J ~ - ~ . ~ : ~ . - C a i i = { ~ ...... , ~ ~ I _ ~ ~ 1 ~ #~. ~ , - ( ~ J - ....... ~ . ~ _ . ~ _ . ~ l ~ _ _

.. ~ k ~ . ~ ~ ~ _ ~ z

" " ~ . . . . . . . . ~ ~ ~ ~ i o ~ . . . . . . . . ~ ~ v ~ . ~ ~ d J . . . . . . . . ~ ~ . ~ ~ .....

~ ~ E ~ / Z ~ ~ / ~ ~ ~ ~ ~,~.. ~ ~ . ~ , . ~ . . . I l-i,, . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~-../ a , - ~ , . ~ / d ~ ~ , ~ . ~ l z l . .. I ~ - ~ , < . : < .~~-..: ~ ... -..- ........ ./ ~ .. ~ - k D ~ w ~ d , ~ ~ / I i " ~ ~ ~ . . . . l . i . ~ . ~ , ~ . . . . . # ~ u , ~ ~ ~ d ~

- ~ ~1~ ~ - - . ~ - - ~ 1 = ~ ~ l l ~ t m ~ s J . A . ~ J - ~L £ .. ~ ~ i ~ . ~ . __-.

. " - I ~ . ~ " 7 . . . . . . : . ' [ : , v ~ " ~ r ~ X x v ~ i ¢ k~ l i d ~ .

. . . . . . . . ~ ~p ~:~ ~Io#~ . ~I~ ~. cJP ~ ~ . ~. ~o ~I_~_~, ~. (~ ". bo=~,n~= ~#~I ~

' ~ " - - " ' " * ~ - - - " - ~ Z ~ l ~ . o r . c ~ c • ~ ~ . ~ - 7 ~ P C ~ K ~ ~ T ~ D ~ : ~ L . , ~ $ ~ ,

.: . . . . . . . . . " . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . /~. : D ~ t ~ / / ~ ~ ~ - ~ . . . ~ . c . 4 ~ ~ o o ¢ ~ ~ / ~

~ i - ; . , . , , ~ ; , . , ; . . ; . ~ : ' . " " " . . . . . . . . . " " . . . . . • - • • . : - , - - : ~ . . . . . . . . . :'1 . . . . . : . . . . . . . . . . . . . . : " . . . . "

I~. H !....... ,,,.,+.......,. L, . . . . . . . . . . . .

. . . . . " . . t < . . . . . . . . . . ~ . ~ . . . . . . . . . . . .

I < . . . . . ,I-.- . . . . . . . . . , . . . . . . . . . . . . . . . . . . . . . . . - • ~ _ - ~ . , , - . L . . . : , ~ - ~ ..... g ~ . . . . . . . . . . . . . ~ ~ . . . . . . . . . . . . . . . . . . . ~ . . . . . . . . • . . . . . . . . . . . .

. . . . . . . . . . . . . . . : . ~ . . ~ e ~ . ~ / ~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~ _ .

.. ~ o ~ . . ~ d ~ ~ l o ~ n : ( ~ ] • . . . . . . . . . ~ l d ~ . ~ . ~ ~ ( ~ 9 . . . . . . . . . . . . . . . . . .

. . . . . . . . ~ ~ : ' ~ : . - ~ o ~ _ . ~ . ~ ~ - ~ B ~ ! ~ . . . . . . . . . . . . . . . . . .

D-~54

E X H I B I T 2

Welded Moment Frame Connect ions with Minimal Residual Stress, A.L. Collin and J.J. Putkey, Steel TIPS, October 1999 1 9

LBL H U M A N G E N O M E L A B S T E E L J O I N T T E S T

~.~.~ ~ ~

NO WELD ~ AT RLLET

W14x211 ~ (A572 GR. 50)

FLAME CUT COVER PLATE / / AND GRIND SMOOTH

24" /

• . • 1" NO WELD TYP. EA. END

" t r ROLLING - ~ I ~ • GRAIN " "~ ~ ~ ~ ± . . . . . . . . . . : = . . . . . . = = ~ =

~ 1 - - - - ~ = = ~ - ~ NO WELD . . . . . . . I "t~

. _ _ . . ~ - ~ PLAN \ ~'~" ~' T~. T • B

t ~ ~ C . J . P . REMOVE BACK-UP BAR

AFTER ROOT IS CLEANED AND INSPECTED, TYP. T & B

I -- I

v

W.P.

2 ERECTION BOLTS - ~

. , ~

SIDES

J . P .

~/~ t: 22"~1~"

~" W30x99 (A36)

~ I . ~ ~.o~ RELD

-.4"x3/8" BAR AT ~ OF COL. (REMOVE BAR AFTER STEEL ERECTION IS COMPLETED)

• SHOP WELD C. J. P. REMOVE BACK-UP BAR

NOTE: ALL WELDS SHALL BE "SR" WELDS - SEE SPECIRCATIONS.

EXHIB IT 3

F i g u r e 6 J O I N T D E T A I L

Forell /Elsesser Engineers, Inc. Morch 24, 1995

Welded Moment Frame Connections with Minimal Residual Stress, A.L. Collin and J.J. Putkey, Steel TIPS, October 1999 20

The Test

A beam and column sub-structure assembly was fabricated into a "T" configuration, shown in the photo below, to mimic the actual configuration found in the building. Laid fiat on the testing floor, the column was restrained and a 150,000 pound hydraulic actuator drove the free end of the beam back and forth over a plus and minus 5 inch range.

Both specimens survived this test regime with the expected yielding occurring in the beam and virtually no signs of distress or fatigue visible or detected in the connection region. Specimen 2 was tested with a pre-existing crack in the column flange, and was successfully cycled I0 times over the 5 inch range, indicating that the connection also provides good protection of the column as well as the welded joint region.

The detailed study and test program, completed at a cost of $60,000, is reported in a paper entitled "Lawrence Berkeley National Laboratory Steel Joint Test Technical Brief' available from Forell/Elsesser Engineers upon request.

150

100

50

-50

LBNL STEEL JOINT TEST SPECIMEN NO. 2

I -

~ - - ~

- - i ~ 1

-150 ~ ,

-0.04 -0.02 0 0.02 0.04

PLASTIC ROTATION (RADIAN)

EXHIBIT 4

Welded Moment Frame Connections with Minimal Residual Stress, A.L. Collin and J.J. Putkey, Steel TIPS, October 1999 21

About the Authors

Alvaro L. Collin was a Consulting Engineer with California registration in Civil Engineering and Metallurgical Engineering. He received a BS degree from the University of California, Berkeley, in 1941 as a Civil Engineering major and a Mechanical Engineering minor. He spent 24 years with Kaiser Steel Corporation as Manager of Engineering of the Fabrication Division, Southern California, and as Senior Development Engineer, Steel manufacturing Division, Oakland, CA. For the past 18 years he consulted on welded construction, heavy equipment design and material handling systems.

Mr. Collin is a life member of the Structural Engineers Association of Northern California. He was a member of the Board of Directors and the Steel and Seismic committees of SEAONC. He was a long-time member of the American Welding Society, having served on the National Board of Directors, on the National Qualification and Certification Committee, and as chairman to the Los Angeles and San Francisco sections. AI was awarded the National, District and Section Meritorious Awards of AWS. He served on AISC and AISI Code Committee Task Groups, the SAC Joint Venture Task Group, and was a member of the Earthquake Engineering Research Institute.

Mr. Collin died during the development of this Steel TIPS. He was able to review the general outline, detailed outline, the first draft, and all concepts presented in the text.

James J. Putkey is a consulting civil engineer in Moraga, California. He received a BCE degree from the University of Santa Clara in 1954. After two years in the U.S. Army, 19 years with the Erection Department of Bethlehem Steel Corporation--Pacific Coast Division, and seven years with the University of California--Office of the President, he started his own consulting business. He has provided consulting services to owners, contractors, attorneys, and steel erectors for the past 18 years.

Mr. Putkey is now "Semi-Retired." However, he still serves as a hearing officer for the University of California-Office of the President, and occasionally writes construction related articles.

Welded Moment Frame Connections with Minimal Residual Stress, A.L. Collin and J.J. Putkey, Steel TIPS, October 1999 22

STRUCTURALSTEELEDUCATIONAL COUNCIL

TECHNICAL INFORMATION & PRODUCT SERVICE

AUGUST 1999

Design of Reduced Beam Section (RBS) Moment Frame Connections

by

Kevin S. Moore, James O. Malley, Michael D. Engelhardt

DESIGN OF REDUCED BEAM SECTION (RBS) MOMENT FRAME CONNECTIONS

ABOUT THE AUTHORS

KEVIN S. MOORE is a Design Engineer with Degenkolb Engineers in San Francisco, Califor- nia. He ea rned his M.S. degree at The University of Texas at Aust in working u n d e r the direc- tion of Dr. J. A. Yura and Dr. M. D. Engelhardt . While conduc t ing research , Kevin ass is ted Dr. Engelhard t with mater ia l test ing for the '~UT Tests," some of the first m o m e n t connect ion tests following the 1994 Northridge ea r thquake . He was the lead engineer for a 5-stolry SMF building utilizing RBS connec t ions cons t ruc ted in San Francisco and is a regis tered Profes- sional Engineer in California.

J A M E S 0. MALLEY is a Senior Principal at Degenkolb Engineers in San Francisco, Califor- nia. He is the Project Director for Topical Invest igations of the SAC Jo in t Venture Par tnership . The SAC Jo in t Venture was created to develop guideline d o c u m e n t s for the design, evaluation, and repair of steel m o m e n t frame bui ldings in response to the damage caused by the North- ridge ear thquake . J i m has been involved with m a n y steel design and peer review projects, inc luding the 5-story SMF bui lding listed above. He is a m e m b e r of the AISC Commit tee on Specifications and Chair of the Seismic Subcommi t t ee and has a u t h o r e d n u m e r o u s papers on steel design and cons t ruc t ion t h roughou t his career. He is also a regis tered S t ruc tu ra l Engi- neer in California.

MICHAEL D. ENGELHARDT is an associate professor of Civil Engineer ing at The University of Texas at Austin. Mike teaches courses on s t ruc tu ra l steel design at The University of Texas and conduc t s r esea rch on seismic res i s tan t steel framing. His previous work inc ludes major cont r ibut ions to the development and val idat ion of eccentr ical ly braced f rames (EBFs). Mike h a s been an active par t ic ipant in m o m e n t connec t ion resea rch since the 1994 Northridge ea r thquake and has worked extensively on RBS rela ted research . Mike is a m e m b e r of AISC Task Commit tee Number 113 on Seismic Design and is a registered Professional Engineer in California.

DESIGN OF REDUCED BEAM SECTION (RBS) MOMENT FRAME CONNECTIONS

C O N T E N T S

I .

.

3.

4.

o

6.

.

I N T R O D U C T I O N . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 D E S C R I P T I O N O F S M F . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1 .2 B A C K G R O U N D O F R B S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

H I S T O R Y O F T H E D E V E L O P M E N T O F R B S S M F C O N N E C T I O N S . . . . . . . . . . . . . . . . . . . . . . . . 3

2.1 INITIAL R E S E A R C H . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

S U M M A R Y O F T E S T R E S U L T S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 3.1 O V E R V I E W O F T E S T R E S U L T S F O R R A D I U S C U T R B S S P E C I M E N S . . . . . . . . . . . 4

R B S D E S I G N P R O C E D U R E F O R S M F S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

4 .1 R B S D E S I G N . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

4 . 2 R B S S I Z I N G . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

4 . 3 S T E P - B Y - S T E P P R O C E D U R E . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

4 . 4 A D D I T I O N A L D E S I G N C O N S I D E R A T I O N S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 R B S D E S I G N E X A M P L E . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 P R O C E D U R E S F O R A C C E P T A N C E O F D E S I G N B Y B U I L D I N G A U T H O R I T I E S . . . 2 1 6. I C O M M U N I C A T I O N . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

6 . 2 M E T H O D O L O G Y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2 6 . 3 C O N S T R U C T I O N D O C U M E N T S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2

F A B R I C A T I O N A N D I N S P E C T I O N I S S U E S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2 7 .1 C U T T I N G A N D G R I N D I N G . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2

7 . 2 W E L D I N G . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 3

R E F E R E N C E S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 5

A P P E N D I X A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A i

LIST O F F I G U R E S

1.1 1.2

2 .1

2 . 2

4 .1

4 . 2 4 . 3

4 . 4

4 . 5

4 . 6

4 . 7

4 . 8 5 .1

5 .2 5 . 3

P R E - N O R T H R I D G E M O M E N T C O N N E C T I O N D E T A I L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

R A D I U S C U T R B S M O M E N T C O N N E C T I O N . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

T A P E R E D C U T R B S M O M E N T C O N N E C T I O N . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

E X A M P L E O F L A B O R A T O R Y B E H A V I O R O F R A D I U S C U T R B S T E S T S P E C I M E N ..... 4

(A) D E T A I L O F T E S T S P E C I M E N . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

(B) R E S P O N S E O F T E S T S P E C I M E N . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 M O M E N T D I A G R A M A N D B E A M G E O M E T R Y F O R R B S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

G E O M E T R Y O F R A D I U S C U T R B S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 T Y P I C A L M O M E N T F R A M E B E A M W I T H R B S C O N N E C T I O N S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

B E A M AT M I N I M U M S E C T I O N O F R B S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

F R E E B O D Y D I A G R A M B E T W E E N C E N T E R S O F R B S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

F R E E B O D Y D I A G R A M B E T W E E N C E N T E R O F R B S

A N D F A C E O F C O L U M N F L A N G E . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

F R E E B O D Y D I A G R A M F O R C A L C U L A T I O N O F C O L U M N M O M E N T S . . . . . . . . . . . . . . . . . . . . . . 13

C O M P A R I S O N O F T E S T R E S U L T S F O R C O V E R P L A T E D A N D R B S C O N N E C T I O N S 17

R B S D I M E N S I O N S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 P O R T I O N O F E X A M P L E B E A M B E T W E E N R B S C E N T E R S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

C O N N E C T I O N D E T A I L F O R D E S I G N E X A M P L E . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

DESIGN OF REDUCED BEAM SECTION (RBS) MOMENT FRAME CONNECTIONS

I. I n t r o d u c t i o n

When subjected to a major ear thquake, build- ings designed to meet the design require- ments of typical building codes, such as the UniI 'orm B u i l d i n g ~ C o d e (1997), are expected to have damage to both s t ruc tura l and non- s t ruc tura l elements. The s t ructura l design for large seismic events m u s t therefore explicitly consider the effects of response beyond the elastic range. The "Special Moment Frame" (SMF) steel building system is designed such tha t the connec t ions be tween the f rame b e a m s a n d c o l u m n s absorb s u b s t a n t i a l energy and provide major contr ibut ions to the d isplacement ductil i ty demand.

1.1 D e s c r i p t i o n o f SMF

Recent s tudies by Lee (1997) a n d others have demons t r a t ed tha t this a s s u m p t i o n is far dif- ferent from the ac tual behavior.

l ~ ~ - - ~ C.P. ~70T-4

I I : I . 7/8" A325-X BOLTS 1

A SMF lateral force resist ing system is often preferred by bui lding owners and archi tec ts because this type of sys tem provides large u n o b s t r u c t e d spaces t h roughou t the build- ing plan. This "open" layout offers the mos t flexibility for p rogramming the spaces as well as a rch i tec tura l appointments . For these rea- sons, steel buildings with SMF sys tems are quite c o m m o n in major commercia l and inst i tut ional s t ruc tures . Fur thermore , the SMF system is considered by m a n y to be one of the most ducti le steel building sys tems available to the engineer. For this reason, SMF sys tems have been widely used in a reas of high seismicity.

SMFs are typically comprised of connec- t ions be tween wide flange b e a m s a n d columns where beam flanges are welded to co lumn flanges utilizing complete joint pene- tration welds. Figure 1.1 shows a typical unreinforced design detail for a beam-to-col- u m n connect ion used in SMF sys tems prior to the 1994 Northridge ear thquake . Common practice prior to the Northridge ea r thquake was to ei ther bolt or weld the web to the col- u m n shear plate, and to weld the beam flanges to the co lumn flange us ing a com- plete joint penet ra t ion groove weld. Histori- cally, designers have a s s u m e d tha t beam shear is t ransferred to the co lumn by the beam web connect ion and the m o m e n t is t r ans fe r r ed t h r o u g h the b e a m flanges.

Figure 1. I Pre-Northridge M o m e n t C o n n e c t i o n Detai l

In the design of SMF connect ions , the engineer m u s t set objectives for both load and deformat ion capacit ies. Usually, the load capaci ty r equ i rement is based on the plastic m o m e n t of the beam. The connec t ion m u s t be s t rong enough to develop the s t rength of the beam, thus reduc ing the r isk of brittle failure in the connect ion. Inelast ic deforma- tion capaci ty is required to a s su re ducti l i ty in p rede te rmined locations when subjec ted to large deformat ion demands .

After some of the problems observed in SMF connec t ions after the Northridge ear th- quake, a c o m m o n phi losophy has been to design the connec t ion to r ema in nominal ly elastic at the co lumn face, and force the inelastic deformat ion of the frame to occur in a portion of the beam, away from the con- nection. This phi losophy is executed by us ing a "capacity design" approach. The plastic m o m e n t and associa ted shea r of the beam is based on probable s t rengths of mater ials . These m a x i m u m s then become t h e design loads for the connect ion. The connec t ion of the beam to the co lumn flange is t hen des igned us ing nomina l mater ia l propert ies.

Most post-Northridge connec t ion designs locate the plast ic h inge (where inelast ic

DESIGN OF REDUCED BEAM SECTION (RBS} MOMENT FRAME CONNECTIONS

de fo rma t ions a re c o n c e n t r a t e d in the SMF beam) a w a y f rom the c o l u m n flange t h r o u g h re inforc ing a s h o r t por t ion of the b e a m n e a r the co lum n . By i n c r e a s i n g the s t r e n g t h of the b e a m in th is region, a p las t ic h inge will t end to form j u s t a d j a c e n t to the re inforced por- t ion of the beam. The i n h e r e n t difficulty wi th ut i l izing a re in forced b e a m - c o l u m n connec - t ion is the i n c r e a s e d ma te r i a l a n d labor cos ts a s soc i a t ed wi th th i s c o n n e c t i o n a n d the SMF sys t em as well as r equ i r ing welds t ha t are difficult a n d cost ly to m a k e a n d inspect .

1.2 B a c k g r o u n d o f RBS

Anothe r type of c o n n e c t i o n deve loped to force the ine las t ic d e f o r m a t i o n a w a y f rom the b e a m - c o l u m n in te r face is re fer red to as a "Reduced B e a m Sect ion" c o n n e c t i o n (RBS) or "dogbone". This c o n n e c t i o n relies on the selective remova l of b e a m flange ma te r i a l a d j a c e n t to the b e a m - t o - c o l u m n connec t ion , typical ly f rom bo th top a n d bo t tom flanges, to r e d u c e the c ross sec t iona l a r e a of the beam. This r e d u c t i o n in c r o s s sec t iona l a r e a will r e d u c e the m o m e n t capac i ty at a d i sc re te locat ion in the beam. Var ious s h a p e s of c u t o u t s a re poss ible , i nc lud ing c o n s t a n t cut , t ape red cut , r a d i u s cu t a n d o thers . Figure 1.2 i l lus t ra tes a r a d i u s c u t RBS connec t ion .

The L u x e m b o u r g - b a s e d steel m a n u f a c - t u r i ng c o m p a n y , ARBED, he ld a 1992 US p a t e n t on the r e d u c e d b e a m sec t ion (RBS).

' A

~L ~ - - . .

F i g u r e 1.2 R a d i u s C u t R B S M o m e n t C o n n e c t i o n

Fol lowing the Nor thr idge e a r t h q u a k e , t h e y wa ived all p a t e n t a n d c la im r igh ts a s s o c i a t e d wi th t he RBS for the benef i t of the profess ion . This g rac ious ges tu re a l lowed f u r t h e r devel- o p m e n t of the concep t for u s e in pos t -Nor th - r idge SMF bui ld ings .

The shape , size a n d locat ion of t he RBS all have a n effect on the c o n n e c t i o n d e m a n d s a n d pe r fo rmance . Var ious s h a p e s h a v e b e e n t e s t ed a n d u s e d in n e w c o n s t r u c t i o n d u r i n g the p a s t severa l years . Test p r o g r a m s have b e e n p e r f o r m e d to invest igate s t r a igh t c u t (Plumier , 1997), t ape r cu t (Chen, et .al . 1996) a n d r a d i u s cu t (Enge lhard t 1997; T remblay , et .al . 1997; Popov, et.al . 1998) r e d u c e d b e a m sec t ions .

The RBS forces yie lding a n d h inge fo rma- t ion to o c c u r wi th in the r e d u c e d sec t ion of the b e a m a n d l imits the m o m e n t t h a t c a n be deve loped at the face of the c o l u m n . By r e d u c i n g d e m a n d s on the b e a m flange groove w e l d s a n d the s u r r o u n d i n g b a s e m e t a l reg ions , the RBS r e d u c e s the poss ib i l i ty of f r a c t u r e s o c c u r r i n g in this v u l n e r a b l e region. A l t h o u g h the RBS essen t ia l ly w e a k e n s the b e a m , i ts i m p a c t on the overa l l l a t e r a l s t r e n g t h a n d st iffness of a s teel m o m e n t f r a m e is genera l ly qui te small .

The ine las t ic de fo rma t ion focused in a n RBS c o n n e c t i o n r e m a i n s in t he r e d u c e d b e a m sec t ion , w h i c h c a n be d e s i g n e d a n d loca t ed s u c h t ha t m i n i m a l pro tec t ive m e a s - u r e s n e e d to be t a k e n at the c o n n e c t i o n of b e a m to c o l u m n . The smal le r m o m e n t gener - a t ed a t the face of the c o l u m n for a n RBS c o n n e c t i o n , in add i t ion to r e d u c i n g s t r e s s levels on the welds , also offers s o m e a d v a n - t ages in sa t is fying s t rong c o l u m n - w e a k b e a m r e q u i r e m e n t s a n d in m i n i m i z i n g c o l u m n d o u b l e r p la te r e q u i r e m e n t s .

Fab r i ca t i on a n d e rec t ion of the RBS con- n e c t i o n avoids the add i t ion of s t r e n g t h e n i n g p l a t e s a n d spec i a l w e l d m e n t s t h a t a r e r e q u i r e d of m a n y pos t -Nor th r idge m o m e n t c o n n e c t i o n s . Consequen t ly , the RBS c o n n e c - t ion is very compet i t ive f rom a cos t p e r s p e c - tive. B e c a u s e of the compet i t ive cos t a n d e s t a b l i s h e d p e r f o r m a n c e b a s e d on ex tens ive t e s t ing a n d ana lys i s , the RBS c o n n e c t i o n a p p e a r s to be a cos t effective, c o n s i s t e n t l y pe r fo rming c o n n e c t i o n for u s e in t he se i smic des ign of SMF bu i ld ing s t r u c t u r e s .

2

DESIGN OFF REDUCED BEAM SECTION (RBS) MOMENT FRAME CONNECTIONS

. H i s t o r y o f t h e D e v e l o p m e n t o f R B S S M F C o n n e c t i o n s

A n u m b e r of significant events led to the cur- rent envi ronment su r round ing SMF design and cons t ruc t ion methodologies. Concerns over mater ia l properties, connect ion geome- try, design pa ramete r s and weld quali ty are j u s t a few issues which became a concern after brittle failures were observed in SMF m o m e n t connect ions after the 1994 North- ridge ear thquake .

SMF s t ruc tu res were still being des igned and reques ted by owners for all the reasons descr ibed earlier. The pre-Northridge con- nect ion detail had become a driving eco- nomic factor for the viability of the SMF sys- tem. To redesign m o m e n t connect ions in a SMF system utilizing expensive connect ion re inforcement t echn iques made this building system less competitive.

2 .1 I n i t i a l R e s e a r c h

A significant a m o u n t of research and t e s t i n g on RBS m o m e n t connect ions has a l ready been completed, a n d addi t ional work is underway. Appendix A provides a listing of tests on RBS connect ions . The list inc ludes key features of each test, inc luding m e m b e r sizes and s t rengths , connect ion details, RBS size and shape, and the plastic rotat ion achieved by each test assemblage. As indi- cated by the data- in Appendix A, successful tests have been conduc ted on cons tan t cut, tapered cut and rad ius cut RBS specimens.

The tapered cut, shown in Figure 2.1, is in tended to allow the section m o d u l u s of the beam to ma tch the seismic m o m e n t gradient in the reduced region, thereby promot ing more uniform yielding within the r educed section. This is in tended to create a reliable, uni form hinging location. However, s t ress concent ra t ions at the re -en t ran t corners of the flange cut may lead to f racture at these locations. After significant p las t i c rotation, both the cons tan t cut and tapered cut RBS connec t ions , have exper i enced f r ac tu re s within the RBS in some laboratory tests. These fractures have occurred at changes in section within the RBS, for example at the

m i n i m u m section of the tapered RBS. These changes of cross-sect ion in t roduce s t ress concen t ra t ions tha t can lead to f rac ture within the highly s t ressed r educed section of the beam.

I ~ ~ = ~

Figure 2. I T a p e r e d C u t RBS M o m e n t C o n n e c t i o n

The rad ius cut RBS appears to minimize s tress concent ra t ions , thereby reduc ing the chances of a f racture occurr ing within the reduced section (Engelhardt , et.al. 1996). F u r t h e r m o r e , tes t r e su l t s i nd i ca t e t h a t inelast ic deformat ions dis t r ibute over tl~e length of the r educed section. The rad ius cu t is also relatively simple to fabricate.

Figure 2.2 shows an example of a labora- tory test of a r ad ius cut RBS specimen. The connect ion detail is shown in Figure 2.2(a) and the m o m e n t ve r sus plast ic ro ta t ion response is shown in Figure 2.2(b). As is typ- ical of mos t r ad ius cu t RBS tests, this speci- m e n showed excellent performance.

As shown in Figure 2.2(a),.it is impor tan t to note tha t mos t RBS test spec imens , in addi t ion to incorpora t ing the RBS, also incor- pora ted significant improvements in welding and in other detai l ing features as compared to the pre-Northr idge connect ion. All speci- m e n s were cons t ruc t ed us ing welding elec- t rodes tha t exhibit improved no tch tough- ness as compared to the E70T-4 electrode c o m m o n l y u s e d pr ior to the Nor thr idge ear thquake .

The major i ty of spec imens also incorpo- ra ted improved pract ices with respec t to backing bars a n d weld tabs. In mos t cases, bot tom flange back ing bars were removed, backgouged a n d sealed with a fillet weld, and top flange back ing bars were seal welded to the column. Weld run-off tabs were removed in mos t cases. In addi t ion to welding re la ted improvements , mos t spec imens also incorpo-

3

DESIGN OF REDUCED BEAM SECTION (RBS} MOMENT FRAME CONNECTIONS

~ ~ / B.U, bar to remain ---.--~"~J / ~ Remove weld tabs

• ~ "~'>~ . . . . . . . i~'i"" Note: i ~ ~ 45 ° ~ All field welds: E71T-8

r ~ ")(S~if~ed CVN = 20.-~

~ ~ \-~W~,lg4 ~ ~ltS: 1" A3 ~ 25 9" C-C ~

' ~Holes: 1-1/16" DIA. J • E ~8" x 6" x 2'-6" / ~ Z ................. ~

. . . .

~ ~ I k cleaned and ins~cted

~ R e a v e B.U. bar IN k Remove ~ l d tabs ~8 ~o o

" 3'-4" Radius

~ ~ /G r i nd Smi th 5 /1~ ~ / ~ / Grind Parallel to Beam Flange

/

~ ~ ~ 2.31"

~ ,~ ~ 9" 27"

(a) Detai l of ~ e s t S p e c i m e n

d CVN = 20 ft-lbs at -20 deg F)

40000 . '

$1:~¢. ~B4 I

i .,0000

i -20000

• .30000 I ~0000

.0.0~ .0.114

~ Moment ~nd RotafJon Computed v, lth Rs~pe¢~ to Faca o~ Col,~nn

I I I ,-0.03 -0.02 .0.01 0 0.01 0.02 0.03 0.04 0.05

Total Plastic Rotation (radian)

(b) Response of Test S p e c i m e n

Figure 2.2 Example of Laboratory Behavior of Radius Cut RBS Test S p e c i m e n

r a t e d a d d i t i o n a l d e t a i l i n g i m p r o v e m e n t s . C o n s e q u e n t l y , a l t h o u g h t h e b e a m f l ange c u t o u t s a re t h e m o s t d i s t i n g u i s h i n g f e a t u r e of t h e RBS c o n n e c t i o n , t h e s u c c e s s of t h i s c o n n e c t i o n in l a b o r a t o r y t e s t s is a l so l ikely r e l a t e d to t h e m a n y o t h e r w e l d i n g a n d deta i l - ing i m p r o v e m e n t s i m p l e m e n t e d in t h e t e s t s p e c i m e n s , i.e. t h e u s e of we ld m e t a l w i t h i m p r o v e d n o t c h t o u g h n e s s , i m p r o v e d p rac - t i ces w i th r e s p e c t to b a c k i n g b a r s a n d we ld t ab s , u s e of c o n t i n u i t y p l a t e s , etc.

3. S u m m a r y o f T e s t R e s u l t s

The t ab l e in A p p e n d i x A p r o v i d e s a l i s t ing of RBS t e s t da ta . Whi le t h i s l ist m a y n o t be e x h a u s t i v e or c o n t a i n every t e s t p e r f o r m e d o n RBS b e a m - c o l u m n s u b a s s e m b l i e s or anc i l l a ry t e s t i n g to s u p p o r t p e r f o r m a n c e , t h e l is t d o e s p rov ide t h e r e a d e r w i t h a s u b s t a n - t ial a m o u n t of d o c u m e n t e d p e r f o r m a n c e con- d i t i on s for t h i s c o n n e c t i o n . The t ab l e a l so i n c l u d e s RBS t e s t s c o m p l e t e d u n d e r t h e SAC P h a s e 2 r e s e a r c h p r o g r a m as of m i d - 1 9 9 9 . T h e s e t e s t r e s u l t s h a v e n o t b e e n fo rmal ly p u b l i s h e d , b u t a re i n c l u d e d b a s e d on avail- ab le t e s t r epo r t s .

T h e AISC Seismic Provisions for Structural Steel Buildings (1997) r e q u i r e qua l i f i ca t ion t e s t i n g for SMF c o n n e c t i o n d e s i g n s . The t e s t r e s u l t s r e p o r t e d in A p p e n d i x A m a y be u s e f u l

in sa t i s fy ing th~se qua l i f i c a t i on t e s t r equ i r e - m e n t s . A p p e n d i x S of t h e Seismic Provisions for Structural Steel Buildings p r o v i d e s gu ide - l ines o n e x t r a p o l a t i n g t e s t r e s u l t s b e y o n d t h e t e s t e d m e m b e r sizes.

A p p e n d i x A i n c l u d e s l i s t ings for 43 RBS tes t s . Th i s n u m b e r does n o t i n c l u d e t e s t s by P l u m i e r (1997), or s h a k i n g t ab le t e s t s by C h e n , Yeh a n d C h u (1996). Add i t i ona l t e s t s h a v e a lso b e e n c o n d u c t e d o n s p e c i m e n s in w h i c h t h e RBS w a s p r o v i d e d in t h e b o t t o m f lange on ly for u s e a s a re t rof i t m e a s u r e for ex i s t i ng m o m e n t f r a m e c o n n e c t i o n s . T h e s e RBS re t rof i t t e s t s a re n o t r e p o r t e d in A p p e n - dix A. I n f o r m a t i o n o n t h e t e s t s is ava i lab le in t h e AISC Steel Design Guide Series Twelve (Gross, et .al . 1999).

3.1 O v e r v i e w o f T e s t R e s u l t s f o r R a d i u s Cut RBS S p e c i m e n s

Thi s s ec t i on p r o v i d e s a n overv iew of t h e t e s t d a t a l i s t ed in A p p e n d i x A for r a d i u s c u t RBS t e s t s p e c i m e n s . T h e r e a re 27 r a d i u s c u t RBS t e s t s l i s ted in t h e table . E x a m i n a t i o n of t h i s d a t a i n d i c a t e s t h a t t h e s e c o n n e c t i o n s devel- o p e d p la s t i c r o t a t i o n s r a n g i n g f rom 0 .029 r a d to b e y o n d 0 .05 rad . T h e s e r e s u l t s s u g g e s t t h a t t h e r a d i u s c u t RBS c o n n e c t i o n c a n deve lop la rge p l a s t i c r o t a t i o n s o n a cons i s - t e n t bas i s . Also n o t a b l e is t h e fact t h a t a

4

DESIGN OF REDUCED BEAM SECTION (RBS) MOMENT FRAME CONNECTIONS

large n u m b e r of rad ius cut RBS connect ions have been tested u n d e r a variety of condi- tions by a n u m b e r of different investigators, and there has not been a single test with poor performance. This suggests the connec- tion is quite robus t and reliable.

The da ta in Appendix A demons t r a t e s the possible u l t imate failure modes for the rad ius cut RBS connect ion. In m a n y tests, spec imen s t rength gradual ly deter iorated due to local and lateral torsional buckl ing, and test ing was te rmina ted due to l imitat ions of the test- ing equ ipment or test setup. However, a n u m b e r of connec t ions have been loaded well past the occur rence of local flange buckl ing within the RBS, and ul t imately failed by low cycle fatigue f racture of the RBS. Only one of the 27 rad ius cut RBS spec imens experi- enced a f racture at the beam- to -co lumn con- nection. This specimen, des ignated "DBBW- C - Beam 2" in Appendix A, f rac tured in the beam bot tom flange base metal ad jacent to the groove weld, with the f racture init iat ing at the weld access hole. However, even this connect ion developed 0.038 rad. of plastic rotat ion prior to fracture.

Most of the rad ius cut RBS spec imens h a v e been tested pseudo statically, us ing a loading protocol in which applied displace- men t s are progressively increased. However, one spec imen ("S-l") was tested monotoni- cally to failure. Two spec imens ("LS-2" and "LS-3") were tested us ing a loading protocol in tended to represen t nea r source g round motions tha t conta in a large pulse. Finally, two spec imens ("S-4" and "SC-2") were tes ted dynamically. The rad ius cu t RBS spec imens have performed well u n d e r all of these load- ing condit ions.

A wide range of beam sizes have been tested with the rad ius cu t RBS. The smal les t beam listed in Appendix A is a W530x82 (Canadian designation) which is rough ly equivalent to a W2 lx50. The heaviest beam tested is a W36x300. All co lumns for r ad ius cut RBS tests have been W14 sections. Most of the co lumns have been sized to provide for a very s trong panel zone, a l though a small n u m b e r of tests have inc luded modera te panel zone yielding. No tests have been con- ducted on spec imens with very weak panel

zones. However, such tests will be completed dur ing 1999.

Of the 27 rad ius cu t RBS spec imens listed in Appendix A, there are no repor ted cases of weld fracture. Beam flange groove welds for all r ad ius cut RBS spec imens have been made by the self shielded flux cored arc welding process (SS-FCAW) us ing electrodes with a m i n i m u m specified CVN toughness of 20 ft.-lbs, a t - 2 0 ° F. Three different electrode des ignat ions have been u sed in these tests: E71T-8, E70TG-K2, and E70T-6. For one of the rad ius cut RBS specimens , details of the backing bars were not reported. However, for the remain ing 26 spec imens in which back- ing bar details were reported, the bot tom flange backing was removed and the top flange backing was left in place. For the majori ty of these specimens , the top flange backing was seal welded to the face of the column, a l though these seal welds were not provided in four spec imens (WG-1 to WG-4). Note tha t only one of the 27 rad ius cu t RBS spec imens u sed cover plates at the beam-to- co lumn connec t ion as a supp lemen t to the RBS.. The remain ing 26 spec imens u sed no supp lementa l re inforcing m e a s u r e s (cover plates, ribs, etc.) at the connect ion.

Dimens ions of the RBS cuts for the 27 rad ius cu t spec imens vary over a fairly small range. The d is tance from the face of the col- u m n to the s tar t of the RBS cut (designated as L 1 in Appendix A) r anged from 50 to 75% of the beam flange width. The lengths of the cuts (designated as LRB S in Appendix A) have varied from 74 to 82% of the beam depth. The a m o u n t of flange width removed at the m i n i m u m sect ion of the RBS (desig- na t ed as FR in Appendix A) has var ied from 38 to 55%.

Two types of web connect ion details have been used for r ad ius cut RBS test specimens: a welded and a bolted detail. In the welded detail, the beam web is welded directly to the co lumn flange us ing a complete jo int pene- trat ion groove weld. For the bolted detail, fully tens ioned high s t rength bolts are used . Approximately half the spec imens have u s e d the bolted detail, a n d half the welded detail. The da ta indicates no significant difference in per formance for r ad ius cu t specimens .

5

DESIGN OF REDUCED BEAM SECTION (RBS) MOMENT FRAME CONNECTIONS

Beam lateral bracing details have also varied among the rad ius cut RBS specimens. Of the 27 specimens, seven are repor ted to have provided a brace at the RBS. For the remain ing 20 specimens, the lateral brace was typically fur ther away from the RBS placed near the point of load application.

Finally, of the 27 rad ius cut spec imens listed in Appendix A, six were tested with a composite concrete floor slab. For Spec imens "SC-1" and "SC-2," a one- inch gap was inten- tionally left be tween the face of the co lumn and the slab, in an a t tempt to minimize com- posi te act ion. For S p e c i m e n s "DBBW-C Beams 1 & 2" and "DBWW-C Beams 1 & 2," no such gap was provided. No de t r imenta l effects of the slab were observed in any of these tests. In some tests, the investigators noted tha t the slab e n h a n c e d overall energy dissipat ion by delaying beam instability. Note tha t for all composite specimens, no shear s tuds were placed in the region of the RBS or between the face of the co lumn and the s tar t of the RBS.

As descr ibed above, a ra ther wide range of condi t ions has been investigated in RBS test- ing completed to-date. Testing of RBS con- nect ions is cont inuing u n d e r the SAC pro- gram and for specific building cons t ruc t ion projects. The reader is encouraged to r emain abreas t of this data, as it becomes available.

Even though m a n y variables have a l ready been invest igated in RBS testing, there are a n u m b e r of condit ions tha t have received less at tention. These condit ions, when they arise in design, should be approached with cau- tion since da ta is lacking in these areas. In such cases, addit ional test ing may be war- ranted. For example, no rad ius cut RBS con- nect ions to the weak axis of a wide flange col- u m n have been tested, a l though da ta for some other RBS connect ions to the co lumn weak axis are available (see Spec imens "COH-3" and "COH-4" in Appendix A). No spec imens with deep co lumns have yet been considered. Fur ther , no tests on spec imens with very weak panel zones have been con- ducted. Fu tu re resea rch is u n d e r w a y to address these and other issues.

4. RBS D e s i g n Procedure for SMFs

The following sect ions contain r ecommenda- tions for the design of new radius cut RBS m o m e n t connect ions. Based on the suc- cesses out l ined above, and the preference of engineers designing new SMF s t ruc tures , the design methodology presen ted herein focuses on the r ad ius cu t RBS shape. Globally impor tan t design pa ramete r s such as panel zone part icipation, beam shear and overall frame drift are addressed as par t of the rec- o m m e n d e d p rocedu re . Many i m p o r t a n t aspects of m o m e n t connect ion design are applicable and m u s t be considered when designing SMF RBS connect ions. The RBS design methodology should be performed in conjunct ion with available test resul ts as par t of the just i f icat ion of the design proce- dure.

The initial par t of the SMF/RBS design is to de termine the configuration of the m o m e n t frames, the typical bay sizes, p lan dimen- sions and frame locations. Many of these r e q u i r e m e n t s are d e t e r m i n e d by o thers , (architects , owners , developers), bu t the engineer should influence these decis ions based on sound design practices. One exam- ple would be to consider the bay size if a SMF/RBS system is to be utilized. Because of the high m o m e n t gradient ratio associa ted with short bays, more beam flange removal in RBS connect ions will be required for shor t bay f rames t han long bay frames. In addi- tion, beam sizes may be affected. With proper guidance, the engineer can supply informa- tion tha t will help the archi tect develop a ra t ional , efficient bu i ld ing design. Upon de terminat ion of the basic s t ruc tura l pa ram- eters, the engineer can begin the m e m b e r and connect ion design process.

4 .1 R B S D e s i g n

The engineer will begin the design of the s t ruc ture by de termining the force level and drift limits to be incorpora ted as par t of the design. These pa ramete r s are typically set by a model building code such as the Uniform Building Code (1997) or, in the future , the

6

DESIGN OF REDUCED BEAM SECTION (RBS) MOMENT FRAME CONNECTIONS

International Building Code. Once the force level is d e t e r m i n e d b a s e d on site cond t ions , s t r u c t u r a l sys tem, se i smic i ty of t he region a n d t a rge t drift l imits, t he eng inee r c a n begin the des ign of t he se i smic s y s t e m u s i n g the AISC Seismic Provisions for Structural Steel Buildings { 1997).

B a s e d on the r e q u i r e d des ign p a r a m e t e r s , the e n g i n e e r will d e t e r m i n e t he b e a m a n d c o l u m n sizes r e q u i r e d to m e e t drift l imits , etc. It is i m p o r t a n t t h a t t h e e n g i n e e r r e m e m - be r t h a t the f r a m e is less stiff d u e to t he RBS des ign , t h a n a "typical" n o n - R B S SMF.

After p r o p e r b e a m - c o l u m n sizes h a v e b e e n d e t e r m i n e d for t he f rame, t he RBS d e s i g n p r o c e d u r e s h o u l d be fo l lowed to develop the p rope r f lange r e d u c t i o n to pro- d u c e t he des i r ed p e r f o r m a n c e . M a n y of t he des ign s t eps a n d r e c o m m e n d a t i o n s para l le l i n fo rma t ion p rov ided in r epo r t s r e f e r e n c e d a t t he e n d of th i s d o c u m e n t .

The s t r e n g t h of t he b e a m at t he m i n i m u m sec t ion of t h e RBS m u s t sa t is fy code requ i re - m e n t s u n d e r all appl icab le load c o m b i n a - t ions i n c l u d i n g gravity, wind , a n d o the r loads a p p r o p r i a t e for t he s t r u c t u r e u n d e r cons ide r - a t ion. B e a m sizes in typica l SMFs are nor - ma l ly governed by code specif ied drift l imits. C o n s e q u e n t l y , even wi th a r e d u c t i o n in b e a m m o m e n t d u e to t h e add i t i on of t h e RBS, t he s t r e n g t h of t h e modi f i ed f r a m e will of ten be sa t i s fac to ry for all load c o m b i n a t i o n s . In some cases , a m i n o r i n c r e a s e in b e a m size m a y be n e e d e d .

The add i t i on of RBS c u t o u t s will r e d u c e t he s t i f fness of a s teel m o m e n t f rame. This r e d u c t i o n in st i ffness, a l t h o u g h genera l ly qui te smal l , m a y affect t h e abil i ty of t he f r ame to sa t is fy code specif ied drif t l imits . A r e c e n t s t u d y by G r u b b s (1997) e v a l u a t e d the r e d u c t i o n in e las t ic la te ra l s t i f fness of s teel m o m e n t f r a m e s d u e to the add i t i on of r a d i u s cu t RBS c o n n e c t i o n s . This s t u d y s h o w e d t h a t over a wide r a n g e of f r a m e h e i g h t s a n d conf igura t ions , t he ave rage r e d u c t i o n in stiff- n e s s for a 50 p e r c e n t f lange r e d u c t i o n w a s on the o rde r of 6 to 7 pe rcen t . For a 40 p e r c e n t f lange r e d u c t i o n , the r e d u c t i o n in e las t ic f r ame st i f fness w a s on the o rde r of 4 to 5 per- cent . If th i s r e d u c t i o n in s t i f fness is a con- cern , drift c a n be c o m p u t e d in t he u s u a l m a n n e r u s i n g a m o d e l t h a t does no t explic-

itly a c c o u n t for t he RBS, a n d t h e n i n c r e a s e d by the a m o u n t s n o t e d above to a c c o u n t for t he RBS c o n n e c t i o n s . Al ternat ively , a re f ined s t r u c t u r a l mode l , i n c l u d i n g the r e d u c e d stiff- n e s s a t e a c h c o n n e c t i o n d u e to t h e RBS, c a n be deve loped to c h e c k the s t i f fness of t h e f rame.

4 . 2 RBS S i z i n g

The loca t ion a n d size of t h e RBS will d ic ta te the level of s t r e s s a t t h e b e a m f l a n g e - c o l u m n f lange c o n n e c t i o n . The RBS se i smic m o m e n t d i a g r a m is p r e s e n t e d in F igure 4.1 a n d indi- ca t e s t he Nomina l Capac i ty , t he Probab le D e m a n d , a n d the Nomina l D e m a n d for t h e RBS b e a m . Note t h a t M ' p RBS is t h e maxi - m u m m o m e n t e x p e c t e d a t l~he face of t he col- u m n f lange w h e n t h e RBS h a s y ie lded a n d s t r a in h a r d e n e d u n d e r c o m b i n e d e a r t h q u a k e a n d gravi ty loads . M' p RBS is d i rec t ly influ- e n c e d by the P robab le i J e m a n d , a n d t h e loca- t ion of t h e RBS. M' P ,RBS is l a te r r e fe r red to as Mf in th i s d o c u m e n t .

r - - ~ r . . . . . . , ~ ; ~ , ~ - ~ , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i , \ I ,

~ ~,~as i

~--~,-,,~ o ~ Moment Diegrem

L ~

~am ¢ , ¢ ~ y

Figure 4. I M o m e n t D i a g r a m a n d B e a m G e o m e t r y for RBS

The overal l goal in s izing t h e RBS c u t is to limit t he m a x i m u m b e a m m o m e n t t h a t c a n develop a t t h e face of t h e c o l u m n to v a l u e s in the r a n g e of a b o u t 85 to 100 p e r c e n t of t h e b e a m ' s a c t u a l p l a s t i c m o m e n t . Th i s a p p r o a c h , in effect, l imi ts t he ave rage maxi - m u m s t r e s s a t t h e b e a m f lange groove we lds to v a l u e s on t h e o rde r of t h e a c t u a l y ie ld s t r e s s of t h e b e a m . E x p e r i m e n t s h a v e s h o w n t h a t c o n n e c t i o n s de t a i l ed in a c c o r d a n c e wi th

7

DESIGN OF REDUCED BEAM SECTION (RBS} MOMENT FRAME CONNECTIONS

t h e r e c o m m e n d a t i o n s p r o v i d e d be low are c a p a b l e of s a f e ly r e s i s t i n g t h i s level of m o m e n t . As a p o i n t of c o m p a r i s o n , t e s t s o n p r e - N o r t h r i d g e m o m e n t c o n n e c t i o n s w i t h o u t RBS c u t o u t s o f ten s h o w m a x i m u m m o m e n t s a t t h e face of t h e c o l u m n of a b o u t 125 per - c e n t of M~ or g r ea t e r (Popov, S t e p h e n 1972; Tsai , PopoPv 1988; E n g e l h a r d t , H u s a i n 1993). C o n s e q u e n t l y , t h e a d d i t i o n of t h e RBS c u t o u t s in t h e b e a m r e s u l t s in a s u b s t a n t i a l r e d u c t i o n in m o m e n t a t t h e face of t h e col- u m n .

M u c h of t h e d e s i g n p r o c e d u r e p r e s e n t e d b e l o w fo l lows r e c o m m e n d a t i o n s of t h e Interim Guidelines: Evaluation, Repair, Modi- fication and Design o f Welded Steel Moment Frame Structures (FEMA 267) (1995) a n d t h e Interim Guidelines Advisory No. 1, Supple- ment to FEMA 267 (FEMA 267A) (1997), w i th severa l excep t ions . Mos t s ign i f i can t of t h e s e e x c e p t i o n s is t h a t FEMA 267A p l a c e s a l imi t on t h e m a x i m u m s t r e s s p e r m i t t e d a t t h e face of t h e c o l u m n e q u a l to n i n e t y p e r c e n t of t h e m i n i m u m spec i f ied y ie ld s t r e s s of t h e col- u m n . For t h e c a s e of a n A992 (A572 Gr. 50) c o l u m n , t h i s r e s u l t s in a l imi t of 45 ksi . Th i s l imi t w a s e s t a b l i s h e d to a d d r e s s c o n c e r n s r e g a r d i n g t h e p o t e n t i a l for t h r o u g h - t h i c k n e s s f a i lu res in c o l u m n f langes . The d e s i g n p roce - d u r e l imi t s t h e m a x i m u m s t r e s s a t t h e face of t h e c o l u m n to a v a l u e o n t h e o r d e r of t h e a c t u a l y ie ld s t r e s s of t h e b e a m . Th i s excep- t ion to t h e r e q u i r e m e n t s of FEMA 267A h a s b e e n a d o p t e d for severa l r e a s o n s . Fi rs t , spec- i m e n s d e s i g n e d a c c o r d i n g to t h e p r o c e d u r e s d e s c r i b e d h e r e i n h a v e p e r f o r m e d well in lab- o r a to ry t es t s . S e c o n d , s a t i s fy ing t h e 45 ks i s t r e s s l imit , w o u l d r e s u l t in la rge f lange c u t o u t s in m a n y cases , or w o u l d r e q u i r e s u p - p l e m e n t a l f lange r e i n f o r c e m e n t s u c h as cover p l a t e s or r ibs. F u r t h e r , r e c e n t l y c o m p l e t e d r e s e a r c h c o n d u c t e d u n d e r t h e SAC P h a s e 2 p r o g r a m s u g g e s t s t h a t t h e p o t e n t i a l for t h r o u g h - t h i c k n e s s f a i lu res is c o n s i d e r a b l y less t h a n p r e v i o u s l y t h o u g h t , a n d t h a t t he c u r r e n t l imit of 45 ks i c a n m o s t l ikely be i n c r e a s e d w i t h o u t p o s i n g a n i n c r e a s e in r i sk of f r a c t u r e in i t ia t ion .

The d e s i g n p r o c e d u r e a s s u m e s t h a t a r a d i u s c u t RBS is p r o v i d e d in b o t h t h e top a n d b o t t o m f l anges a t t h e m o m e n t c o n n e c - t i on a t e a c h e n d of a m o m e n t f r a m e b e a m .

T h e p r o c e d u r e a l so a s s u m e s t h e m i n i m u m spec i f ied y ie ld s t r e s s of t h e b e a m is 50 ks i or l e ss (Gr. 50 b e a m s ) , a n d t h a t t h e m i n i m u m spec i f i ed y ie ld s t r e s s of t h e c o l u m n is 50 ks i or g r ea t e r (Gr. 50 or Gr. 65 c o l u m n s ) .

F igu re 4 .2 s h o w s t h e g e o m e t r y of a r a d i u s c u t RBS, a n d F igu re 4 .3 s h o w s t h e en t i r e m o m e n t f r a m e b e a m . The key d i m e n s i o n s

I ~ ~ 1 ~ a

4 c ~ + d R = rad ius of cut 8c

C

~1 - - 1

b

Figure 4 . 2 G e o m e t r y o f R a d i u s C u t RBS

t h a t m u s t be c h o s e n by t h e d e s i g n e r a re a, t h e d i s t a n c e f rom t h e face of t h e c o l u m n to t h e s t a r t of t h e RBS cu t , b, t h e l e n g t h of t h e RBS cu t , a n d c, t h e d e p t h of t h e RBS c u t a t i t s m i n i m u m sec t ion . The r a d i u s of t h e c u t R c a n be r e l a t e d to d i m e n s i o n s b a n d c b a s e d o n t h e g e o m e t r y of a c i r cu l a r arc , u s i n g t h e e q u a t i o n in Fig. 4 .2 . T h e a m o u n t of f lange m a t e r i a l t h a t is r e m o v e d a t t h e m i n i m u m s e c t i o n of t h e RBS is s o m e t i m e s r e f e r r ed to the percent f lange removal w h i c h is c o m - p u t e d a s (2c/bf.) x 100, w h e r e bf i s the u n r e - d u c e d f l ange v~idth of t h e beam~

In p a s t r e s e a r c h t e s t s , t h e d i m e n s i o n s a a n d b h a v e gene ra l l y b e e n c h o s e n b a s e d o n t h e j u d g m e n t of t h e r e s e a r c h e r s . In genera l , t h e s e d i m e n s i o n s s h o u l d be k e p t a s sma l l a s

• w = uniform beam gravity load ~ II II RBS RBS

_ _ ~ ~.~_.£1l.~r.! ~ ~ 1 I } I I t ~ ~ 1 t I } ~ l ~ l ~ ~ . ! ? . . t . ~ . ! . | ~ [ ~ ]

' &4 i i ~ ,- ,n -~ ,n - ~

•

,, l l a + ~ " L' = distance be~een ~nters of RBS ~ts ~a+ ~ ~

I ~ L : distance between column ¢entedines

Figure 4 . 3 T y p i c a l M o m e n t F r a m e B e a m w i t h

RBS C o n n e c t i o n s

8

DESIGN OF REDUCED BEAM SECTION (RBS) MOMENT FRAME CONNECTIONS

possible in order to minimize the increase of m o m e n t between the plastic hinge located in the RBS and the face of t_he column.

The d imens ion a should be large enough, however, to permit s t ress in the reduced sec- tion of the beam to spread uniformly across the flange width at the face of the column. Similarly, the d imens ion b should be large enough to avoid excessive inelast ic s t ra ins within the RBS. Based on an evaluat ion of successful pas t tests, the following sugges- t ions are made for selecting these d imen- sions:

(o.s to o.Ts) bf tl)

b ~ (65 to 0 .85 )d (2)

where by and d are the beam flange width and delSth. Examina t ion of RBS test da ta indicates tha t successfu l connect ion per- formance has been obta ined for a wide range of values for a and b. Consequent ly , a great deal of precis ion in choosing these values does not appear just if ied and Equa t ions 1 and 2 should be cons idered an approximate guide.

The remain ing d imens ion tha t m u s t be chosen w h e n sizing the RBS is c, the depth of the cut. The value of c will control the maxi- m u m m o m e n t developed within the RBS, and therefore will control the m a x i m u m m o m e n t genera ted at the face of the column. As noted above, the final d imens ions should be chosen so tha t the m a x i m u m m o m e n t at the face of the co lumn is in the range of abou t 85 to 100 percent of the beam's ac tua l plastic moment . At present , it is sugges ted to avoid utilizing flange reduc t ions greater t han about 50 per- cent. Thus, the value of c should be chosen to be less t han or equal to 0.25bf.

The basic approach t aken in "this proce- dure is to choose pre l iminary values for a, b, and c, then compute the m o m e n t at the face of the co lumn, and check this m o m e n t against the limit no ted above. Some iteration in the RBS d imens ions may be needed to arrive upon a sat isfactory design. Fur ther design checks are comple ted upon satisfac- tory sizing of the RBS.

The beam size will typically be chosen for drift requi rements , followed by some a m o u n t of flange reduct ion. The designer m u s t exam- ine the effect of all applied loads at the RBS

location. It is possible tha t beam size may need to be adjus ted , and different RBS sizing and location m u s t be de termined , to meet all design criteria.

This RBS sizing de te rmina t ion is also applicable when retrofit t ing existing SMF s t ruc tures . Access is l imited or impossible at the uppe r flange of the beam, due to the p resence of a floor slab, so RBS modificat ions typically occur at the bot tom flange of the m o m e n t beam only. If access is available to the top flange of the beam, it is r e c o m m e n d e d to apply the RBS design methodology to both flanges. There h a s been a great deal of effort and resea rch spent on the use of RBS modi- fications to existing SMFs. The AISC Design Guide Series Twelve (1999) tha t summar izes this work, conta ins a significant a m o u n t of informat ion regarding retrofit of SMFs utiliz- ing RBS connect ion modifications. It is rec- o m m e n d e d tha t des igners us ing an RBS approach to retrofit an existing SMF refer to the AISC d o c u m e n t prior to utilizing the design methodology con ta ined herein.

Upon selection of the b e a m - c o l u m n com- binat ion to be util ized in the SMF design and the location, shape a n d size of the RBS, fur- ther connect ion design checks are requi red to ensu re the design will perform in a ducti le manne r .

The first check shou ld be the "Strong Col- u m n - W e a k Beam" confirmation. This check is i n t ended to limit inelast ic deformat ions of co lumns outs ide of thei r pane l zone regions. It is generally recognized tha t co lumn yield- ing is an undes i r ab le mode because of the possible effect on the co lumn, and in turn , the global stability of the s t ruc tu ra l frame. The AISC Seismic Design Provisions (1997) out l ine an acceptable design level for the b e a m / c o l u m n relat ionship. As a m i n i m u m , this AISC proviso shou ld be met.

RBS connec t ion des ign m u s t also address the panel zone. The pane l zone is subjec ted to large shear forces as the b e a m s reach their full capacity. Based on FEMA 267A (1997), the panel zone m u s t be s t rong enough to develop at least 80% of the shea r s associa ted with Mfl The pane l zone r e q u i r e m e n t s can be met in one of two ways. One way is to provide a co lumn with a th ick e n o u g h web to resis t the requi red shear in acco rdance with the

9

DESIGN OF REDUCED BEAM SECTION (RBS) MOMENT FRAME CONNECTIONS

d e s i g n r e q u i r e m e n t s . The o t h e r way to s u p - ply suf f ic ien t p a n e l zone s h e a r r e s i s t a n c e is to a d d d o u b l e r p l a t e s to t h e s e l ec t ed sec t ion . D o u b l e r p l a t e s s h o u l d c o n s i s t of t h e r e q u i r e d a d d i t i o n a l t h i c k n e s s of steel , a d d e d to o n e or b o t h s ides of t h e c o l u m n web. F a b r i c a t o r s i n d i c a t e t h a t t h e u s e of a heav ie r c o l u m n sec- t ion, i n s t e a d of d o u b l e r p l a t e s a n d o t h e r l ab o r i n t e n s i v e r e in fo r c ing de ta i l s , m a y r e s u l t in a m o r e e c o n o m i c a l s t r u c t u r a l f r ame .

The f inal d e s i g n c h e c k to be p e r f o r m e d on t h e s e l ec t ed b e a m - c o l u m n c o m b i n a t i o n is t h e b e a m s h e a r . T h e m a x i m u m b e a m s h e a r is d e v e l o p e d in t h e s ec t i on of t h e b e a m b e t w e e n t h e RBS a n d t h e c o l u m n f lange face, w h e r e gravi ty s h e a r a n d s e i s m i c s h e a r co inc ide . At t h i s loca t ion , s h e a r c a p a c i t y of t h e b e a m sec- t ion n e e d s to be c h e c k e d to e n s u r e t h a t t h e b e a m will h a v e a d e q u a t e s h e a r c a p a c i t y af ter t h e p l a s t i c h i n g e in t h e b e a m deve lops d u e to a p p l i e d l a te ra l loads .

The fol lowing s t e p - b y - s t e p p r e s e n t a t i o n o u t l i n e s t h e RBS d e s i g n p r o c e d u r e r e l a t i ng to t h e r e m o v a l of t h e b e a m f lange a n d t h e c h e c k s r e q u i r e d to e n s u r e p r o p e r b e h a v i o r a n d c o r r e l a t i o n w i t h t e s t a n d r e s e a r c h r e s u l t s .

4 . 3 S t e p - b y - s t e p P r o c e d u r e

STEP 2 C o m p u t e t h e p las t i c s ec t i on m o d u - lu s a t t h e m i n i m u m sec t ion of t h e RBS.

F igu re 4 .4 s h o w s a c r o s s - s e c t i o n of t h e b e a m at t h e m i n i m u m sec t ion of t h e RBS.

b~

"~'~"""''~P~ions cut from flange d/2 ~ ~ t w

Plastic Neutral Axis

d/2 /./.~Portions cut from flange

/ _ _ _ ~ ,~ , '~ t

~ ~.~ c c

F i g u r e 4 . 4 B e a m at M i n i m u m S e c t i o n o f R B S

B a s e d o n t h e d i m e n s i o n s s h o w n in t h i s fig- u re , Z R B S c a n be c o m p u t e d a s follows:

STEP 1 C h o o s e t r ia l v a l u e s for RBS d i m e n - s i o n s a, b, a n d c.

The t r ia l v a l u e s for a a n d b s h o u l d be c h o s e n w i t h i n t h e l imi t s of E q u a t i o n s 1 a n d 2. To e s t a b l i s h a t r ia l v a l u e of c, a f l ange r e d u c t i o n of a b o u t 40 p e r c e n t is s u g g e s t e d for t h e in i t ia l d e s i g n i t e ra t ion . T h u s , c h o o s e c ~ 0 .20 b f As n o t e d earl ier , v a l u e s for c in e x c e s s o f a p p r o x i m a t e l y 0 . 2 5 b f a re n o t rec- o m m e n d e d .

a (O.S to 0.75) bf

b ~ (0. 6 5 to O. 85) d

10

Z ~ s = Z b - 2 c t.f (d - t.f ) (3)

Where :

Z R B S = p la s t i c s e c t i o n m o d u l u s a t m i n - i m u m s e c t i o n of RBS

(1)

= p la s t i c s ec t i on m o d u l u s for full b e a m c r o s s - s e c t i o n

(i.e. w i t h o u t f l ange c u t o u t s )

o t h e r va r i ab l e s a s s h o w n in F igure 4.4.

DESIGN OF REDUCED BEAM SECTION (RBS} MOMENT FRAME CONNECTIONS

STEP 3 E s t a b l i s h t h e e x p e c t e d y ie ld s t r e s s of t h e b e a m .

T he e x p e c t e d yie ld s t r e s s for t h e b e a m c a n be d e t e r m i n e d f rom S e c t i o n 6 .2 of t h e AISC Seismic Provisions for Structural Steel Buildings (1997). A c c o r d i n g to t h e s e provi- s ions :

Fy e = Ry Fy (4)

whe re :

Fy e = e x p e c t e d y ie ld s t r e s s

= m i n i m u m spec i f i ed y ie ld s t r e s s

= ra t io of e x p e c t e d to m i n i m u m spec i f i ed y ie ld s t r e s s

= 1.5 for A36 s tee l

T h e fac to r of 1.15 in E q u a t i o n 5 a c c o u n t s for s t r a i n h a r d e n i n g , a n d is b a s e d o n s t r a i n h a r d e n i n g v a i a e s m e a s u r e d in RBS tes t s .

STEP 5 C o m p u t e t h e s h e a r force a t t h e c e n t e r of t h e RBS c u t s a t e a c h e n d of t h e b e a m .

T h e s h e a r a t t h e c e n t e r of t h e RBS c a n be c o m p u t e d f r o m a free b o d y d i a g r a m of t h e m o m e n t f r a m e b e a m t a k e n b e t w e e n RBS c e n t e r s . S u c h a f ree b o d y d i a g r a m is i l lus- t r a t e d in F i g u r e 4 .5 for t h e c a s e of a u n i - fo rmly d i s t r i b u t e d grav i ty l oad w.

f R~BS RBS I w = uniform beam gravity ~oad • l!.~.,~ ~ ~ t ~ I t t t t I t t I t ~ ~ I t I I I t t ~ t . ! . . ! , {

. . . . . .

RBS RBS! i RBS RBS

i L' = distance between centers of RBS ' -I

Figure 4 .5 Free Body Diagram B e t w e e n

C e n t e r s o f RBS

= 1.1 for A572 Gr. 50 a n d A992 s tee l

T h e v a l u e of Fve r e c o g n i z e s t h a t t h e a c t u a l y ie ld s t r e n g t l ~ o f s t r u c t u r a l s tee l c a n s ign i f i can t ly e x c e e d t h e m i n i m u m spec i f i ed va lue .

S u m m i n g m o m e n t s a b o u t e a c h e n d of t h i s f ree b o d y d i a g r a m r e s u l t s i n t h e follow- ing:

2MRBs wL' V~S - L ' + - ~ - (6a)

S T E P 4 C o m p u t e t h e m a x i m u m m o m e n t e x p e c t e d a t t h e c e n t e r of t h e RBS.

MRB S = 1.15 ZRB S Fy e (5) 2 MRB s wL'

V~O~S - L ' 2 (6b)

w h e r e : w h e r e :

MRB S =

ZRB S =

m a x i m u m m o m e n t e x p e c t e d a t t h e c e n t e r of t h e RBS

p l a s t i c s e c t i o n m o d u l u s a t m i n - i m u m s e c t i o n of t h e RBS

e x p e c t e d y ie ld s t r e s s of b e a m

VRBS V' BS = s h e a r force a t t h e c e n t e r of t h e RBS a t e a c h e n d of b e a m

L ' = d i s t a n c e b e t w e e n c e n t e r s of RBS

W = u n i f o r m l y d i s t r i b u t e d g r av i t y l o a d o n b e a m

11

D E S I G N OF RE DUC E D B E A M S E C T I O N (RBS} MOMENT FRAME CONNECTIONS

For gravi ty load c o n d i t i o n s o t h e r t h a n a u n i f o r m load, t h e a p p r o p r i a t e a d j u s t m e n t c a n eas i ly be m a d e to t h e free b o d y d i a g r a m a n d to E q u a t i o n s 6 a a n d 6b.

E q u a t i o n s 6 a a n d 6b a s s u m e t h a t p l a s t i c h i n g e s will fo rm a t t h e RBS a t e a c h e n d of t h e b e a m . If t h e gravi ty load o n t h e b e a m is ve ry large, t h e p l a s t i c h i n g e a t one e n d of t h e b e a m m a y m o v e t o w a r d t h e in te r io r p o r t i o n of t h e b e a m s p a n . If t h i s is t h e case , t h e free b o d y d i a g r a m in F igu re 4 .5 s h o u l d be mod i - fied to e x t e n d b e t w e e n t h e a c t u a l p las t i c h i n g e loca t ions . To c h e c k if E q u a t i o n s 6 a a n d 6b a re valid, d r a w t h e m o m e n t d i a g r a m for t h e s e g m e n t of t h e b e a m s h o w n in F igu re 4 .5 , i.e., for t h e s e g m e n t of t h e b e a m b e t w e e n t h e c e n t e r s of t h e RBS cu t s . If t h e m a x i m u m m o m e n t o c c u r s a t t h e e n d s of t h e s p a n s , t h e n E q u a t i o n s 6 a a n d 6b a re valid. If t h e m a x i m u m m o m e n t o c c u r s w i t h i n t h e s p a n , a n d e x c e e d s Mp. e of t h e b e a m (see E q u a t i o n 8), t h e n t h e m o d i f i c a t i o n d e s c r i b e d above will be n e e d e d .

STEP 6 C o m p u t e t h e m a x i m u m m o m e n t e x p e c t e d a t t h e face of t h e c o l u m n .

M f = Mp, B s + VRB s a +

where :

(7)

= m a x i m u m m o m e n t e x p e c t e d a t t h e face of t h e c o l u m n

a l l o t h e r va r i ab l e s a s p r e v i o u s ~ de f i ned

E q u a t i o n 7 n e g l e c t s t h e gravi ty load on t h e p o r t i o n of t h e b e a m b e t w e e n t h e c e n t e r of t h e RBS a n d t h e face of t h e c o l u m n . Th i s s impl i f ies t h e e q u a t i o n a n d i n t r o d u c e s little error . If d e s i r e d , t h e grav i ty l oad on th i s sma l l p o r t i o n of t h e b e a m c a n be i n c l u d e d in t h e free b o d y d i a g r a m a n d in E q u a t i o n 7.

STEP 7 C o m p u t e t h e p l a s t i c m o m e n t of t h e b e a m b a s e d on t h e e x p e c t e d y ie ld s t r e s s .

Mpe = Z b Fy e (8)

T h e m o m e n t a t t h e face of t h e c o l u m n c a n be c o m p u t e d f rom a free b o d y d i a g r a m of t h e s e g m e n t of t h e b e a m b e t w e e n t h e c e n t e r of t h e RBS a n d t h e face of t h e c o l u m n f lange. S u c h a free b o d y d i a g r a m is i l l u s t r a t e d in F i g u r e 4 .6 .

RBS

- - M f ..... "~". VRB s MRB s

~ ,

I- b - - - N a +.-Z-

Figure 4 . 6 Free B o d y D iagra m B e t w e e n C e n t e r o f

RBS a n d Face o f C o l u m n F lang e

S u m m i n g m o m e n t s a b o u t t h e left e n d of t h i s f ree b o d y d i a g r a m r e s u l t s in t h e follow- ing:

whe re :

Mpe = p la s t i c m o m e n t of b e a m b a s e d on e x p e c t e d y ie ld s t r e s s .

STEP 8 C h e c k t h a t M f i s in t h e r a n g e of 85 to 100 p e r c e n t of Mpe.

M.f ~0 .85 to 1.0 (9) m pe

If E q u a t i o n 9 is n o t sa t i s f ied , m o d i f y t h e v a l u e s of c a n d / o r a a n d b a s n e e d e d , a n d r e p e a t S t e p s 2 t h r o u g h 8. Note t h a t t h i s c h e c k on m o m e n t a t t h e face of t h e c o l u m n is s impl i f i ed for d e s i g n p u r p o s e s , b a s e d on m o r e d e t a i l e d a n a l y s e s a n d p a s t t e s t r e su l t s . The a c t u a l force t r a n s f e r m e c h a n i s m a n d s t a t e of s t r e s s a n d s t r a i n a t t h i s l oca t ion is qu i t e c o m p l e x d u e to t h e c o n s t r a i n t gene r - a t e d by t h e c o n n e c t i o n to t h e c o l u m n f lange. For m o r e d e t a i l e d i n f o r m a t i o n on t h e i s sue , t h e r e a d e r is r e f e r r ed to (Lee, et .al . 1997).

12

DESIGN OF REDUCED BEAM SECTION (RBS) MOMENT FRAME CONNECTIONS

STEP 9 Strong Column-Weak Beam Check

To c h e c k s t rong c o l u m n - w e a k b e a m Z Mc requi rements , the p rocedure presen ted in FEMA 267A (1997) will be used , with minor Where: modifications. The equat ion to be used to c h e c k th is r e q u i r e m e n t (from E q u a t i o n Vc = 7.5.2.5-1 of FEMA 267A (1997)) is as follows:

= M c t + Me b (14)

s h e a r force in the c o l u m n s above and below the connect ion

~ Z¢(F~c - J~) > 1.0 (10) Mct ZMc

= c o l u m n m o m e n t above connec t ion

immedia te ly

where: Mcb = c o l u m n m o m e n t immedia te ly below connec t ion

plast ic sect ion m o d u l u s of the c o l u m n sec t ion above a n d below the connect ion

ht dis tance from top of b e a m to point of inflection in the col- u m n above the connec t ion

YMc = m i n i m u m specified yield s t ress of the co lumn

= axial s t r e s s in the c o l u m n above and below the connect ion

~VM c s u m of the co lumn m o m e n t s at the top a n d bot tom of the panel zone c o r r e s p o n d i n g to the development of M R B S at the c e n t e r of the RBS in the a t t ached beams

Figure 4.7 shows a free body d iagram tha t can be u sed to es t imate co lumn m o m e n t s w h e n checking Equat ion 10. This free body cuts the beams at the RBS centers and cuts the co lumns at a s s u m e d points of inflection (often t aken as mid-height of the ad jacent stories for design purposes).

Based on Figure 4.7, £'M c can be esti- ma ted from the following equat ions:

, ,(de _,~ Z M R~s + (VR~s + V ~ s ) ~ - + a +

2J V~ : (11)

h t + d b + h b

Mct = Vch t (12)

Mcb = Vch b (13)

d c = depth of co lumn

hb dis tance from bot tom of b e a m to point of inflection in the col- u m n below the connec t ion

d b = depth of beam

All o ther variables as previously defined.

Mct

~ -,,~-.-.~ V ~i C

i Mcb

I

l I I

a+(b/2) d c a+(b/2)

Figure 4 . 7

~ MRBS V RBS

Free B o d y D i a g r a m for C a l c u l a t i o n o f C o l u m n M o m e n t s

ht

d b

hb

13

D E S I G N OF REDUCED B E A M S E C T I O N (RBS) MOMENT FRAME CONNECTIONS

T he a p p r o a c h p r e s e n t e d in FEMA 267A (1997) a c c o u n t s for t h e d i f fe rence in c o l u m n s h e a r forces a bove a n d be low t h e c o n n e c t i o n , w h e r e a s t h e s i m p l i f i e d a p p r o a c h a b o v e a s s u m e s t h e s a m e s h e a r force is p r e s e n t in t h e c o l u m n s above a n d be low t h e c o n n e c - t ion. A l t h o u g h t h e a p p r o a c h in FEMA 267A (1997) m a y be s o m e w h a t m o r e a c c u r a t e , t h e c o m p u t a t i o n of V c p r e s e n t e d in E q u a t i o n 11 above is s i m p l e r to i m p l e m e n t , a n d is still r e a s o n a b l y a c c u r a t e for in i t ia l d e s i g n p u r - p o s e s c o n s i d e r i n g t h e n u m e r o u s u n c e r t a i n - t ies invo lved in t h e s t r o n g c o l u m n - w e a k b e a m d e s i g n p h i l o s o p h y . T h e r e a d e r is r e f e r r ed to Sec t i on 7 .5 .2 .5 of FEMA 267A (1997) to i m p l e m e n t a m o r e a c c u r a t e ca lcu- l a t ion for V c to be u s e d in t h e f inal d e s i g n check .

S T E P 10 C h e c k Pane l Z o n e

To c h e c k t h e c o l u m n p a n e l zone , t h e p ro- c e d u r e u s e d in Sec t i on 6 . 6 . 6 . 3 . 7 of FEMA 2 6 7 A (1997) will be u s e d . T h i s s e c t i o n r e q u i r e s t h a t t h e p a n e l zone h a v e suf f ic ien t s t r e n g t h to deve lop t h e s h e a r force d e v e l o p e d by 0 .8 £'M/: B a s e d on t h i s a p p r o a c h , t h e p a n e l z o n e ' s h e a r force c a n be c o m p u t e d as follows:

M? = m a x i m u m m o m e n t e x p e c t e d a t o p p o s i t e c o l u m n face

All o t h e r va r i ab l e s as p r e v i o u s l y def ined .

The v a l u e of My c o m p u t e d a c c o r d i n g to E q u a t i o n 7 c o m b i n e s the , s e i s m i c m o m e n t d u e to (2XMRBs)/L' w i t h t h e m o m e n t d u e to grav i ty load. O n t h e s ide of t h e c o l u m n oppo- s i te to t h a t w h e r e My is deve loped , t h e m o m e n t a t t h e face of" t h e c o l u m n will be s o m e w h a t s m a l l e r s ince t h e gravi ty load m o m e n t will o p p o s e t h e s e i s m i c m o m e n t . T h i s s o m e w h a t s m a l l e r m o m e n t is c a l c u l a t e d u s i n g E q u a t i o n 17.

The s t r e n g t h of t h e p a n e l zone c a n be cal- c u l a t e d a s follows:

3b c t~ V = 0.55Fycdct 1 + dbdc--~ ~ (18)

where :

V = p a n e l zone s h e a r s t r e n g t h

M'f = M ~ S + V~S a + (15)

•Mf= Mf+ M~r (16)

o.8Z Vpz - 0.8V c (17)

0.95 d b

Where :

b c = w i d t h of c o l u m n f lange

tc f = t h i c k n e s s of c o l u m n f lange

= to ta l t h i c k n e s s of p a n e l zone i n c l u d i n g d o u b l e r p l a t e s

All o t h e r va r i ab l e s as p r e v i o u s l y def ined .

S T E P 11 C h e c k B e a m S h e a r

Vpz p a n e l zone s h e a r force corre- s p o n d i n g to t h e d e v e l o p m e n t of 80 p e r c e n t of t h e m a x i m u m e x p e c t e d c o l u m n face m o m e n t s

m a x i m u m m o m e n t e x p e c t e d a t t h e face of t h e c o l u m n , ca lcu- l a t ed a c c o r d i n g to E q u a t i o n 7

The f inal d e s i g n c h e c k s h o u l d be m a d e to e n s u r e t h a t t h e b e a m h a s a d e q u a t e c a p a c i t y for s h e a r a s s s o c i a t e d w i th l a te ra l a n d grav i ty loads . Th i s c h e c k c o m b i n e s t h e b e a m s h e a r a s s o c i a t e d w i t h t h e p l a s t i c m o m e n t w i t h i n t h e RBS u s i n g E q u a t i o n 6a, c o m b i n e d w i th t h e p o r t i o n of gravi ty load a d d i n g s h e a r to t h e b e a m w i t h i n t h e s e c t i o n b e t w e e n t h e RBS

14

DESIGN OF REDUCED BEAM SECTION (RBS) MOMENT FRAME CONNECTIONS

center and the co lumn flange. This can be calculated us ing Equat ion 19:

VRB s q

( /- / , ) W - -

2 (19)

2

4 . 4 A d d i t i o n a l D e s i g n C o n s i d e r a - t i o n s

In addi t ion to es tabl ishing the d imens ions of the RBS cut, there are a n u m b e r of addi- t ional design and detail ing features tha t may significantly affect connect ion per formance and economy of this system. These i tems are d i scussed below.

The p rocedure p resen ted above for sizing the RBS cut permits a range of acceptable values for the d imens ions a, b and c. Fabri- cation can likely be simplified by s tandardiz- ing these d imens ions over a large n u m b e r of beams on a project. Making small changes on the RBS d imens ions from beam to beam is not likely to improve connect ion perform- ance and may unnecessa r i ly increase fabri- cat ion costs. The designer may wish to con- sult with a fabricator before finalizing the RBS d imens ions to identify ways of r educ ing fabrication costs. For example, if the fabrica- tor is m a k i n g RBS cu t s u s i n g a to rch m o u n t e d on a guide with a fixed radius , the economy of the connec t ion may be improved by main ta in ing a cons tan t rad ius of cu t R over a large n u m b e r of connect ions .

The RBS cut is normal ly made by the rmal cut t ing in the fabrication shop. The cut should be made to avoid nicks, gouges, a n d other discont inui t ies . After the cut is made , the surface should be ground, to aid in reduc ing the potent ial for f ractures occurr ing in the RBS at h igh plastic rotat ions and low cycle fatigue. The grinding should be done to avoid p roduc ing grind marks perpendicu la r to the beam flange, since they are perpendi- cular to the direction of principal stress. These m a r k s can act as s t ress risers. Varia- t ions on grinding me thods may be possible to reduce fabrication effort.

Another cons idera t ion for design of RBS m o m e n t connec t ions is welding. Research

conduc ted since the Northridge ea r thquake has demons t r a t ed the impor tance of weld metal t oughness in the groove welds of seis- mic res i s tan t m o m e n t connect ions (Kauf- m a n n , et.al. 1996; Tide 1998 I. The AISC Seis- mic Provis ions (1997) r e c o m m e n d s the use of a filler metal with a m i n i m u m specified ten- sile s t rength of 70 ksi, (assuming a 50 ksi base mater ia l specified yield) and a m i n i m u m specified CVN value of 20 ft.-lb, at -20 ° F. Previous resea rch tes ts on RBS connect ions have generally employed the self-shielded flux cored arc welding process (FCAW), us ing E70TG-K2, E71T-8 or E70T-6 electrodes. All of these electrodes provide a m i n i m u m spec- ified CVN of 20 ft.-lb, at -20 ° F. A n u m b e r of other FCAW electrodes are available tha t pro- vide this m i n i m u m CVN value. In addition, successful tes ts on other types of connec- t ions have employed the shielded metal arc welding {SMAW) process us ing an E7018 electrode. The final choice of welding process and electrode is bes t left to the fabricator. Other factors, s u c h as the mixing of different filler meta ls in the same weld jo in t may resul t in lower CVN values for the combinat ion, t han for one of the filler meta ls alone. A paper wri t ten on this subject , "The Effects of Intermixed Weld Metal on Mechanical Prop- erties" (Johnson, Q u i n t a n a 1998), may be useful to the engineer w h e n consider ing the inter-mixing of weld filler metals .

At the beam flange complete joint pene- t rat ion welds, it is r e c o m m e n d e d tha t the weld run-off tabs be removed at both the top and bot tom flanges, a n d tha t the edges of the groove welds be g round smooth. The pre- ferred final profile of the weld tab g round surface is r ad iused , to fur ther reduce the possibility of f rac ture at these locations. This will minimize any potent ia l no tches intro- duced by the p re sence of the weld tabs, or by discont inui t ies con ta ined in the weld meta l within the run-off regions. In addit ion, it is r e c o m m e n d e d tha t the bot tom flange steel backing be removed a n d a reinforcing fillet be placed at the base of the weld after the jo in t is backgouged to s o u n d metal . This require- men t is in tended both to e l iminate the no tch effect p roduced by the steel backing, and to permit bet ter inspec t ion and u l t rasonic test- ing of the weld. At the top flange groove weld,

15

DESIGN OF REDUCED BEAM SECTION (RBS) MOMENT FRAME CONNECTIONS

it is r e c o m m e n d e d t h a t t h e s teel b a c k i n g be sea l w e l d e d to t h e face of t he c o l u m n u s i n g a m i n i m u m size fillet weld, typica l ly a 5 / 1 6 " fil- let. Ana lys i s h a s i n d i c a t e d t h a t t h e n o t c h effect of t h e s tee l b a c k i n g is n o t as severe a t t h e top f lange, a n d t h a t w e l d i n g t h e s tee l b a c k i n g to t h e c o l u m n f u r t h e r r e d u c e s t h e n o t c h effect. F u r t h e r , de fec t s a re l ess l ikely a t t h e t op f lange we ld s ince t h e groove weld is n o t i n t e r r u p t e d by t h e b e a m web, as it is a t t h e b o t t o m f lange.

M a n y r e s e a r c h e r s a n d d e s i g n e r s bel ieve t h a t t h e we ld a c c e s s ho le h a s a n i m p o r t a n t effect o n c o n n e c t i o n p e r f o r m a n c e . A l t h o u g h c u r r e n t r e s e a r c h is a d d r e s s i n g i s s u e s r e l a t e d to t h e we ld a c c e s s hole , t h e r e a p p e a r s to be n o c o n s e n s u s a s of ye t on t h e o p t i m u m size a n d s h a p e . C o n s e q u e n t l y , p e n d i n g f u r t h e r r e s e a r c h , a c c e s s ho le g e o m e t r y s h o u l d con- fo rm to t h e r e q u i r e m e n t s s h o w n in F igu re 5 .2 of AWS D 1 . 1 - 9 8 (AWS 1998). T h e r e is n o i n d i c a t i o n t h a t we ld a c c e s s ho le size, w i t h i n t h e AWS l imi ts , will adve r se ly affect t h e pe r - f o r m a n c e of RBS m o m e n t c o n n e c t i o n s . There fore , size a n d s h a p e of t h e a c c e s s ho le s h o u l d be left to t h e f ab r i ca to r to c o n f o r m to AWS r e c o m m e n d a t i o n s .

A n o t h e r i m p o r t a n t a s p e c t of w e l l - b e h a v e d m o m e n t c o n n e c t i o n s a r e t h e c o n t i n u i t y p l a t e s b e t w e e n t h e c o l u m n f langes . All of t h e s u c c e s s f u l t e s t s o n RBS c o n n e c t i o n s for n e w c o n s t r u c t i o n (Appendix A) h a v e e m p l o y e d c o n t i n u i t y p la tes . However , n o RBS t e s t s to d a t e h a v e o m i t t e d c o n t i n u i t y p la tes , so it is u n c l e a r u n d e r w h a t c o n d i t i o n s c o n t i n u i t y p l a t e s a re a c t u a l l y r equ i r ed . P e n d i n g t h e ou t - c o m e of f u r t h e r r e s e a r c h , it is r e c o m m e n d e d t h a t c o n t i n u i t y p l a t e s be p r o v i d e d for all RBS c o n n e c t i o n s , w i t h a c o n t i n u i t y p l a t e t h i ck - n e s s s imi l a r to t h e b e a m f lange t h i c k n e s s . Welds t h a t a t t a c h a c o n t i n u i t y p l a t e to t h e c o l u m n f lange or web, s h o u l d be m a d e wi th a n e l ec t rode w i t h a r a t e d CVN of a t l ea s t 20 ft.-lb, a t -20 ° F. B a s e d on e x p e r i m e n t a l r e s u l t s , r e m o v a l of b a c k i n g b a r s f rom cont i - n u i t y p l a t e we lds , however , d o e s n o t a p p e a r to be n e c e s s a r y . W h e n w e l d i n g t h e c o n t i n u i t y p l a t e s to t h e c o l u m n , w e l d i n g in t h e "k-area" of t h e c o l u m n s h o u l d be a v o i d e d (AISC 1997}.

All w e l d i n g s h o u l d be spec i f ied to be in c o n f o r m a n c e w i t h t h e l a t e s t ed i t i on of AWS

D 1.1. A c c e p t a n c e c r i te r ia for u l t r a s o n i c tes t - i ng of groove we lds is r e c o m m e n d e d to be in c o n f o r m a n c e w i t h Table 5.2 of AWS D 1.1-98. Add i t i ona l u s e f u l i n f o r m a t i o n on w e l d i n g m o m e n t c o n n e c t i o n s c a n be f o u n d in a n u m - be r of r e f e r e n c e s l i s ted at t h e e n d of t h i s doc- u m e n t .

R e c e n t t e s t s h a v e s h o w n t h a t RBS con- n e c t i o n s w i t h b o l t e d web de ta i l s c a n m e e t t h e r e c o m m e n d e d p la s t i c r o t a t i o n d e m a n d s of FEMA 267 (1995). However , it s h o u l d be n o t e d t h a t a t la rge r o t a t i o n d e m a n d s , t h e bo l t ed de ta i l a p p e a r s to be m o r e s u s c e p t i b l e to f r a c t u r e i n i t i a t i ng n e a r t h e weld a c c e s s hole . Th i s i s s u e is t h e s u b j e c t of f u r t h e r SAC s p o n s o r e d r e s e a r c h . Unt i l m o r e def ini t ive g u i d a n c e is p r o v i d e d in t h e u p c o m i n g SAC Guidelines, t h e e n g i n e e r s h o u l d ca re fu l ly c o n s i d e r r e q u i r e d c o n n e c t i o n a n d SMF per- f o r m a n c e w h e n c h o o s i n g a b e a m web con- nec t ion .

The m a j o r i t y of t h e w e l d e d web c o n n e c - t ion t e s t s h a v e u t i l i zed a c o m p l e t e j o i n t p e n - e t r a t i on (CJP) groove weld b e t w e e n t h e b e a m web a n d c o l u m n f lange over t h e full d e p t h of t he web. The s h e a r tab , w h i c h is w e l d e d to t h e c o l u m n a n d bo l t ed to t h e b e a m web, is still p rov ided . T h i s s h e a r t ab se rves severa l p u r p o s e s . Fi rs t , i t a c t s a s b a c k i n g for t h e C J P groove weld. S e c o n d , it ca r r i e s e r e c t i o n l oads a n d h e l p s m a i n t a i n t h e f r a m e in a p l u m b p o s i t i o n u n t i l w e l d i n g a t t h e c o n n e c - t ion is c o m p l e t e d . S ince t h e s h e a r t ab is p ro- v ided for e r e c t i o n p u r p o s e s only, it is r e c o m - m e n d e d t h a t t h e d e s i g n of t h e s h e a r t ab be left to t h e fabr ica to r . However , to e n s u r e t h a t t h e s h e a r t ab d o e s n o t r e s i s t l oads in t h e even t t h a t excess ive p l a s t i c r o t a t i o n s c a u s e t h e web c o n n e c t i o n to f r ac tu re , t h e d e s i g n e r c o u l d c o n s i d e r i n d i c a t i n g t h a t t h e s h e a r t ab be f a b r i c a t e d w i t h s h o r t h o r i z o n t a l s lo t t ed holes .

T r a d i t i o n a l l y t h e s h e a r t a b w o u l d be w e l d e d on b o t h s ides . However , w h e n uti l iz- i ng a web C J P weld, t h e "~backside" fillet we ld m a y p o s e p o t e n t i a l filler m e t a l m i x i n g a n d fit u p p r o b l e m s . The e n g i n e e r s h o u l d w o r k w i t h t h e f a b r i c a t o r to g e n e r a t e a n a c c e p t a b l e we ld ing s e q u e n c e . As a n a l t e rna t ive to a C J P groove weld, t h e b e a m web c o n n e c t i o n c a n a lso be m a d e u s i n g a h e a v y fillet w e l d e d s h e a r tab . The s h e a r t ab is typ ica l ly w e l d e d

16

DESIGN OF REDUCED BEAM SECTION (RBS) MOMENT FRAME CONNECTIONS

to the co lumn us ing either fillet welds or a CJP groove weld. The shear tab, in turn, is then welded to the beam web with fillet welds. An example of such a connect ion can be found in "Moment Frame Connection Development and Testing for the City of Hope National Medical Center" (Zekioglu, et.al. 1997).

If the engineer chooses to use a bolted web connection, all aspects of the connect ion should be designed to resist the full shear applied to the beam due to gravity and earth- quake loads. Short slotted holes may be uti- lized to futher protect the shear tab and beam web from possz'bie excesive deflections when the connect ion in subjected to large rotat ions as the system undergoes inelastic action dur ing an ear thquake. It should be noted tha t s t ructural steel erectors prefer s tandard holes to slotted holes to aid in erec- tion.

One of the most d iscussed aspects of RBS design, and one of the most important , is the supplementa l lateral bracing required for this system. FEMA 267A (1997) r ecommends tha t a lateral brace be provided near the RBS. The following discussion presents an analysis of test resul ts that did not have lat- eral bracing provided near the RBS.

Virtually all m o m e n t connect ions that dissipate energy by yielding of the beam are subject to varying degrees of beam instability at large levels of inelastic rotation. This is true both for reinforced connect ions (cover plates, ribs, haunches , etc.) and for RBS con- nections. This instabili ty generally involves a combinat ion of flange buckling, web buckl ing and lateral torsional buckling and typically resul ts in deteriorat ion of the beam flexural strength, with increasing inelastic rotations. In the experience of some researchers , the degree of instabili ty and associated s t rength deterioration for RBS connect ions tested in the laboratory have been no more severe, and perhaps somewhat less severe than for many types of reinforced connect ions . This is demons t ra ted by the connect ion test results shown in Figure 4.8.

This figure shows a plot of beam tip load versus beam tip d isp lacement for two differ- ent test specimens. These two spec imens were virtually identical, except for the con-

nect ion detail. Both specimens were con- s t ruc t ed wi th the same m e m b e r sizes (W36xlS0 beam and W14x426 column) and hea ts of steel, and tested in the same test setup with identical member lengths, identi- cal member end support conditions, and identical lateral bracing. Both spec imens were subjected to the same loading history. The only difference was that one specimen was cons t ruc ted with a cover plated connec- tion and the other with an RBS connection. Both spec imens were provided with a single beam lateral support near the point of load application.

250

200

150

100 .

~ 5 0 .

~ o .

.~ -~0. - 1 0 0 ,

- 1 5 0 .

- 2 0 0 ,

-250

-6

Cover'Pla~ed Connectlon ~.______,~_ -~ - -~ - - ,~

RBS Connection ] * ~

\ ' ~ ~ -

_ _ _

- - - -

~ '~"'~'~'({~:;e ~ • I I

.~ -2

~ . ~ . ~ :~-~ :°~* °" ~ ~ '°~ ~

, , Displacement (inches)

Figure 4 .8 Compar i son o f Tes t Resu l t s for

Cover Plated and RBS C o n n e c t i o n s

As can be seen from Figure 4.8, the peak s t rength of the RBS connect ion is less t han that of the cover-plated connection. This, of course, is expected and is in fact a potential advantage of the RBS in that it reduces the m o m e n t genera ted at the connect ion and the m o m e n t delivered to the column. After reach- ing their peak s trength, both connect ions exhibited some s t rength deteriorat ion due to combined flange, web and lateral torsional buckl ing in the beam. Note however tha t the rate of deteriorat ion is less for the RBS spec- imen. In fact, at large inelastic deformations, the RBS exhibits the same s t rength as the cover-plated connect ion. This compar i son demons t ra tes the observation m a d e above, i.e., RBS c o n n e c t i o n s exhibi t no more s t rength deterioration, and pe rhaps some- what less deteriorat ion than reinforced con- nections.

17

DESIGN OF REDUCED BEAM SECTION (RBS) MOMENT FRAME CONNECTIONS

The t e s t d a t a s u m m a r i z e d in A p p e n d i x A i n d i c a t e s t h a t m a n y RBS c o n n e c t i o n t e s t s h a v e b e e n c o n d u c t e d w i t h o u t a n a d d i t i o n a l l a t e ra l b r a c e a t t h e RBS. T h e r e is n o i n s t a n c e w h e r e a n i n v e s t i g a t o r r e p o r t e d u n u s u a l l y severe or u n a c c e p t a b l e s t r e n g t h d e t e r i o r a t i o n d u e to t h e a b s e n c e of a l a te ra l s u p p o r t n e a r t h e RBS. F u t h e r , a s d i s c u s s e d above , s t r e n g t h d e g r a d a t i o n in t h e RBS is c o m p a r a - ble to t h a t s e e n in m a n y o t h e r c o n n e c t i o n t y p e s for w h i c h n o a d d i t i o n a l l a t e ra l b r a c i n g is p r e s e s n t l y r equ i r ed . C o n s e q u e n t l y , b a s e d on c u r r e n t l y ava i lab le da t a , a n a d d i t i o n a l lat- eral b r a c e at t h e RBS d o e s n o t a p p e a r n e c e s - s a r y in o r de r to ach i eve a c c e p t a b l e p e r f o r m - ance . However , t h e d e s i g n e r s h o u l d still a d h e r e to t h e n o r m a l code p r o v i s i o n s for b e a m la te ra l s u p p o r t a n d for b e a m f lange a n d web s l e n d e r n e s s l imits . La te ra l b r a c i n g for b e a m s in Spec ia l M o m e n t F r a m e s s h o u l d be p r o v i d e d a t a m a x i m u m s p a c i n g of 2 5 0 0

/FY, a s r e q u i r e d by Sec t ion 9 .8 of t h e AISC is~nic Provisions ( 1997}. As d e s c r i b e d ear l ier , m o s t m o m e n t con-

n e c t i o n s s h o w g r a d u a l s t r e n g t h d e g r a d a t i o n a t la rge levels of p l a s t i c r o a t a t i o n d u e to com- b i n e d local a n d la te ra l t o r s i o n a l b u c k l i n g of t h e b e a m . Th i s o c c u r s for t h e RBS as well a s for m o s t o t h e r c o n n e c t i o n types , a s i l lus- t r a t e d in F igu re 4.9. R e d u c i n g t h e la te ra l s u p p o r t s p a c i n g in t h e reg ion of t h e p l a s t i c h i n g e f r o m t h a t r e q u i r e d in Sec t ion 9 .8 of t h e AISC Seismic Provisions m a y t h e r e f o r e r e d u c e t h e r a t e of s t r e n g t h d e g r a d a t i o n for m o s t t y p e s of m o m e n t c o n n e c t i o n s . F u r t h e r def in i t ive r e c o m m e n d a t i o n s a n d r e s e a r c h r e s u l t s will be p r o v i d e d in t h e u p c o m i n g SAC Guidelines.

If a d e s i g n e r s h o u l d c h o o s e to p rov ide a l a te ra l b r a c e a t t h e RBS, t h e b r a c e s h o u l d n o t be l oca t ed w i t h i n t h e r e d u c e d s ec t i on of t h e b e a m . Welded or bo l t ed b r ace a t t a c h e - m e n t s in t h i s h i g h l y s t r a i n e d r eg ion of t he b e a m m a y se rve as f r a c t u r e in i t i a t ion si tes. C o n s e q u e n t l y , if a l a te ra l b r ace is p rov ided , it s h o u l d be l oc a t ed a t or b e y o n d t h e e n d of t he RBS t h a t is f a r t h e s t f rom t h e face of t h e col- u m n . If b r a c i n g is to be p r o v i d e d as p a r t of t h e des ign , r e q u i r e m e n t s a n d r e c o m m e n d a - t ions c a n be g a t h e r e d f rom d o c u m e n t s s u c h as FEMA 267A (1997) a n d " F u n d a m e n t a l s of B e a m Brac ing" (Yura 1993).

5 RBS Design Example

Description of Design Example Project

• C o m m e r c i a l Office B u i l d i n g / M e d i c a l Office B u i l d i n g

• L o c a t e d in S a n F ranc i sco , Ca l i forn ia • D i s t a n c e f r o m N e a r e s t E a r t h q u a k e

Fau l t : ~ 9 k i l o m e t e r s (San Andreas ) • High Se i smic i ty Z o n e w i th Near F a u l t

C h a r a c t e r i s t i c s

Description of Design Example Frame

P e r i m e t e r M o m e n t F r a m e s F r a m e c e n t e r l i n e d i m e n s i o n s :

s to ry h e i g h t = 13' - 0" b a y w i d t h = 22 ' - 8"

B e a m : W 2 4 x 1 1 7 A572 Gr. 50 (A992) Fy b = 50 ks i

C o l u m n : W14x311 A572 Gr. 50 (A992) Fy c = 50 ks i

Gravi ty load o n b e a m : (1.2D + .5L p e r Sect . 9 .2c of AISC Seismic Provisions):

2 k i p s / f t (0.17 k i p s / i n )

Gravi ty l oads are d u e to floor t r i b u t a r y l o a d s a s well a s ex te r io r wal l loads .

D e s i g n typ ica l in te r io r m o m e n t c o n n e c t i o n of p e r i m e t e r f rame.

I ~ V l ~ a

R = radius of cut = 4c~+ b ~

8c

_1 - - I b

Figure 5 .1 RBS D i m e n s i o n s

18

DESIGN OF REDUCED BEAM SECTION (RBS) MOMENT FRAME CONNECTIONS

S e c t i o n Proper t i e s : From Equat ion 5:

W 2 4 x 1 1 7 :

d b = 2 4 . 2 6 in. b f = 12 .80 in.

fw = 0 .85 in. = 0 . 5 5 in.

Zxb = 3 2 7 in. 3 W 1 4 x 3 1 1 :

d c = 17 .12 in. bc f = 16 .23 in. t c f = 2 . 2 6 in. t cw = 1.41 in. Zxc = 6 0 3 in. 3

STEP 1 C h o o s e t r i a l v a l u e s for RBS d i m e n - s i o n s a, b a n d c

MRB S = 1.15 ZRBS_Fy e = 1 1 5 x 2 1 8 x 5 5 = 13789 i n - k i p

STEP 5 C o m p u t e t h e s h e a r fo rce a t t h e c e n t e r s of t h e RBS a t e a c h e n d of t h e b e a m

L ' = L - d c - 2 a+ =272-17 .12-2 7+ =222in.

F r o m E q u a t i o n s 6 a a n d 6b:

2Me~ s wL' 2×13789 0.17x222 Vm~ s - - - + - ~ =143kips

L' 2 222 2

a -~'(0.5 to 0.75) b f ~ 6 in. to 10 in. Try: a = 7 in.

b ~ ( 0 . 6 5 to 0.85) d b ~ 16 in. to 21 in. Try: b = 19 in.

c ~ 0 . 2 b f ~ 2 . 6 in. Try: c = 2 . 7 5 in.

STEP 2 C o m p u t e t h e p l a s t i c s e c t i o n m o d u - l u s a t t h e m i n i m u m s e c t i o n of t h e RBS

F r o m E q u a t i o n 3:

ZRB S = Zxb- 2 c t f ( d b - t ~ = 327 - 2 x 2.75 x 0.85 x (24.26 - 0.85) = 218 in.3

STEP 3 E s t a b l i s h t h e e x p e c t e d y ie ld s t r e s s of t h e b e a m

For A572 Gr. 50 s tee l , Ry = 1.1.

F r o m E q u a t i o n 4:

V~ s _ 2M~s wL'_ 2×13789 0.17×222 =105kips L' 2 222 2

F i g u r e 5 .2 s h o w s t h e s h e a r fo rce d i a g r a m , t h e b e n d i n g m o m e n t d i a g r a m , a n d t h e f ree b o d y d i a g r a m t h e for t h e p o r t i o n o f t h e b e a m b e t w e e n R B S c e n t e r s . O b s e r v e t h a t t h e m a x - i m u m m o m e n t o c c u r s a t t h e e n d s , i .e. , a t t h e c e n t e r s of t h e RBS. If t h e g r a v i t y l o a d w e r e e x t r e m e l y la rge , c o m p a r e d to t h e m o m e n t

143 105

V (k i p )

M ( k i p - i n )

13789

-13789

Fy e = RyFy b = 1 . 1 x 5 0 = 5 5 k s i

STEP 4 C o m p u t e t h e m a x i m u m m o m e n t e x p e c t e d a t t h e c e n t e r o f t h e RBS

~ REDS w = 0.17 kips/in. ~ RIBS

Ii . , . l . . i ~ I i ~ I i I I I I ~ t I t i i I I I t I ~ I i i . l . . ! j

. . . . . . . . t J 143 ' "~05k ~

, L' ~ 222 in.

F i g u r e 5 . 2 P o r t i o n o f E x a m p l e B e a m

b e t w e e n R B S C e n t e r s

19

D E S I G N O F R E D U C E D B E A M S E C T I O N (RBS) M O M E N T FRAME CONNECTIONS ,

d e v e l o p e d d u e to a p p l i e d l a te ra l l oads , t h e c u r v e d p o r t i o n of t h e m o m e n t d i a g r a m c o u l d dr ive t h e p las t i c h i n g e t o w a r d t h e c o l u m n , a w a y f rom t h e RBS. Th i s e x a m p l e i n d i c a t e s t h a t t h e gravi ty l oad is n o t la rge e n o u g h to fo rm a p l a s t i c h i n g e w i t h i n t h e s p a n , a w a y f rom t h e RBS. C o n s e q u e n t l y , t h e ca lcu la - t i on s above for t h e m o m e n t a n d s h e a r forces , a t t h e RBS cu t s , a r e valid.

S T E P 6 C o m p u t e t he m a x i m u m m o m e n t e x p e c t e d a t t h e face of t h e c o l u m n

M s

F r o m E q u a t i o n 7:

=Mees + Veas(a + 2b-/= 13789 + 143(7 + ~ ) = 16149in - kip

S T E P 7 C o m p u t e t h e p l a s t i c m o m e n t of t h e b e a m b a s e d on t h e e x p e c t e d yie ld s t r e s s

F r o m E q u a t i o n 8:

Mpe = Zxb Fy e = 327 x 55 = 17985 in -k ip

S T E P 8 C h e c k t h a t Mfis in t h e r a n g e of 85 to 100 p e r c e n t of Mpe

F r o m E q u a t i o n 9:

ZMc > 1.0 (Equa t i on 10)

R e t u r n i n g to t h e e x a m p l e , a s s u m i n g t h a t p o i n t s of in f lec t ion in t h e c o l u m n s o c c u r a t t he i r m i d - h e i g h t s , a n d a s s u m i n g a n axial s t r e s s (fa) of 15 ks i in t h e c o l u m n s u n d e r c o m b i n e d e a r t h q u a k e a n d gravi ty loading , t h e fol lowing c a l c u l a t i o n s resu l t .

F r o m E q u a t i o n s 11, 12, 13 a n d 14:

h~ + d b + h b

2 x 13789+ (143 + 105(17;12 + 7 + ~ )

156 = 217kips

Met

Mcb

= V c h t = 2 1 7 x (156 - 2 4 . 2 6 ) / 2 = 14294 in -k ip

14294 in -k ip

= 2x14294 = 28588 in - k i p

M f 16149 - -

Mpe 17985 - - - 0.90 OK

T h u s , t h e p r e l i m i n a r y d i m e n s i o n s a re OK.

Use: a = 7 i n . b = 1 9 i n . c = 2 .75 in.

S T E P 9 S t r o n g C o l u m n - W e a k B e a m C h e c k

To c h e c k s t r o n g c o l u m n - w e a k b e a m r e q u i r e m e n t s , t h e p r o c e d u r e p r e s e n t e d in FEMA 267A (1997) will be u s e d , w i th t h e m i n o r m o d i f i c a t i o n s n o t e d in Sec t ion 4. The f inal e q u a t i o n to be u s e d to c h e c k t h i s r e q u i r e m e n t ( f rom E q u a t i o n 7 . 5 . 2 . 5 - 1 of FEMA 267A) is as follows:

~Zc(Fyc-.f~) 2×603(50-15) - = 1.5 > 1 OK

~ M ~ 28588

S T E P 10 C h e c k C o l u m n P a n e l Z o n e

To c h e c k t h e c o l u m n p a n e l zone , t h e p ro- c e d u r e d i s c u s s e d in Sec t i on 4 will be u s e d .

B a s e d on t h e e x a m p l e , t h e c o l u m n p a n e l zone s h e a r is c o m p u t e d a s follows:

Mf = 16149 in -k ip (Equa t i on 7)

F r o m E q u a t i o n s 15, 16 a n d 17:

27Mf = Mf+ M:f = 16149 + 15522 = 3 1 6 7 4 in -k ip

, Mf=M~Bs+V~Bs a+ =13789+105 7+ =15522 in - kip

i | 1

2 0

DESIGN OF REDUCED BEAM SECTION (RBS} MOMENT FRAME CONNECTIONS

Vez - 0.8z..,~'Mr 0.8Vc Vc - 0.8x31671 0.8x217 = 926kips 0.95dt) 0.95 × 24.26

Panel zone s t rength is computed as fol- lows:

From Equation 18:

= 0.55F~,~d~tIlL + 3b~ft~d+d~t 1

I 3 x 16"23 x (2"26)~ ] = 0.55xSOx17.12x1.41 1+ 24.26xlT.12xl.41J = 946 kips

946 > 926 .'.No doubler plates required

STEP 11 Check Beam Shear

From Equat ion 19:

w ( l - l ' ) /272~222/ V~ 4 2 0.17 -

' 2 143 ÷ 2

= 145kips

V, = A,,,Fy = (0.55)(24.26)(5 O) = 667 k ips > 145 k ips

RBS flange reduct ion is approximately 43 percent. Consequently, it is expected that the inclusion of tlae RBS the beams will increase interstory drift by about 5 percent.

S ~ e ~ c Abut

,~ ~ . B.U. bar to remain

I / ~ ~ Remove weld tabs IE 718" x 6" ~,.~,,.T-~-~'~"r-.~ / ~ IP {B.S.) ~ I ! [ I / _1 16 ~ Weld B.U. bar Io coiutnn

• ~ l ~ . l I~ . ~ _ 5 . ° - - N ....

~ l / * ~

I.t' i I w2,.,,7 ~i I.I I'\i I g;-~'~-------------~,~,:,,d~,,,~,,~,,oo,~d,

~, tose~v a s b a c i~g C~, ~ - - ~ I Z . . . . ooo,0,.to , . ] ~ , ~ , _ ~ ~ ~ \ , ~ , ~ . ~ : ~ column and beam byfabdcato~.

I I 5/16 \ cleaned and inspected

.

Configure plate comes to \ ~ 17 75" Radius =.o,o0, . . . . . . / . of column Grind Smooth

~ ~ J ~ 1 ~ 2.75" 7.3" 2.75"

5 / ' I ' ~ NI welds: ET0 ~lI groove welds: electrodes must be rat~;I for

'° CVN of at teast 20 It-fos at -20 deg. F. All welding shall conform to AWS D1.1

Figure 5 .3 C o n n e c t i o n Detai l for Des ign E ~ m p l e

6 P r o c e d u r e s f o r A c c e p t a n c e o f D e s i g n b y B u i l d i n g A u t h o r i t i e s

Continuity Plates

Use cont inui ty plates with a th ickness approximately equal to the beam flange thickness . The beam flange th ickness is 0.85 inches. Therefore, use 7 /8" thick cont inui ty plates (0.875"). Connect cont inui ty plates to co lumn flanges us ing CJP groove welds, and the web us ing double fillet welds. The cor- ners of cont inui ty plates should be config- ured to avoid welding into the k-area of the column.

Beam Web Connection

Connec t b e a m web to co lumn flange us ing CJP groove weld over full depth of web (between weld access holes).

A drawing of a generic final connect ion detail is shown in Figure 5.3. The resul t ing frame should be checked for all code speci- fied s t rength and drift limits. Note tha t the

The design of SMF building systems require that the design account for inelastic defor- mat ion d e m a n d s on the connection. The AISC Seismic Provisions for Structural Steel Buildings (1997), Section 9.2, p resen ts the requ i rements for SMF structures . The RBS connec t ion is an opt ion tha t can m e e t requ i rements set by bui lding codes and con- s ensus documents . The following c o m m e n t s are in tended to describe actions that can be followed to help facilitate the permi t t ing process for a SMF building system.

6.1 C o m m u n i c a t i o n

It is r e c o m m e n d e d that early in the process, the Structural Engineer of Record communi - cate with the bui lding official regarding the proposed use and per t inent aspects of the RBS m o m e n t connect ion. The engineer may need to provide background documen ta t ion to the bui lding official if he or she is unfamil- iar with the design and terminology relating

21

DESIGN OF REDUCED BEAM SECTION (RBS} MOMENT FRAME CONNECTIONS

to the design. The use of this d o c u m e n t may aid the bui lding official in u n d e r s t a n d i n g the design intent.

6 . 2 M e t h o d o l o g y

Once the bui lding official u n d e r s t a n d s the design in tent and sys tem behavior, it is impor tan t to clearly state the design method- ology to be used early so tha t any misunder - s tandings can be avoided. This d o c u m e n t presen ts a general design methodology, uti- lizing some simplifying a s s u m p t i o n s and some of the bet ter aspects of m a n y different design methods . There are other ways to design an RBS m o m e n t connect ion and SMF system than tha t represen ted in this docu- ment . If other me thods are utilized, the engi- neer should be sure to clearly indicate the me thod u sed and the impor tan t aspects tha t show design compliance with the governing building code.

Any design methodology utilized should correlate well with other publ i shed methods , test resul ts and research papers. Section 9.2 of the AISC Seismic Provisions require tha t the design be based on qualifying cyclic tests. The table in Appendix A will help to satisfy this r equ i rement for the RBS connect ion. Any significant deviation from es tabl ished methodologies or tests should be justified. It is impor tan t to u n d e r s t a n d tha t m a n y rec- ommenda t ions conta ined in this d o c u m e n t are based on exper imenta l research. Design equat ions and RBS sizing values are based on successful research , both analytically and experimentally. Therefore, any new design equat ions should be comparable to estab- l ished equat ions.

6 .3 C o n s t r u c t i o n D o c u m e n t s

After a design is complete, it is imperative to convey the information accurate ly on con- s t ruct ion documents . While calculat ions are impor tan t and describe the final cons t ruc ted connect ion, cons t ruc t ion documen t s provide direction to the fabricator and erector. The e lements expressed on the drawings will be more impor tan t to the final quality of the design than any calculation.

The documenta t ion related to the RBS connect ion should be clear and concise, yet provide enough detail for the fabricator to properly incorporate all the difficult and impor tan t aspects of the connection. The information should be such tha t any fabrica- tor or erector can utilize the information pro- vided, and cons t ruc t the final connect ion in such a m a n n e r tha t the performance will directly correlate with the design intent.

Impor tan t aspects of the design to be inc luded in the drawing details are welding detai ls , RBS shape a n d locat ion, no tes regarding grinding of the RBS after cutting, shear tab detail information and beam web to co lumn flange connect ion details. It is rec- o m m e n d e d to provide a set of notes specific to the RBS connect ions , relat ing to welding pract ices and connect ion cons t ruc t ion proce- dures to help the contractor u n d e r s t a n d the connect ion and the impor tance it has on the building sys tem performance. Reference to applicable port ions of AWS D I.1 and other AWS or AISC documen t s should be inc luded in these notes to clearly state a level of expected quality. This level of informat ion will also facilitate obtaining the appropria te level of inspect ion required for this type of connection.

7 Fabricat ion and I n s p e c t i o n I s sues

A n u m b e r of fabr icat ion and inspec t ion i ssues are impor tan t to ensu re a well-con- s t ruc ted RBS connect ion. As d i scussed ear- lier proper fabrication and erect ion of this connect ion is a critical port ion of the sys- t em 's pe r fo rmance . If welds are poorly placed, the s t ress at which f racture init iates and propagates is m u c h lower than the stress a tough weld metal , placed with care, can resist. Cut t ing and grinding are critical aspects of fabrication which m u s t be well executed to p roduce a high quality final con- nection.

7.1 C u t t i n g a n d G r i n d i n g

The cut portion of both the curved RBS sec- tion, as well as the prepara t ion of the end of

22

DESIGN OF REDUCED BEAM SECTION (RBS} MOMENT FRAME CONNECTIONS •

the beam, needs to be smooth and free of notches . This smoothness is impor tan t for reasons d i scussed earlier. Many fabrication shops have the ability to make virtually no tch free the rmal cuts. While this is a ben- efit to reduce the n u m b e r of pe rpendicu la r notches , which may presen t s t ress risers, small imperfect ions exist tha t may affect con- nect ion performance.

Therefore, it is impor tan t to clearly iden- tify wha t is the adequa te a m o u n t of mater ia l to remove (by grinding) from the cut surface. FEMA 267A (1997) d i scusses a level of acceptable surface roughnes s value less t han or equal to 1000 as defined in ANSI/ASME B46.1. This level is difficult to de termine wi thout a significant a m o u n t of equ ipment and expertise. Therefore, this d o c u m e n t rec- o m m e n d s tha t the the rmal cuts be ground smooth in the following manne r : "It is impor- t an t tha t the pa t te rn of any cuts m a d e in the flange be propor t ioned so as to avoid sharp cut corners. All c o m e r s should be rounded to minimize no tch effects and in addit ion, cut edges should be cut or g round to have a sur- face roughness meet ing the requ i rements of AWS C4.1-77 class 4, or smoother."

The designer should d i scuss the in tent with the fabricator and develop criteria for an acceptable mock-up to be made for reference du r ing fabr icat ion inspec t ions . The final grinding tha t the engineer and fabricator have agreed upon , shou ld be inspec ted by the fabricator 's representa t ive as well as the owner 's test ing agency, to ensu re compl iance with the accepted mock-up .

Many beams u s e d for SMF sys tems are large with th ick flanges and webs. Shear p u n c h i n g holes in these th ick port ions of the member could lead to localized de laminat ion or tearing. In s i tua t ions where hole diame- ters are smaller t h a n the base mater ia l th ickness , the des igner m a y consider tha t holes required for fabricat ion of e lements and port ions of the RBS beam be drilled ra ther t han punched . No resea rch resul ts indicate tha t a reduc t ion in connect ion per formance is a t t r ibutable to p u n c h i n g holes in RBS beams.

7.2 W e l d i n g

Welding is a very critical par t of the proper fabrication of this connect ion. A significant a m o u n t of effort ha s been made to produce a b e a m wi th a r e d u c e d sec t ion m o d u l u s , de s igned to yield pr ior to developing m o m e n t s which deliver very high s t resses to beam flange - co lumn flange welds. However, if the welding requi red for this connect ion is done poorly, the s t ress at wh ich brittle behavior m a y occur is m u c h lower than the engineer expects. Good welds, us ing tough filler metal , will resis t h igher loads than sur- round ing base metal . Therefore, it is impera- tive tha t the welding for this type of connec- t ion be of h igh qual i ty , to p r o d u c e a connec t ion tha t will perform as designed.

Any specific i s sues re la ted to welds, such as weld profiles, s equence , submi t t a l of mater ia ls or cert if ications tha t are consid- ered impor tan t for compl iance of the fabrica- tor's work to mee t the design intent , should be clearly s ta ted in the cons t ruc t ion docu- ments . I tems s u c h as p rehea t shou ld be addres sed in the project specifications and cons t ruc t ion drawings. Typically, AWS will adequate ly addres s mos t i ssues , and for new design will provide the fabr icator ample direction to complete the cons t ruc t ion in a safe and high quali ty m a n n e r .

The engineer shou ld be clear in the proj- ect specifications a n d cons t ruc t ion drawings tha t filler meta l s shal l no t be mixed in s u c h a way as to p roduce a CVN value below tha t specified a n d requ i red for a single filler metal. Most fabricat ion shops present ly use gas shielded FCAW m e t h o d s for welds to co lumns a n d beams . The erect ion crews, especially w h e n welding complete joint pene- t ra t ion groove welds , typical ly u s e self shielded FCAW. Also, the re are different filler metals u sed for the flat posit ion as well as other positions. Some combina t ions of filler meta ls in the same jo in t m a y p roduce a com- bined CVN value, wh ich could p resen t "brit- fie behavior". The engineer should carefully review the in format ion provided in "The Effects of In termixed Weld Metal on Mechan- ical Properties" (1998) and the submi t t ed WPS prior to fabr icat ion to ensu re tha t the fabricator and erector are not creat ing a

23

DESIGN O F REDUCED BEAM SECTION (RBS} MOMENT FRAME CONNECTIONS

potential problem by inappropriately mixing filler metals.

Parameters should be set for quali ty con- trol of shop welding and fabrication. The fab- ricator m u s t have an acceptable Quali ty Con- trol (QC) procedure in place th roughout the fabrication of the project. In addition, Quality Assurance measures should be t aken to help ensure tha t the QC procedure is being imple- mented and followed. Typically QA or Verifi- cat ion Inspect ion is provided by special inspectors , hired by the owner. It is the responsibi l i ty of the engineer to es tabl ish inspect ion protocol, request a pre-fabrication and pre-erection meeting, and impress upon the fabricator and erector the impor tan t

i s sues s u r r o u n d i n g the RBS connec t ion details and construction. Complete joint pen- etrat ion groove welds should be inspected by a Level II qualified NDT inspector as defined in the AWS D 1.1. Each joint should be ultra- sonically tested and all welds associated with the connect ion should receive cont inuous special inspection. Field inspect ion should be sensitive to such i ssues as weld preparat ion and fi t-up, weld profile and weld p a s s sequence, back-up bar removal and grinding of run-of f tabs . The inspec to r s shou ld develop an acceptable protocol for inspect ion and reports in regards to welding and con- nect ion completion.

24

DESIGN OF REDUCED BEAM SECTION (RBS) MOMENT FRAME CONNECTIONS

References

"AISC Initiates Research Into k Area Crack- ing," Modern Steel Construction, Vol. 37, No. 9, September 1997, pp.23-24.

Grubbs, K.V., "The Effect of the Dogbone Connection on the Elastic Stiffness of Steel Moment Frames," M.S. Thesis, Department of Civil Engineering, the Uni- versity of Texas at Austin, Austin, Texas, August 1997.

Blodgett, O., Funderburk, S., and Miller, D., "Fabricators ' and Erectors ' Guide to Welded Steel Construction," The Lincoln Electric Company, Cleveland, 1997.

International Conference of Building Officials (ICBO), The Uniform Building Code (UBSC), April 1997.

Chen, S.J., Yeh, C.H. and Chu, J.M, "Ductile Steel Beam-to-Column Connections for Seismic Resistance," Journal of Structural Engineering, Vol. 122, No. 11, November 1996, pp. 1292-1299.

Iwankiw, N., "Ultimate Strength Considera- tions of Seismic Design of the Reduced Beam Section (Internal Plastic Hinge)," Engineering Journal , American Institute of Steel Construction, Inc., Vol. 34, No. 1, First Quarter 1997.

Engelhardt, M.D. and Husain, A.S., "Cyclic Loading Performance Of Welded Flange - Bolted Web Connections," Journal o f Structural Engineering, ASCE, Vol. 119, No. 12, December 1993.

Engelhardt, M.D., Winneberger, T., Zekany, A.J. ,and Potyraj, T., ~The Dogbone Con- nection: Part II." Modern Steel Construc- tion, August 1996.

Engelhardt, M.D., Winneberger, T., Zekany, A.J. ,and Potyraj, T., "Experimental Inves- tigation of Dogbone Moment Connec- tions," Proceedings: 1997 National Steel Construction Conference, American Insti- tute of Steel Construction, Chicago, May 1997.

Johnson, M., Quintana, M., '~The Effects of Intermixed Weld Metal on Mechanical Properties, Part III," Proceedings, Interna- tional Conference on Welded Construc- tions in Seismic Areas, AWS, October 1998.

Kaufmann, E., Xue, M., Lu, L., and Fisher, J. , "Achieving Ductile Behavior of Moment Connections," Modern Steel Con- struction, Vol. 36, No. 1, American Insti- tute of Steel Construction, J anua ry 1996.

Lee, K., Goel, S.C., Stojadinovic, B., "Bound- ary Effects in Welded Steel Moment Con- nections," Research Report No. UMCEE 97-20, December 1997.

Engelhardt, M.D. and Sabol, T.A., "Reinforc- ing of Steel Moment Connections with Cover Plates: Benefits and Limitations," Engineering Structures, Vol. 20, No. 6, pp. 510-520, 1998.

Noel, S. N., "Reduced Beam Section Design for Seismic Retrofit of Steel Moment Frame Connections," M.S. Thesis, Divi- sion of Structural Engineering, University of California, San Diego, 1997.

Gross, J., Engelhardt, M., Uang, C., Kasai, K., and Iwankiw, N., "Modification of Existing Steel Welded Moment Frame Connect ions for Seismic Resistance," Steel Design Guide Series Twelve, Ameri- can Institute of Steel Construction, Inc., Chicago, 1999.

Plumier, A., "The Dogbone: Back to the Future," Engineering Journal, American Institute of Steel Construction, Inc. 2nd Quarter 1997.

25

DESIGN O F REDUCED BEAM SECTION (RBS} MOMENT FRAME CONNECTIONS

Popov, E. and Stephen, R., "Cyclic Loading of Full Size Steel Connections," Bulletin No. 21, American Iron and Steel Institute, 1972.

SAC Joint Venture, Background Reports on Metallurgy, Fracture Mechanics, Welding, Moment Connections and Frame Systems Behavior, Published by the Federal Emer- gency Management Agency, Report FEMA 288, 1996.

SAC Joint Venture, Interim Guidelines: Eval- uation, Repair, Modification and Design of Welded Steel Moment Frame Structures, Published by the Federal Emergency Management Agency, Report FEMA 267, August 1995.

SAC Joint Venture, Interim Guidelines Advi- sory No. 1 - Supplement to FEMA 267, Published by the Federal Emergency Management Agency, Report FEMA 267A, March 1997.

Seismic Provisions for Structural Steel Build- ings, American Institute of Steel Con- struction, Inc., Chicago, April 15, 1997.

"Structural Welding Code - Steel," AWS D 1.1- 98, American Welding Society, Miami, 1998.

Tide, R., "Stability of Weld Metal Subjected to Cyclic Static and Seismic Loading," Engi- neering Structures, Vol. 20, Nos. 4-6, April-June 1998.

Tsal, K.C. and Popov, E.P., "Steel Beam-Col- umn Joints In Seismic Moment Resisting Frames", Report No. UCB/EERC - 88/19, Earthquake Engineering Research Cen- ter, University of California at Berkeley, 1988.

Yura, J.A., "Fundamentals of Beam Bracing," Proceedings, Structural Stability Research Council Conference, "Is Your Structure Suitably Braced?," 1993.

Zekioglu, A., Mozaffarian, H. and Uang, C., "Moment Frame Connection Development and Testing for the City of Hope National Medical Center," Proceedings; Structures Congress XV, Portland, April 13-16, 1997, American Society of Civil Engineers, 1997.

26

APPENDIX A

Summary of Experiments on Reduced Beam Section Moment Connections for New Construction

Ref

[1]

[1]

[1]

[1]

[1]

Spec.

YC-1

YC-2

PC-1

PC-2

PC-3

Beam

Built-up W shape d=23.6", b~=l 1.8", tf=0.79", tw=0.47"

Lb=73" A36 steel Fy_f =40 ksi Fo.~ =66 ksi Fy.w =40 ksi Fu.w =65 ksi

Column

Built-up Box: 19.7"xl 9.7"x.79"

Lc = 87" A572 Gr. 50

Fy =56 ksi Fu =82 ksi

Flange Welds

SS-FCAW E70T-7

No weld tabs used

Web Connection

Bolted: 7-7/8" A325

RBS Details and Other

Flange Modifications

Tapered cut L1=2"

LRBS=I 3.8" FR=20%

Tapered cut L~=2"

LRBS=17.7" FR=25%

Tapered cut L1=4.7"

LRBS=I 5.7" FR=34%

Tapered cut L1=4.7"

LRSS = 17.7" FR=42%

Tapered cut L1=4.7"

LRss=I 7.7" FR=42%

Op (%)

2.4

2.9

4.1

4.8

3.8

Comments

Fracture of beam flange initiating at weld access hole

Fracture of beam flange initiating at weld access hole

Fracture of beam flange initiating at weld access hole

Fracture of beam flange initiating at weld access hole

Fracture of beam flange initiating at weld access hole

I m~

Ref

[2]

[2]

[2]

[2]

[3,4]

[3,4]

Spec.

DBT- 1A-99-

176

Beam

W30x99 A572 Gr. 50

L~= 138"

Column

W14x176 A572 Gr. 50

Lc=168"

Flange Welds

SS-FCAW E70TG-K2;

backing bar removed

Web Connection

Bolted: 7-1" A325

RBS Details and Other

Flange Modifications

Tapered cut L1=7.5"

LRBS=20.25 ''

DBT- 1 B-99-

176

DBT- 2A-150-

257

DBT- 2B-150-

257

ARUP- 1

Fy.w = 61.6 ksi Fu.w = 82.8 ksi

W30x99 A572 Gr. 50

Lb=138" Fy. w = 51.5 ksi Fu.w = 72.1 ksi

W36x150 A572 Gr. 50

Lb=138" F~.w = 60.2 ksi Fu.w = 72.3 ksi

W36x150 A572 Gr. 50

Lb=138" Fy.w = 62.9 ksi Fu.w = 83.1 ksi

W36x150 A572 Gr. 50

Lb=132"

Fy.w =55.6 ksi Fu.w =70.7 ksi

W14x176 A572 Gr. 50

Lc=168" Fy.w =55.5 ksi Fu.w =71.8 ksi

W14x257 A572 Gr. 50

Lc=168" Fy.w =59.6 ksi Fu.w =75.2 ksi

W 14x257 A572 Gr. 50

Lc=168" Fy.w =64.5 ksi Fu.w =83.2 ksi

W 14x426 A572 Gr. 50

Lc=136"

at bottom flange

SS-FCAW E70TG-K2

backing bar left in

Bolted: 9-1" A325

welded (heavy shear tab groove

FR=45%

Tapered cut L1=7.5"

LRBS=20.25 " FR=45%

Tapered cut L1=9"

LaBs=24" FR=45%

Tapered cut L1=9"

LRBS=24 '' FR=45%

Tapered cut L1 =9"

LABS=24"

COH-1

Fy.f =55.5 ksi Fu4 =73 ksi

Fy.w =62.5 ksi Fu-w =77 ksi

W27x178 A572 Gr. 50

Lb= 132" Fy.f =44 ksi Fu.f =62 ksi Fy.w =46 ksi Fu-w =62 ksi

W 14x455 A572 Gr. 50

Lc=136" Fy.f =55 ksi Fu4=84 ksi Fy.w =54 ksi Fu-w =86 ksi

place w/seal weld at top flange;

backing bar removed at bottom flange

welded to column and fillet welded

to beam web)

FR=44% top & bottom

flanges reinforced with

vertical ribs Tapered cut

L~=7" LABS=20" FR=38%

top & bottom flanges

reinforced with vertical ribs

0p (%)

2.8

4.0

Comments

no failure; test stopped due to limitations in test setup

no failure; test stopped due to limitations in test setup

3.5 ' Fracture of beam top flange near groove we d

1.7 Fracture of beam top flange we d; propagated to divot- type fracture of column flange

3.5 Flange fracture at minimum section of RBS

3.5

A-2

[3,4]

[3,4]

[3,4]

[3,4]

COH-4 ~¢ =~

COH-5 |~

[5,6]

[5,6]

Spec. Beam Column Flange Welds Web Connection

RBS Details and Other

Flange Modifications

COH-2 (~ =¢ ~

COH-3 Wl 4x455 A572 Gr. 50

Lc=136" Fy.f =55 ksi Fu.f =84 ksi Fyow =54 ksi Fu-w =86 ks i

Beam connected to column web

W33x152 A572 Gr. 50

Lb=132" Fy.f =57.6 ksi Fu.f =78.5 ksi Fy.w =62 ksi

Fu-w =84.5 ksi

Tapered cut L1=9"

LRBS=26" FR=43%

top & bottom flanges

reinforced with vertical side

plates

Ref

DB1 Wl 4x426 A572 Gr. 50

Lc=136"

W 14x426 A572 Gr. 50

Lc=136" Fy.f =50 ksi

Fu4 =74.5 ksi Fy.w =50 ksi Fu.w =75 ksi

W33x152 A572 Gr. 50

Lb=132" F~4 =62.8 ksi Fu.f =86 ksi

F~.w =69.1 ksi Fu.w =93.7 ksi

SS-FCAW E71T-8

backing bar left in place w/seal weld at

top flange; backing bar removed

at bottom flange

W36x160 L~=134"

Fy.f =54.7 ksi Fu4 =75.6 ksi Fy.w =53.5 ksi Fu-w =79.2 ksi

welded (beam web

W36x150 Lb=134"

Fy.f =41.4 ksi Fu4=58.7 ksi Fy.w =47.1 ksi Fu-w =61.8 ksi

DB2

Constant cut L1=9"

groove welded to column)

LRBS=I 9.5" FR=40%

Radius cut L1=9"

L~Bs=27" FR=40%

Gp Comments (O/o)

3.8

3.2

4.0

1.8

2.0 Flange fracture at RBS

3.0 Testing stopped due" to limitations of test setup

A-3

Ref

[5,6]

[5,6]

[5,6]

[7]

Spec.

DB3

DB4

DB5

DB1

Beam

W36x170 L~=134"

Fy.f =58 ksi Fu.f =73 ksi

Fy,w =58.5 ksi Fu.w =76.7 ksi

W36x194 Lb=134"

Fy.f =38.5 ksi Fu4 =58.6 ksi Fy,w =43.6 ksi Fu.w =59.8 ksi

W30x148 Lb=134"

Fy.f =46.6 ksi Fu.f =64.5 ksi Fy.w =48.5 ksi Fu.w =65.4 ksi

W36x135 A36 Steel Lb=134.5"

Column

W 14x426 A572 Gr. 50

Lc=136"

W 14x426 A572 Gr. 50

Lc=136" Fy4=50 ksi

Fu4 =74.5 ksi Fy,w =50 ksi Fu.w =75 ksi W 14x257

A572 Gr. 50 Lc=136"

Fy.f =48.7 ksi Fu.f =69 ksi

Fy. w =49.4 ksi Fu.w =66.2 ksi

W 14x257 with 1-5/16" thk.

cover plates (cover plates welded

across flanges of W14x257 to form

box) A572 Gr, 50

L~=132"

Flange Welds

SS-FCAW E71T-8

(details of backing and weld tabs not

available)

Web Connection

Not Available

RBS Details and Other

Flange Modifications

Radius cut L1=9"

LRBS=27 '' FR=40%

Radius cut L1=9"

LRBS=27 " FR=38%

Radius cut L1 =5"

LRas=25 " FR=38%

Radius cut L1=8"

LRBS=28 '' FR=40%

~p (%)

3.8

3.7

4.0

3.0

Comments

Testing stopped due to limitations of test setup; significant column panel zone yielding

Testing stopped due to limitations of test setup

A-4

Ref

[8]

[8]

[8]

[8]

[8]

Spec. Beam Column

S-1

S-2A

SC-1

S-3

S-4

W530x82 (Canadian Designation)

d=20.8", bf=8.2", tf=0.52", tw=0.37"

wt.=54 Ib/ft. Lb= 142"

CSA G40.41-350W steel

Fy.f =52.4 ksi Fo.f =76.6 ksi Fy.w =57.5 ksi

Fu.w =81 ksi (~

W 14x 120 A572 Gr. 50

Lc=120"

Flange Welds

SS-FCAW E71T-8

backing bar left in place w/seal weld at

top flange; backing bar removed

at bottom flange

Web Connection

Bolted: 5-1" A325

RBS Details and Other

Flange Modifications

Radius cut L1=4.7"

LRss=l 5.7" FR=55%

0p (%)

9.0

3.6

3.4

note (8)

note (9)

Comments

Specimen loaded monotonically; testing stopped due to limitations of test setup

Testing stopped due to limitations of test setup Composite slab included (6); testing stopped due to limitations of test setup statically applied simulated earthquake loading (7); testing stopped due to reaching end of simulated earthquake loading; no connection failure dynamically applied simulated earthquake loading (7); testing stopped due to reaching end of simulated earthquake loading; no connection failure

A-5

Ref

[8]

[11]

[11]

[11]

[11]

[12]

[12]

Spec.

SC-2

LS-1

Beam Column

W30x99 A572 Gr. 50

W14x176 A572 Gr. 50

Flange Welds

SS-FCAW E70T-6

Web Connection

welded (Beam web

RBS Details and Other

Flange Modifications

Radius cut L1 = 7"

LS-2

LS-3

LS-4

DBBW Beam 1

Lb = 141" Fy.f = 54.0 ksi Fu4= 71.9 ksi Fy.w = 58.0 ksi Fu.w = 74.8 ksi

W36x150 A572 Gr. 50

Lb = 141"

Lc = 150" Fy.f= 55.5 ksi Fu4 = 74.0 ksi Fy.w= 54.0 ksi Fu.w= 73.1 ksi

(~

W 14x398 A572 Gr. 50

Lc = 146"

backing bar left in place w/seal weld at

top flange; backing bar removed

at bottom flange ~

SS-FCAW E70T-6

backing bar left in

groove welded to column)

Bolted: 10 - 1" A490

LaB s = 20" FR = 50%

Radius cut L1 = 9"

LaBS = 27" FR = 50%

DBBW

Beam 2 m

Fy.f = 54.3 ksi Fo.f = 68.8 ksi Fy.w = 59.4 ksi Fu.w= 72.0 ksi

Fy = 53.0 ksi Fu = 73.0 ksi

(based on CMTR)

place w/seal weld at top flange;

backing bar removed at bottom flange

.

0p (%)

Note (9)

Comments

Composite slab included (6); dynamically applied simulated earthquake loading (6); testing stopped due to reaching end of simulated earthquake loading; no connection failure

4.0 No connection failure

+1.0 note (12) /-5.0 -1.0/ note (12) +5.0

4.0 No connection failure; testing stopped due to limitations of test setup

4.0 No connection failure; test stopped due to limitations of test setup;

see note (13)

4.0

A-6

Ref

[12]

[12]

[13]

[13]

[13]

[13]

Spec.

DBBW- C

Beam 1 DBBW-

C

Beam 2

DBWW

Beam 1

DBWW

Beam 2 DBWW

-C

Beam 1 DBWW

-C

Beam 2

Beam Column Flange Welds Web Connection

W36x150 A572 Gr. 50

Lb= 141" Fy.f= 54.3 ksi Fu.f = 68.8 ksi Fy.w = 59.4 ksi Fu.w= 72.0 ksi

¢¢

W 14x398 A572 Gr. 50

Lc = 144" F v = 53.0 ksi Fu = 73.0 ksi

(based on CMTR)

SS-FCAW E70T-6

backing bar left in place w/seal weld at

top flange; backing bar removed

at bottom flange ( (

welded (Beam web

groove welded to column)

RBS Details and Other

Flange Modifications

Op (%)

5.0

3.8

3.5

Comments

Low cycle fatigue fracture in RBS;

see note (14) Fracture of bottom beam flange adjacent to groove weld; fracture initiated at weld access hole;

see note (14) No connection failure; test stopped due to limitations of test setup

see note (13)

3.5

5.0 Low cycle fatigue

5.0

fracture in RBS

see note (14) Low cycle fatigue fracture in RBS

A-7

Ref Spec.

[14] WG-1

[14] WG-2

[14] WG-3

[14j

Notes:

Beam

W33x201 A572 Gr. 50 Lb = 160.5"

F~.f = 52.0 ksi Fu-f = 72.8 ksi Fy.w = 51.5 ksi Fu-w = 68.0 ksi

W36x300 A572 Gr. 50

Lb = 159" F~.f = 56.0 ksi Fu4 = 72.9 ksi Fy.w = 56.7 ksi Fu.w = 74.5 ksi

WG-4 "

Column

W14x311 A913 Gr. 65

Lc = 152" Fy.f = 69.0 ksi Fu4 = 88.3 ksi Fy-w = 68.0 ksi F..w= 86.5 ksi

5/8" doubler plates (A572 Gr. 50)

provided on each side of column web

W14x550 A913 Gr. 65

Lc = 152" Fy.f = 67.0 ksi Fu4= 86.8 ksi Fy.w= 68.1 ksi Fu.w = 87.6 ksi

Flange Welds

SS-FCAW E70TG-K2;

backing bar removed at bottom flange

Web Connection

Bolted: 13-1" A490

Bolted: 20 - 1" A490 (2 rows of 10 bolts each)

RBS Details and Other

Flange Modifications

Radius cut L1 = 9.3"

LRBS = 25" FR = 54%

Radius cut L1 = 10"

Lass = 27" FR = 51%

~p (%)

2.9

2.9

3.5

Comments

fracture of RBS at local buckle in RBS

see note (15)

No connection failure; test stopped due to limitations of test setup

1~

" 4.5 "

1. All specimens are single cantilever type, except DBBW, DBBW-C, DBWW, and DBWW-C 2. All specimens are bare steel, except SC-1, SC-2, DBBW-C and DBWW-C 3. All specimens subject to quasi static cyclic loading, with ATC-24, SAC or similar loading protocol, except S-1, S-3, So4, SC-2, LS-2 and LS-3 4. All specimens provided with continuity plates at beam-to-column connection, except Popov Specimen DB1 (Popov Specimen DB1 was provided with

external flange plates welded to column). 5. Specimens ARUP-1, COH-1 to COH-5, S-1, S-2A, S-3, S-4, SC-1, SC-2 and LS-4 provided with lateral brace near loading point and an additional

lateral brace near RBS; all other specimens provided with lateral brace at loading point only. 6. Composite slab details for Specimens SC-2 and SC-2:118" wide floor slab; 3" ribbed deck (ribs perpendicular to beam) with 2.5" ~oncrete cover;

normal wt. concrete; welded wire mesh reinforcement; 3.4" dia. shear studs spaced at 24" (one stud in every other rib); first stud located at 29" from face of column; 1" gap left between face of column and slab to minimize composite action.

A-8

7. Specimens S-3, S-4 and SC-2 were subjected to simulated earthquake loading based on N10E horizontal component of the Llolleo record from the 1985 Chile Earthquake. For Specimen S-3, simulated loading was applied statically. For Specimen S-4 and SC-2; simulated loading was applied dynamically, and repeated three times.

8. Specimen S-3: Connection sustained static simulated earthquake loading without failure. Maximum plastic rotation demand on specimen was approximately 2%.

9. Specimens S-4 and SC-2: Connection sustained dynamic simulated earthquake loading without failure. Maximum plastic rotation demand on specimen was approximately 2%.

10. Tests conducted by Plumier not included in Table. Specimens consisted of HE 260A beams (equivalent to W10x49) and HE 300B columns (equivalent to W12x79). All specimens were provided with constant cut RBS. Beams attached to columns using fillet welds on beam flanges and web, or using a bolted end plate. Details available in Refs. 9 and 10.

11. Shaking table tests were conducted by Chen, Yeh and Chu [1] on a 0.4 scale single story moment frame with RBS connections. Frame sustained numerous earthquake records without fracture at beam-to-column connections.

12. Specimens LS-2 and LS-3 were tested using near field loading protocol. The specimen was subjected to peak pulses corresponding to 6% story drift ratio. Loading was repeated six times for LS-2 and four times for LS-3. The specimens eventually failed due to low cycle fatigue fracture at the narrowest section in the RBS.

13. Specimens DBBW and DBWW were cruciform t~,pe specimens with beams attached to each column flange. 14. Specimens DBBW-C and DBWW-C were cruciform type specimens with composite floor slab. Composite slab details:

96" wide slab; 2" ribbed metal deck (ribs parallel to beam) with 3.5" topping of normal weight concrete; concrete compressive strength at time of testing = 3600 psi for DBBW-C and 6800 psi for DBWW-C; slab reinforced with #4 Gr. 60 bars in each direction; 3.4" dia. shear studs spaced at 12"; first stud located at 36" from face of column (at end of RBS).

15. Specimens WG-1 to WG-4: Test report provided slightly conflicting data on location along length of beam where displacement was measured. Values of plastic rotation reported above are based on an estimated location for displacement measurements.

A-9

Notation: Fy.f = flange yield stress from coupon tests Fu_f = flange ultimate stress from coupon tests Fy_w = web yield stress from coupon tests Fu-w = web ultimate stress from coupon tests Lb = Length of beam, measured from load application point to face of column Lo = Length of column L~ = distance from face of column to start of RBS cut EBBS = length of RBS cut FR = Flange Reduction = (area of flange removed/original flange area) xl00

(Flange Reduction reported at narrowest section of RBS) ep = Maximum plastic rotation developed for at least one full cycle of loading, measured with respect to the face of the column (based on occurrence

of fracture or based on end of loading)

References: [1] Chen, S.J., Yeh, C.H. and Chu, J.M, "Ductile Steel Beam-to-Column Connections for Seismic Resistance," Journalof Structural Engineering, Vol.

122, No. 11, November 1996, pp. 1292-1299. [2] Iwankiw, N.R., and Carter, C., "The Dogbone: A New Idea to Chew On," Modern Steel Construction, April 1996. [3] Zekioglu, A., Mozaffarian, H., and Uang, C.M., "Moment Frame Connection Development and Testing for the City of Hope National Medical Center,"

Building to Last- Proceedings of Structures Congress XV, ASCE, Portland, April 1997. [4] Zekioglu, A., Mozaffarian, H., Chang, K.L., Uang, C.M. and Noel, S., "Designing After Northridge," Modem Steel Construction, March 1997. [5] Engelhardt, M.D., Winneberger, T., Zekany, A.J. and Potyraj, T.J., "Experimental Investigation of Dogbone Moment Connections," Proceedings; 1997

National Steel Construction Conference, American Institute of Steel Construction, May 7-9, 1997, Chicago. [6] Engelhardt, M.D., Winneberger, T., Zekany, A.J. and Potyraj, T.J., "The Dogbone Connection, Part II, Modem Steel Construction, August 1996. [7] Popov, E.P., Yang, T.S. and Chang, S.P., "Design of Steel MRF Connections Before and After 1994 Northridge Earthquake," International

Conference on Advances in Steel Structures, Hong Kong, December 11-14, 1996. Also in: Engineering Structures, 20(12), 1030-1038, 1998. [8] Tremblay, R., Tchebotarev, N. and Filiatrault, A., "Seismic Performance of RBS Connections for Steel Moment Resisting Frames: Influence of

Loading Rate and Floor Slab," Proceedings, Stessa '97, August 4-7, 1997, Kyoto, Japan. [9] Plumier, A., "New Idea for Safe Structures in Seismic Zones," IABSE Symposium - Mixed Structures Including New Materials, Brussels, 1990. [10] Plumier, A., "The Dogbone: Back to the Future," Engineering Journal, American Institute of Steel Construction, Inc. 2nd Quarter 1997. [11] Uang, C.M., Unpublished preliminary test reports for SAC Phase 2 RBS tests, University of California at San Diego, December 1998 and February

1999. [12] Engelhardt, M.D. and Venti, M., Unpublished preliminary test reports for SAC Phase 2 tests, University of Texas at Austin, 1999. " [13] Fry, G., Unpublished preliminary test reports for SAC Phase 2 tests, Texas A & M University, 1999. [14] Unpublished report of connection proof tests for building construction project in southern California; project title withheld at request of building owner,

January, 1999.

A-10

June 2002

Use of Deep Columns In

Special Steel Moment Frames

By

Jie-Hua Jay Shen, Ph.D., P.E., S.E. Associate Professor

Department of Civil and Architectural Engineering Illinois Institute of Technology

Abolhassan Astaneh-Asl, Ph.D., P.E.

Professor Department of Civil and Environmental Engineering

University of California, Berkeley

David B. McCallen, Ph.D. Director

Center for Complex Distributed Systems Lawrence Livermore National Laboratory

____________________________________________________________________________

(A copy of this report can be downloaded free of charge for personal use from www.aisc.org)

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 1

Use of Deep Columns in Special Steel Moment Frames By Jie-Hua Jay Shen, Abolhassan Astaneh-Asl and David B. McCallen This report discusses some of the issues related to the use of “deep columns” in special moment frames. Since 1994 Northridge earthquake significant amount of research and development projects have been done in U.S., Japan and elsewhere on seismic behavior and design of steel moment frames. In almost all of these research projects, the column used in testing or analyses have been W14 or smaller sections. One of the most important research projects during this period was the SAC Steel joint Venture Project where a large number of moment connections were tested and analyzed and design recommendations were formulated. In this project, almost all specimens had a column with depth of no more than 14-16 inches. However, since in many cases of moment frames, the governing design requirement is the stiffness to control the drift, the use of deep columns with a depth of 24, 27 and even 30 inches, becomes very economical. Unfortunately, there is no extensive and reliable information on actual cyclic behavior and design of moment frames with deep columns. This report discusses: (a) the issues that need to be considered in using deep columns in moment frames, (b) a comparison of seismic behavior of two 10 story moment frames designed using W14 and W27 respectively, (c) the results of a series of realistic non-linear finite element analysis of moment-rotation behavior of connections with deep columns and; (d) the conclusions. First Printing, June 2002. __________________________________________________________________________________ Jie-Hua Jay Shen, Ph.D., P.E., S.E. Associate Professor, Department of Civil and Architectural Engineering, Illinois Institute of Technology, 3201 South Dearborne Street, Chicago, IL, 60616. Phone: (312) 567-5860, Fax: (312) 567-3579. E-mail: shen@iit.edu. ____________________________________________________________________________________________ Abolhassan Astaneh-Asl, Ph.D., P.E., Professor, 781 Davis Hall, Univ. of California, Berkeley, CA 94720-1710, Phone: (510) 642-4528, Fax: (925) 946-0903, E-mail: astaneh@ce.berkeley.edu , Web page: www.ce.berkeley.edu/~astaneh ____________________________________________________________________________________________ David B. McCallen, Ph.D., Director, Center for Complex Distributed Systems, Lawrence Livermore National Laboratory, 7000 East Avenue, MS L-151, Livermore, CA 94550. Phone: (925) 423-1219 E-mail: mccallen2@llnl.gov. ____________________________________________________________________________________________ Disclaimer: The information presented in this publication has been prepared in accordance with recognized engineering principles and is for general information only. While it is believed to be accurate, this information should not be used or relied upon for any specific application without competent professional examination and verification of its accuracy, suitability, and applicability by a licensed professional engineer, designer or architect. The publication of the material contained herein is not intended as a representation or warranty on the part of the Structural Steel Educational Council or of any other person named herein, that this information is suitable for any general or particular use or of freedom from infringement of any patent or patents. Anyone making use of this information assumes all liability arising from such use. Caution must be exercised when relying upon specifications and codes developed by others and incorporated by reference herein since such material may be modified or amended from time to time subsequent to the printing of this document. The Structural Steel Educational Council or the authors bears no responsibility for such material other than to refer to it and incorporate it by reference at the time of the initial publication of this document.

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 2

ACKNOWLEDGMENTS The publication of this report was made possible in part by the support of the Structural Steel Educational Council (SSEC). The authors wish to thank all SSEC members for their valuable comments. Particularly, special thanks are due to Fred Boettler, Jeff Eandi, Lanny Flynn, Pat Hassett, William Honeck, Brett Manning and James Putkey for their valuable and detailed review comments. The authors also appreciate the review comments provided by James Malley of Degenkolb Engineers and Dr. Farzad Naeim of John A. Martin Associates. The opinions expressed in this report are solely those of the authors and do not necessarily reflect the views of the Illinois Institute of Technology, the University of California Berkeley, the Lawrence Livermore National Laboratory where authors are employed nor the Structural Steel Educational Council or other agencies and individuals whose names appear in this document. A portion of this work was performed at Lawrence Livermore National Laboratory under the auspices of DOE Contract W-7405-Eng-48. The analyses and design of the 10-story frames were done using the latest version of the SAP-2000n program. The generous donation of the program by Computers and Structures Inc. of Berkeley (www.csiberkeley.com) is sincerely appreciated. The finite element analyses of connections were conducted using ABAQUAS and NIKE-3D program.

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 3

USE OF DEEP COLUMNS IN SPECIAL STEEL MOMENT FRAMES By: JAY SHEN, Ph.D., P.E., S.E. Associate Professor Department of Civil and Architectural Engineering, Illinois Institute of Technology , Chicago ABOLHASSAN ASTANEH-ASL, Ph.D., P.E. Professor Department of Civil and Environmental Engineering, University of California, Berkeley DAVID B. McCALLEN, Ph.D. Director Center for Complex Distributed Systems, Lawrence Livermore National Laboratory, Livermore

____________________________________________

TABLE OF CONTENTS

ABSTRACT / Page 1 ACKNOWLEDGMENTS / Page 2 TABLE OF CONTENTS / Page 3

NOTATIONS / Page 4 CHAPTER 1. INTRODUCTION / Page 5 CHAPTER 2. USE AND BEHAVIOR OF FRAMES WITH DEEP COLUMNS / Page 8 CHAPTER 3. ANALYSIS OF CYCLIC BEHAVIOR OF DEEP COLUMN CONNECTIONS / PAGE 17 CHAPTER 4. CONCLUSIONS / Page 33 REFERENCES/Page 36

ABOUT THE AUTHORS / Page 38 LIST OF PUBLISHED “STEEL TIPS” REPORTS / Page 39

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 4

_________________________________________________________________________

Notations _________________________________________________________________________

In preparing the following notations, whenever possible, the definitions are taken from various references as indicated inside the parentheses whenever applicable. bf Width of flange. E Modulus of elasticity. Fy Specified minimum yield stress of the type of steel to be used, ksi. As used in the LRFD

Specification, "yield stress" denotes either the minimum specified yield point (for those steels that have a yield point) or the specified yield strength (for those steels that do not have yield point). (AISC, 1997).

Fyw Specified minimum yield stress of the web. h Depth of web. J Torsion constant, cross section property. in. Inch, 1 inch= 25.4mm. Ix Moment of inertia about x-axis. Iy Moment of inertia about y-axis. ksi Kilo-pounds per square inches, 1 ksi=6,895 kilo-Pascal. rx Radius of gyration about x-axis. ry Radius of gyration about y-axis. Sx Section modulus about x-axis. Sy Section modulus about y-axis. tf Thickness of flange. tw Thickness web. Zx Plastic modulus about x-axis. Zy Plastic modulus about y-axis. λp Limiting slenderness parameter for a compact element. (AISC, 1997). λr Limiting slenderness parameter for a non-compact element. (AISC, 1997). λf Equals bf /2tf for flange. λw Equals h/tw for web. θc Twisting of column.

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 5

1. Introduction 1.1. Introduction Moment-resisting frames are one of the frequently used lateral load resisting systems in many steel building structures. During the 1994 Northridge earthquake, a large number of welded steel moment frames developed cracks in their beam-to-column welds at or near joints. Although, none of the damaged structures developed any partial collapse or even injuries, the structural engineering and steel construction community undertook an extensive effort to study the phenomenon and mitigate it. In the aftermath of the 1994 Northridge earthquake and during 1994-2000 periods, a comprehensive research and technology development project was undertaken by SAC Steel Joint Venture (FEMA-350, 2001) primarily funded by the Federal Emergency Management Agency to address this problem. The main goal of the project, sometimes denoted as simply the SAC project, was to develop technologies for design, construction, inspection, evaluation and retrofit of the moment frames subjected to seismic effects.

As part of the SAC Project, a large number of cyclic tests of beam-to-column connections of moment frames were conducted. The aim was to establish the actual behavior of existing as well as the improved beam-to-column moment connections. Most of these tests were done on specimens where the columns were W14 sections with a maximum depth of column being about 14-16 inches. When the studies were completed, SAC Project produced a set of reports (FEMA-35, 2001) on various aspects of the problem and its solutions. One of the important items in the FEMA reports was the introduction of “pre-qualified” moment connections. The pre-qualified connections have specific ranges of material properties and geometry, which are based on tested connections. It is expected that if properties of a designed connection fall within these ranges, the designed connection will behave in a manner similar to those tested within the SAC Program.

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 6

Almost all the pre-qualified connections in SAC reports have a W14 column traditionally used in many structures. However, in today’s design offices, structural engineers in many projects find it more economical to use columns that are deeper than the W14 sections. In recent years, it has been recognized that there is a strong economic incentive for the design engineer to use deep columns to satisfy increasingly more stringent drift limitations. Using W14 columns to satisfy drift limitations specified by the codes often results in unnecessarily heavy columns. Structural engineers have, from time to time, used deeper columns for some steel building projects, when they had resources to carry out the physical tests of project-based connections. The deep columns would be more extensively used for moderate-rise to high-rise buildings if the time consuming and costly physical tests could be avoided. So far, limited research has been done regarding the behavior and design of a beam-to-column connection with deep columns. Two reports (Gilton et al, 2000) and (Ricles et al., 2000) include the results of cyclic testing of a few beam-to-column connection specimens where the column was a deep wide flange section. Therefore, there is a need for information on the performance of beam-to-column moment connections with deep columns. A deep column in this context is a column with a depth of greater than 21 inches.

1.2. Background on This Study After the 1994 Northridge earthquake, extensive studies were conducted to improve the performance of the steel moment-resisting frame when subjected to strong ground motions. Since then, the Reduced Beam Section (RBS), where a portion of the beam flange is removed in order to force the plastic hinge in the beam away from the column face, has become one of the frequently used welded moment connections. Researchers have studied the behavior of the RBS connections when connected to W14 columns (FEMA-350, 2001), and have found that the connections with RBS have larger cyclic rotational ductility than the same connections without RBS. This type of beam-to-column connection assembly has been pre-qualified by FEMA-350 for seismic design of moment-resisting frames along with a number of other configurations of welded and bolted connections

In 2000, a report by Gilton et al. (2000) presented the results of cyclic tests of three RBS moment connections where deep columns were used. The authors have reported twisting of the deep columns. Although the twisting of deep column in their tests appears to have been observed during the late stages of loading and after rotations in excess of 0.03 radian, the authors have expressed concern about twisting of the deep columns and have formulated and proposed limitations on the geometry of the column cross section to prevent the observed twisting of deep columns. A review of the report by Gilton et al (2000) indicates that the lateral movement of RBS hinge and the resulting twisting of deep columns in their tests may have been due to

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 7

unrealistic boundary conditions and lack of bracing normally provided to top flange by the floor beams.

To investigate this, non-linear cyclic behavior of RBS moment connections with W14 and deep columns were studied and the results are summarized here. The analyses began with building the model of a beam-to-column sub-assemblage that had been physically tested (Gilton et. al., 2000). After the results of a tested specimen was well simulated by a finite element model, a group of more realistic beam-to-column sub-assemblages with other deep column configurations were analyzed, and the results were evaluated. The results confirmed that indeed column twisting in Gilton et al. (2000) tests might have occurred primarily because of the way the specimen was tested. In these tests, there was no flange bracing which normally is provided to the top flange of the beam by the floors in actual buildings.

The authors hope the information presented here can be useful in better understanding the actual behavior of moment connections with deep columns in buildings. In addition, we hope the information can assist future researchers in planning their test set-up to test moment connections with deep columns in a realistic and proper manner.

1.3. Objectives of this Report

The main objectives of this Steel Technical Information and Product Services (Steel TIPS) report are:

1. To review the use of frames with deep columns (Section 2).

2. To conduct pushover and inelastic time history analyses of frames with W14 as well as deep columns and compare their seismic behavior (Section 2).

3. To conduct a critical review of the results of a few cyclic tests available at this time on “deep columns”. The deep columns are defined as columns with a depth of 21 inches or greater, particularly columns with 24, 27, 30 and 33 inch depths (Section 3).

4. Using realistic models of the connections with deep columns, to conduct simulated cyclic tests of these connections and compare the results of computer analyses to actual test results to ensure that the computer analyses predict the actual test results well (Section 3).

5. To conduct more analyses of moment connections with different beam and deep columns sections and with floors being present or not (Section 3).

6. To formulate tentative recommendations for the use of deep columns in moment frames. Such recommendations can be verified by selective, well-planned and correctly executed testing (Section 4).

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 8

2. USE AND BEHAVIOR OF FRAMES WITH DEEP COLUMNS

2.1. Introduction

In most cases of design of moment frames, drift limitations, and not strength, govern the design. One of the efficient ways of reducing the drift of a moment frame is to increase the bending and shear stiffness of its columns. Using deeper cross sections than the W14’s traditionally used in many moment frames will accomplish this. The following text provides a discussion of the issues related to the use of deep columns.

2.2. Issues Related to the Use of Deep Columns

2.2.a. Stiffness of the Moment Frame

Deep columns with W21 to W30 sections provide larger moment of inertia for the same weight compared to traditional W14 column sections. For example, the weight/ft of a W27 section will be less than ½ of the weight/ft of a W14 section with comparable moment of inertia. Relatively large bending stiffness of the deep columns results in increasing the global stiffness of the moment frame, which in turn results in reducing the drift and damage.

2.2.b. Strength

In moment frames subjected to relatively large lateral forces, bending strength of the columns is one of the important parameters. Deep columns provide larger plastic moment capacity than the equivalent W14’s, making it possible to more easily meet the strong column-weak beam design requirements. For example, the weight/ft of a W27 section will be less than 70% of the weight of a W14 section having the same plastic moment capacity. In using deep columns with relatively small weak axis moments of inertia, one has to check the possibility of lateral torsional buckling of the deep column, especially for tall floors.

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 9

yww FEth /45.2/ ≤

ywyp FErL /7.1=

ywy FErL /7.1/ ≤

According to AISC Specification (AISC, 2001), if un-braced length of compression flange of a beam in bending is less than the Lp given by the following Equation 2.2, lateral-torsional buckling is not expected to occur before the beam reaches its plastic moment capacity.

If L ≤ Lp the beam is compact for lateral-torsional buckling, where:

(AISC-LRFD Manual, 2001, P. 16.1-33) (2.1)

By rearranging the above equation we can obtain a limit for L/ry of the column, Equation 2.2, that below this limit lateral-torsional buckling is not expected and need not be checked.

(2.2)

For A36, Grade 50 and Grade 65 steel, the above limit of L/ry is equal to 48, 41 and 36 respectively.

2.2.c. Panel Zone Issues

Deep column sections have deeper webs than the W14 columns and provide more web area than the W14 for the same weight. This means that shear strength and stiffness of the panel zone in a deep column is greater than the corresponding values in a W14 column with the same weight. The larger shear strength of the panel zone in deep columns can help reduce the need for doubler plates. The larger shear stiffness of the panel zone in deep columns can help reduce panel zone distortions. As a result, the contribution of panel zone distortions to the story drift can be smaller when deep columns are used. In deep columns, where the web is relatively slender, shear buckling of panel zone should be investigated. Shear buckling of web can be avoided by limiting the h/tw of the column web to the following value from the AISC Specification (AISC, 2001).

(AISC-LRFD Manual, 2001, P. 16.1-35) (2.3)

If h/tw of the column web satisfies the above equation, it is expected that the column web can reach shear yielding before buckling. The term on the right side of the Equation 2.3 above for A36, grade 50 and Grade 65 steel (Fy=36, 50 and 65 ksi) is equal to 69, 59 and 52

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 10

respectively. A check on currently available rolled shapes indicate that all rolled wide flange shapes tabulated in the first part of the current AISC-LRFD Manual (AISC, 2001) have h/tw less than 59 therefore satisfy the limit of Equation 2.3 above for A36 and grade 50 steels. For grade 65 steel, with the exception of a few sections, almost all rolled shapes have h/tw less than 52 satisfying the limit of Equation 2.3.

2.2.d. Local Buckling

As far as local buckling is concerned, deep columns have a disadvantage compared to W14 columns. In general, b/t ratio of flanges and h/tw of webs of deep columns are larger than the W14’s with the same weight. However, most deep column sections with grade 50 steel have compact webs and flanges and can be used in high seismic areas.

2.3. Comparison of Behavior of a Frame with W14 and Deep Columns

In order to identify benefits and limitations of using deep columns in moment frames, a limited comparative study was done. In the study, a typical building was selected and was designed using W14 columns. Then, the same building was designed using W27 columns. Both frames had the same girders. The results of analyses of these two frames indicated that in all respects, the frames behaved similarly. However, the weight of the frame with W27 columns was considerably less. Of course, one should not generalize the outcome of this one case of comparison, but as an example, it sheds some light on seismic behavior of similar frames with W14 and W27 columns. In addition, it shows the extent of saving in the weight of columns for this building if one uses deep columns.

2.3.a. Building Used in the Comparative Studies

The building selected for the comparative study was a 10-story perimeter frame building. This building structure, using W14 columns, was almost the same as the structure of a 10-story “study” building designed by the SAC Joint Venture (SAC, 1996) and provided to researchers in 1996. For these studies, the building was assumed located in seismic areas of California within a 10 km distance of a major fault. Hayward fault ground motions were the used in the nonlinear time history analyses. SAC designed the study buildings to comply with the UBC-97 (ICBO, 1997). Figures 2.1 shows framing plan and elevation of the 10-story study structure.

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 11

2.3.b. Design of the Building Used in the Comparative Studies

As indicated earlier, the building used in the study was adapted from one of the study buildings that was developed and used in the SAC Joint Venture program (SAC, 1996). The ten-story building designed by SAC for a Los Angeles site had W14 columns. The SAC-designed structure complied with the UBC-97 and its maximum inter-story drift (for 18 feet tall ground floor, see Figure 2.1) was 1.7%, which is less than the 2% limit given by the UBC-97 for this structure. The frame on column line 6 of SAC structure was selected as one of our two study frames and was denoted as “W14 Study Frame”. Then, we replaced the W14 columns with W27 columns while keeping the same beams and denoted this frame “W27 Study Frame”. Since in moment frames, usually drift is the governing design parameter, the replacement W27 were selected such that the frame had still a drift value less than 2% and both W14 and W27 study frames had comparable stress level in their members. Figure 2.2 shows cross sections of the girders and columns used in both frames. Figure 2.3 shows Demand/Capacity ratios for members of study frames. Instead of LRFD methods, in the design we used AISC-ASD design option of the SAP2000n software and the nominal loads. This was done to be able to compare the stresses and deformations generated in each frame by the combined design forces at service load level and not at factored-load levels. The use of ASD methods here is not to advocate its use in design, which is best done using LRFD methods. To the authors, the ASD method provided a better feel about service level (unfactored) stresses and deformations in the frames.

Figure 2.1. Plan and Elevation Views of the 10-Story Structure

18 ft

8 @ 13 ft

12 ft

ELEVATION PLAN

30 ft

30 ft

30 ft

30 ft

30 ft

30 ft 30 ft 5 6

A

B

C

D

E

F

30 ft 30 ft 30 ft 2 3 4 1

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 12

The analysis of the frame with W27 columns showed that the maximum inter-story drift in the frame was 1.2% and for the frame with W14 columns was 1.7%. Both drift values were less than the limit of 2% as per UBC-97 (IBC, 1997) and occurred at the 18 feet tall ground floor. Figure 2.3 shows values from the interaction equation for the two frames, which indicates the stress level at code service level forces to be similar in both frames and relatively low as expected in a moment frame.

Figure 2.2. Girders and Columns of W14, and W27 Study Frames

W14 Study Frame

W27 Study Frame

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 13

Figure 2.3. Values of ASD M-P Interaction Equation for the W14 Study-Frame

(upper frame) and the W27 Study Frame (lower frame)

W14 Columns

W27 Columns

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 14

Push-over Analyses:

In order to compare the performance of two frames, using the SAP 2000n program, pushover analyses of the frames shown in Figure 2.2 were conducted. In the pushover analyses, both frames were subjected to ever-increasing first mode pushover displacements. Figure 2.4 shows the push over curves. Both frames were able to reach a roof displacement of about 2.5 feet before collapse. Figure 2.5 shows the hinges at the time of collapse. The frame with W14 columns showed soft story formation while the frame with W27 columns had more yielding in the columns at the time of collapse. The columns in the frame with W27 columns were considerably lighter than the columns in the frame with W14 columns.

Figure 2.4. Pushover Curves for the Frames with W14 and W27 Columns

Roof Displacement, ft.

Base Shear, kips.

0

1000

3000

1.0 2.0 3.0

2000

W27

W14

Roof Disp.

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 15

Note: Indicates a plastic hinge with partial yielding

Indicates a plastic hinge with full yielding of the cross section

Figure 2.5. Hinges in the Frames Just Prior to Collapse

Inelastic Time History Analyses:

In order to compare the dynamic response of two frames, inelastic time history analyses of both frames were conducted. The dead and live load as well as the mass applied to both frames were the same as given by SAC (FEMA-350, 2001). The inelastic models of the frames shown in Figure 2.2 were subjected to the E-W acceleration component of the Hayward Seismic Evaluation Earthquake (SEE) generated by Bolt and Gregor (1993). Figure 2.6 shows the time history of displacement of the first floor for the two frames. The drift values for the first floor can be obtained by dividing displacements by 18 feet, the height of ground floor. The inter-story drift of the frames with W27 and W14 columns were 1% and 1.2% respectively. The drift values calculated using UBC-97 (ICBO, 1997) provisions were 1.2% and 1.7% for frames with W14 and W27 columns respectively. Plastic hinges formed in both frames at the RBS areas. However, since the girders in both frames were the same, it was not expected that non-linear behavior of frames would be much different.

In previous sections, it was shown that the drift values and stresses in two study frames, one with W14 columns and the other with W27 columns, were essentially the same. However, for

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 16

this 10-story building with a 150ft by 150ft plan, the weight of the steel using W27 deep columns was about 1.3 lbs/ft2 less than the steel in the same frame but with W14 columns. According to a leading steel fabricator, the 1.3 lbs/ft2 equals to about 6-8% in total material saving based on 16-18 psf of steel for a typical structure of this type. Of course as mentioned earlier, this 10-story building was just an example to demonstrate that using deep columns instead of W14 can result in improvement in lateral load resisting behavior, much better drift and damage control as well as possible savings in the cost of construction of steel frames.

Figure 2.6. Time History of Horizontal Displacement of First Floor to Hayward SEE Earthquake

W14 Frame

W27 Frame

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 17

3. ANALYSIS OF CYCLIC BEHAVIOR OF DEEP COLUMN

CONNECTIONS 3.1. Introduction

This Chapter investigates, analytically, the cyclic behavior of beam-to-column connections with deep column sections ranging from W14 to W33. A compact beam section was used for most of parametric studies, since; almost all available wide flange sections are compact. For comparison, a non-compact section beam was also included. Detailed nonlinear finite element analyses were conducted to address the issues that influence the cyclic performance and design considerations of one of the most commonly used connections pre-qualified by FEMA-350 (2001), namely the RBS connection, whit the column becoming deeper and deeper. In the following sections, a summary of the results of these studies is presented.

3.2. Simulation of Cyclic Behavior of Tested Specimen

3.2.a Computer Model of Test Specimen

As indicated in previous chapter, two of the three specimens tested by Gilton et al. (2000) had web doubler plates added to the column panel zone. The third specimen without the doubler plate, assumed to more realistically represent the current design practice, was therefore selected to be modeled and analyzed in this study. This specimen was Specimen DC-2 (Gilton et. al., 2000). A nonlinear finite element model of this specimen was constructed with the nonlinear finite element program, ABAQUS (ABAQUS, 2001). The specimen was a standard beam-to-column assembly consisting of a W27×194 column and a W36×150 beam, both specified as A572 Gr.50 steel. A reduced beam section (RBS) was introduced to make the beam side of the connection pre-qualified by FEMA 350 (FEMA 350, 2001). The details of the RBS, the column

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 18

stiffeners and web shear tab plate are shown in Figure 3.1. The test setup of the beam-to-column assembly connection is shown in Figure 3.2.

Figure 3.1. Non-Linear Computer Model of the Specimen

Figure 3.2. Model of Test Set-up Used by Gilton et al. (2000)

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 19

The computer model, denoted here as ABQ-DEEP, used fully integrated six-node and eight-node three-dimensional solid elements (Element types C3D6 and C3D8 in ABAQUS).

A finer mesh was used in the RBS area, panel zone and shear tab plate areas. Rigid links were used to connect the beam tip to the actual loading point (reference node), which was also

restrained to prevent out-of-plane translation (Figure 3.2). The material properties of the steel, yield strength and ultimate strength, were specified from the mill certified coupon test of the

Specimen DC-2 (see Table 3.1). Stress-strain curve for the steel was a tri-liner curve with three segments: (a) first segment, (the elastic segment) from the origin to the yield point, (b) the second segment from the yield point to ultimate strength point with stress equal to Fu and strain of 0.20;

and (c) the last segment, a horizontal line at stress level of Fu.

Table 3.1. Properties of Specimen DC-2 Tested by Gilton et al., (2000)

Cyclic loading pattern in the test, controlled by the displacement at the tip of the beam, was of a standard small-to-large displacement cycles as shown in Figure 3.3. At small displacements, the cycles were repeated four times. At larger inelastic displacements, the cycles were repeated twice.

Figure 3.3. Loading History Used in the Test and Analysis

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 20

3.2.b. Simulated Cyclic Behavior of Connection

When simulated cyclic loading was applied to the nonlinear model of specimen, the specimen remained virtually elastic before 1% drift cycles, when some yielding was observed. Though such elastic deformation cycles might be desirable for physical testing, a finite element analysis does not record any effects of elastic cyclic loading and unloading on the assembly. Thus, in the simulation analysis, the cyclic loading history for the analysis started from the cycles immediately before any yielding was observed. The number of inelastic cycles appears to have a significant influence on the post-buckling behavior in terms of strength degradation. The actual test of specimen DC-2 indicated that strength was reduced considerably when the inelastic cycle was repeated. Such cycle-related strength reduction became more significant when a larger inelastic cycle was repeated, apparently due to the Bauschinger effect leading to local buckling and low cycle fatigue phenomenon.

3.2.3. Comparison of Analytic and Experimental Results

Figure 3.4 shows the load-displacement curves from the test specimen DC-2 tested by Gilton et al. (2000), and from the analysis discussed here. The overall cyclic responses from the analysis and the test match reasonably well. There are some noticeable discrepancies in unloading and reloading regions, particularly at large inelastic deformation levels. The unloading curve of the tested specimen was highly nonlinear, significantly different from the linear unloading curve conventionally used as analytical models of hysteretic behavior. The reloading in an opposite direction after a full inelastic unloading made the specimen softer. The softening in unloading and reloading appear to have been responsible for an accelerated strength reduction from its peak value after each cycle with the same or higher level of displacement.

The deformed shapes of specimen from analysis model at 5% drift level are presented in Figures 3.5, showing an isometric view of the buckling shape near the beam-to-column joint. The deformed shape is similar to the final buckling shape observed in the test (Gilton et. al., 2000), especially large deformations in the RBS area. Figure 3.6 shows top and end views of the deformed specimen at 5% story drift.

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 21

Figure 3.4. Load-Displacement Curve of Specimen DC-2 and Analytical Results

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 22

Figure 3.5 Buckling Shape of the Specimen Model at 5% Story Drift

Figure 3.6. Deformed Shape of Web and Flanges at 5% Story Drift

3.3. Parametric Study of Cyclic Behavior of Deep Column Connections

Having successfully simulated the cyclic behavior of the tested specimen, the ABAQUS model, ABQ-DEEP as the prototype, was used to model the connection assembly with various column sizes. In the seismic design of steel moment-resisting frames based on improved connection details summarized in recent FEMA publications (FEMA-350, 2001), there are some concerns related to the connection strength reduction after its peak strength is reached. Slower reduction might indicate a more stable connection performance, and vice versa. It has been

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 23

observed that strength reduction after the peak strength is reached heavily depends on the number of inelastic cycles. The main goals of the parametric studies were:

1. To investigate whether or not there are any significant characteristics in a connection with deep column sections that are not considered in current design practice;

2. To investigate the effects of floor slab and transverse beams in bracing the connection and preventing lateral movement of hinge areas.

Six beam-to-column connection assemblies were studied analytically. Five of them had the columns listed in Table 3.2, and the W36x150 beam section. The five columns were selected to construct the connection assemblies within a practical range. The column sections were selected based on their plastic section modulus (Zx) and moments of inertia (Ix and Iy), so that the comparison could be made with respect to lateral movement of the hinge areas and twisting of columns with different combinations of Zx, Ix, and Iy.

Table 3.2. Section Properties of the Studied Column Sections

In addition, the effect of lateral bracing on the connection assembly performance was also investigated by introducing actual lateral supports from transverse beams and the concrete with metal deck floor that exists in almost all steel framed buildings. To study bracing effects of the floor slab, in some analytical cases, the beam was laterally braced along the beam top flange outside the RBS. Two different boundary condition cases were considered: (1) Unbraced case – where the beam had no lateral restraints similar to specimens tested by Gilton et al (2000); and (2)

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 24

Braced case – where the beam was laterally restrained in its panel zone and top flange except in the RBS region. For comparison, an additional beam-to-column connection with a non-compact beam section, W30x90, and a W27x194 column, was also included in this study. The cyclic analyses applied a maximum displacement of 6% story drift ratio in the same manner as conducting a physical test per FEMA-350 (2001). The following sections will present a summary of the analytical results together with discussions of various issues.

3.3.1. Overall Cyclic Behavior of Deep Column Connections

Figures 3.7, 3.8, and 3.9 show the cyclic behavior of the connection assemblies with W30x191, W33x169, and W201x201, respectively. The cyclic loops of the connections demonstrated that the connections with deeper columns were stable. With lateral bracing (the solid-blue lines in the figures), the connections did not have any significant strength reduction before the 4% drift ratio. Under the cyclic loading, the strength degradation occurred upon the

Figure 3.7. Cyclic Behavior of the Connection with W30x191 Column and

W36x150 Beam.

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 25

load reversal in both positive and negative deformation regions after the plastic hinge formed in the RBS region at about 3% drift ratio, mainly due to inelastic local web and flange buckling. Without lateral bracing (the dashed-red lines in Figures 3.7, 3.8, and 3.9), the connections experienced column twisting and beam lateral torsional buckling after 4% drift ratio, demonstrating a larger strength reduction than those with lateral bracing.

It seems apparent that the lateral supports to the beam flange under compression improved the inelastic behavior of connections with deep columns. In particular, the post-buckling strength degradation was reduced considerably by lateral supports provided by the floor, as shown in Figures 3.7, 3.8, and 3.9. The lateral supports to the beam prevented lateral movement of plastic hinge area and extended the deformation prior to the onset of strength degradation. The local buckling of the flanges and web was mainly responsible for a slow degradation in strength at a later deformation stage for the braced connections. A larger strength degradation under negative bending moment, when the beam top flange was in tension, in the above figures indicates that extra lateral supports to the bottom flange can help to enhance inelastic cyclic behavior. Note that all cases involved a compact beam section, W36x150 with Fy=50 ksi. If any non-compact beam section were used, the strength degradation would have been more significant, as discussed later.

Figure 3.8. Cyclic Behavior of the Connection with W33x169 Column and W36x150 Beam

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 26

3.3.2. Effect of Column Size/Depth

Figure 3.10(a) shows the plan views of deformed RBS connections, with no floor

slab and transverse beams present, at a relatively large story drift ratio of 6%. The large story drift was selected to show the deformations at very late stages of cyclic behavior and at drift values much beyond what can be expected in major seismic event. The figure shows RBS connections with deep columns where no lateral bracing was provided in order to reveal the effect of the column size on the lateral stability of the connection assembly. The larger lateral torsional deformation of the beam was observed when the column was weaker in out-of-plane stiffness. For example, there was no lateral torsional buckling of the same beam when the column was changed to a W14x426.

It seems that in this case, due to lack of floor slab and transverse beams, the deep column was the only element responsible to resist the torque applied to it by the beam. Being subjected to such twisting effects, the deep columns with no floor underwent twisting as shown in Figure 3.10(a) for four study cases. The values of θc given in Figure 3.10 are approximate values of column twisting alone in degrees.

Figure 3.9. Cyclic Behavior of the Connection with W33x201 Column and W36x150 Beam.

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 27

Figure 3.10. Lateral Deformation of the RBS Area and Column Deformations for:

(a) Connections with no Floor Slab and Transverse Beam; and (b) Connections with Floor Slabs and Transverse Beams

Figure 3.10(b) shows the same four connections as in Figure 3.10(a) but this time the

connections have floor slab attached to the top flange of the beam at shear stud locations and a transverse beam is attached to the panel zone of the column. As the figure indicates, by having the floor slab and transverse beam, the column twisting was negligible.

As can be seen in Table 3.2, the torsional stiffness and weak-axis flexural stiffness of a W14 sections are greater than the corresponding values for deeper columns with comparable strong axis flexural stiffness. When a beam-column connection specimen is tested with no slab and transverse beam, there is no lateral restraint to prevent lateral movement of the highly yielded and locally buckled RBS hinge as shown in Figure 3.10(a). When the hinge area, not attached to the floor, moves laterally, it can apply large enough moment to “bare” column to twist it as shown in Figure 3.11.

W27x194 Column W30x191 Column W33x169 Column W33x201 Column

Note: All Beams: W36x150)

W27x194 Column W30x191 Column W33x169 Column W33x201 Column

Note: (All Beams: W36x150) , (a)

(b)

c ≅ 0.0 c ≅ 0.0 c ≅ 0.0 c ≅ 0.0

c ≅ 1.5° c .≅ 2.5° c ≅ 3.0° c ≅ 2.5°

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 28

Figure 3.11. Torque acting on Column due to Lateral Movement of RBS

We believe that the lack of floor slab in Gilton et al’s (2000) tests is the main reason for

development of column twisting in their tests. Had the floor slab been present, as is the case in almost all buildings, or at least the restraining effects of floor slab been represented by bracing in the test set-up, most likely the twisting of columns would have been minor and non-consequential. It is strongly recommended that in future tests of beam-column connections particularly RBS connections with deep columns, the restraining effects of the floor be represented either by having the actual floor cast with the specimen or by attaching to top flange appropriate bracing mechanisms to represent the floors.

3.3.3. Effect of Beam Section Compactness

It is necessary to use a compact beam section in the earthquake-resistant moment frame to ensure a stable cyclic performance during a strong earthquake. The limit of bf/2tf ratio for a compact flange, λp, is equal to 52 ⁄ √(Fy). In practice, most wide flange sections are compact sections. In this study, all previous discussions have been based on a compact beam section, W36x150 (λf = bf/2tf = 6.4; λf/λp=0.87). In this section, a non-compact section, W30x90 (bf/2tf = 8.5; λf/λp=1.16), was selected to compare the behavior of the deep-column connection assembly with compact and non-compact beam sections. For definitions of terms, see “Notations” in Page 4. Figure 3.12 shows the cyclic response of the assembly with W30x90 beam and W27x194 column. The strength reduction rates are 35% and 50% at 4% and 5% story drift levels, respectively, which are twice as much as those observed from previous analyses based on W36x150 beam. An early local buckling of the flanges, as well as the lateral torsional buckling might be responsible for such accelerated strength degradation. Figure 3.13 and 3.14 present the buckling shape of the assembly at 5% story drift level. It is apparent that the buckling of the flange is much more extensive with a non-compact flange than the compact one. However, even in the case with a non-compact beam, after considerable local buckling and distortion of the RBS hinge, the column did not develop twisting.

Torque=(Flange Force)x( Eccentricity.)

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 29

Figure 3.12. Load-displacement curve of the assembly with W30x90 beam and W27x194 column

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 30

Figure 3.14 Buckling shape of the assembly with W30x90 beam and W27x194 column (the top flange view).

Note: 1 kN= 0.225 kips, 1mm=0.0394 inch.

Figure 3.15. Cyclic Behavior of Connection with W14x426 Column and W36x150 Beam

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 31

3.3.4. Lateral Stability of the Connection with W14 Column

For comparison with deep-column connections, an RBS connection assembly with W14x426 column and W36x150 beam was used. Four cases were investigated. The first case, named as ABQ-Fu, involved no RBS. Other three cases involved RBSs with different eccentricities and flange reduction rates. The eccentricity is measured from the column flange face to the near end of the RBS, and the flange reduction rate is the ratio of the cut flange area of the smallest RBS to the original flange area. Figure 3.14 shows analytical and experimental responses of the assembly with ABQ-e1 RBS. There was practically no strength reduction visible from the load-displacement curve. The deformed shapes of the four cases are given in Figure 3.15. There is no lateral torsional buckling in all but one case. The case with a large eccentricity RBS suffered lateral torsional buckling primarily due to a distant RBS from the column. In none of the cases, there was any torsion or weak-axis flexural deformation visible in the column.

Figure 3.16. Deformed Shapes of Connections with W14x425 Column and W36x150 Beam: (a)

No RBS; (b) Small eccentricity and moderate flange reduction RBS; (c) Large eccentricity and moderate flange reduction RBS; and (d) Moderate eccentricity and large flange reduction RBS

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 32

In order to compare behavior of connections with W14 and deep columns, a connection with W14x426 also was analyzed. The beam at this connection was the same as the others, a W36x150. Figure 3.15 shows cyclic moment-rotation behavior of this connection established by non-linear finite element analysis. The connection was analyzed with and without the bracing provided by floor slab. In addition, a third case was also analyzed where the beam did not have the RBS. The analyses indicated that in this case, presence or absence of floor slab did not make much difference. The RBS area of the beam did not move laterally and the column did not show tendency to twist as shown in Figure 3.16. It appears that in this case, the W14 column alone, because of its large stiffness in torsion and lateral bending, was able to brace the RBS hinge and prevent its lateral movement. This may be the reason why in more than 100 tests of connections conducted within the SAC Program, and almost all were without the slab, very few specimens showed tendency for column twisting. As a result, the SAC tests using W14 columns, by default, ended up being valid tests even though there was no floor to brace the beam. Simply put, the column alone provided the bracing. However, in case of connections with deep columns, the columns were not able to provide the bracing that the floor normally provides. As a result, the RBS area of these specimens moved in lateral direction causing twisting of column making these tests somewhat unrealistic and the results questionable. Based on studies summarized in previous sections, it can be concluded that the twisting of the deep columns during the tests conducted by Gilton et al (2000) most likely was the result of the way the tests were done rather than a realistic behavioral phenomenon. The test specimens did not have the lateral bracing provided by the floors that exists in almost all steel structures. Had Gilton, Chi and Uang (Gilton et al, 2000) done the tests with correct boundary conditions and representative bracings, the results would have been realistic representation of actual condition in the field and most likely the twisting of deep columns would have been negligible and non-consequential to the behavior and design. This was clearly the case with tests done by Ricles, Mau, Lu and Fisher (Ricles et al, 2000), where the boundary conditions in the test set-up were correctly presented. No twisting of deep columns were reported for deep column specimens tested by Ricles et al.

Currently, a series of cyclic tests on RBS moment connections with deep columns is in progress at Lehigh University by Professor Ricles and his research team. The results of such tests, expected to be done properly as the earlier tests at Lehigh (Ricles et al, 2000) and the design recommendations stemming from such results, will be a valuable addition to the field.

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 33

4. CONCLUSIONS 4.1. Introduction Based on the results of non-linear analyses of steel moment frames with RBS connections and with W14 through W33 columns, the following conclusions were reached. The conclusions herein should not be used or relied upon for any specific application without competent professional examination and verification of its accuracy, suitability, and applicability by a licensed professional engineer, designer or architect. As indicated in the “Disclaimer” section, anyone making use of the information herein assumes all liability arising from such use. 4.2. Conclusions

1. Based on the observed performance of the frames with deep columns and the behavior of their connections, there were no considerable reasons found to suggest preventing the use of deep column sections in any moment frame including special moment frames.

2. The inelastic analyses of connections with deep columns indicated that the study connections should be able to provide the required strength and especially the rotational ductility in excess of those required by FEMA-350 (2001) for pre-qualified connections. Figure 4.1 shows the FEMA requirement for minimum moment-rotation envelope curve (curve OYF) as well as representative envelop curve for connections with deep column studied herein (curve OYA). As the figure indicates, the connections with deep column clearly satisfy the FEMA requirement.

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 34

3. In reference to deep columns, FEMA-350 (2001), Page 2-23, states: “The pre-qualified

connections should only be used with W12 and W14 column sections”. According to FEMA-350, this statement is based on the results of only two tests of deep column specimens that were done at the time of development of FEMA reports. In these two tests, the deep columns showed a tendency to twist. A critical review of the test set-up, as discussed in previous sections, revealed that most likely such column twisting would not have occurred had the test set-up and the specimens been realistic representative of actual buildings. The specimens had no transverse beams connected to the panel zone of the columns and had no floor slabs. Almost all moment frame steel structures have floors (typically steel deck/concrete slab) and transverse beams, which provide significant lateral bracing. This investigation indicated that presence of the floor was enough to provide necessary bracing and to eliminate or to reduce the column twisting to insignificant and non-consequential levels.

4. The cyclic behavior of RBS connections with deep columns was found to be similar to the behavior of the same connection with W14 columns. Our studies indicated that there is no difference in bracing requirement for RBS connections with W14 and deep columns of up to W33 when there is a floor slab at least on one side of the beam.

5. By using deep columns, in a moment frame, the drift limits can be met with less steel tonnage compared to W14 column sections. This is due to considerably large moment of inertia of deep sections for the same weight per foot as a comparable W14 column.

Figure 4.1. Comparison of the M-θ Curve of Connections with Deep Columns to the M-θ Required by FEMA for Special Moment Connections

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 35

6. An added advantage of using deep column is a potential for saving in the cost of material and construction. In the 10-story study frames, the weight of the steel using W27 deep columns was about 1.3 lbs/ft2 less than the steel in the same frame but with W14 columns. According to a leading steel fabricator, the 1.3 lbs/ft2 equals to about 6-8% in total material saving based on 16-18 psf of steel for a typical structure of this type. Of course as mentioned earlier, this 10-story building was just an example to demonstrate that using deep columns instead of W14 not only can result in increasing lateral load resisting strength, decreasing drift, and reducing the cost. In other cases, the amount of saving may vary but most likely still there will be some economic gain in using deep columns.

7. The specimens without floor bracings, Figure 4.2(a), tested by Gilton, Chi and Uang (2000), cannot be considered representative of the actual structures. Design procedures and recommendations based on such test results cannot be justified. Future testing of the connections with deep columns need to be done such that the bracing effects provided by the floors and transverse beam(s) are represented. An example is shown in Figure 4.2(b). .

Figure 4.2. (a) Unrealistic Test Set-up used by Gilton, Chi, Uang (2000) and (b) Realistic Set-up

(a) (b)

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 36

________________________________________________________________________

References ________________________________________________________________________

ABAQUS (2001), User Manual I, II and III, Version 6.2, Hibbitt, Karlsson & Sorensen, Inc., Providence, RI.

AISC (1998), Load and Resistance Factor Design Specification for Structural Steel Buildings, American Institute of Steel Construction, Chicago, IL.

AISC (2002), Seismic Provisions for Structural Steel Buildings, (in review at this writing), American Institute of Steel Construction Inc., Chicago, IL.

Astaneh-Asl, A. (1995), “Seismic Behavior and Design of Bolted Steel Moment-Resisting Frames”, Steel TIPS, Structural Steel Educational Council, Moraga, CA.

(This report can be downloaded free from www.aisc.org web site.)

Bolt, B. and Gregor, N., (1993), “Synthesized Strong Ground Motion for the Seismic Condition Assessment of the Eastern Portion of the San Francisco Bay Bridge”, Report No. UCB/EERC-93/12, University of California, Berkeley, CA.

FEMA-350 (2001), “Seismic Design Criteria for Steel Moment-Frame Structures”, Report, Federal Emergency Management Agency, MD.

(This report can be downloaded free from www.fema.gov web site.)

Flynn, L., (2000), “Letter to the Editor,” Modern Steel Construction, American Institute of Steel Construction, November, Chicago, IL.

Gilton, C., Chi, B. and Uang, C. M. (2000), Cyclic Response of RBS Moment Connections: Weak-Axis Configuration and Deep Column Effects, Report No. SSRP-2000/03, Structural Systems Research Project, Department of Structural Engineering, University of California, San Diego, La Jolla, CA.

ICBO (1997), Uniform Building Code, International Conference of Building Officials, Whittier, CA.

Use of Deep Columns in Special Steel Moment Frames, J. Shen, A. Astaneh-Asl and D. B. McCallen, 2002. 37

Kitjasateanphun T. (2001), Seismic performances of Reduced Beam Section Frames, Ph.D. Thesis, Department of Civil and Architectural Engineering, Illinois Institute of Technology, Chicago, IL.

Moore, K.S., Malley, J.O., and Engelhardt, M.D., (1999), “Design of Reduced Beam Section (RBS) Moment Frame Connections”, Steel TIPS, Structural Steel Educational Council, Moraga, CA.

(This report can be downloaded free from www.aisc.org web site.)

Ricles, J.M., Mao, C., Lu, L-W and Fisher, J.W., (2000) “Development and Evaluation of Improved Details for Ductile Welded Unreinforced Flange Connections”, ATLSS Report No. 00-04, ATLSS Engineering Research Center , Lehigh University, Bethlehem, PA.

SAC, (1996), “Northridge Model Buildings”, Internal Report for SAC Researchers, SAC Joint Venture, Sacramento.

SCI, (2000), “Structural and Earthquake Engineering Software”, SAP-2000 Software, Computers and Structures, Berkeley.

SDI (1989), LRFD Design manual for Composite Beams and Girders with Steel Deck, No. LRFD1, The Steel Deck Institute.

SEAOC, (1999), “Recommended Lateral Force Requirements and Commentary”, Seventh Ed., Structural Engineers Association of California, Sacramento, CA.

SteelTIPS

STRUCTURAL STEEL EDUCATIONAL COUNCIL

TECHNICAL INFORMATION & PRODUCT SERVICE

August 2002

Cost Considerationsfor

Steel Moment FrameConnections

by

Patrick M. Hassettand

James J. Putkey

(A copy of this report can be downloaded free of charge for personal use from www.aisc.org)

Acknowledgments

The Authors wish to thank the members of the Structural Steel Educational Council (SSEC) fortheir support in publishing this Steel TIPS. Special thanks are due to the following SSECmembers and their firms for providing reviews and specific comments:

Bill Honeck, Forell/Elsesser EngineersBrett Manning, The Herrick Corporation

• Rick Wilkinson, Gayle Manufacturing

Disclaimer. The information presented in this publication has been prepared in accordance withrecognized engineering principles and is for general information only. While it is believed to be accurate,this information should not be used or relied upon for any specific application without competentprofessional examination and verification of its accuracy, suitability, and applicability by a licensedprofessional engineer, designer, or architect. The publication of the material contained herein is notintended as a representation or warranty on the part of the Structural Steel Educational Council or of anyother person named herein that this information is suitable for any general or particular use or of freedomfrom infringement of any patent or patents. Anyone making use of this information assumes all liabilityarising from such use.

Caution must be exercised when relying upon other specifications and codes developed by other bodiesand incorporated by reference herein since such material may be modified or amended from time to timesubsequent to the printing of this publication. The Structural Steel Educational Council and the authorsbear no responsibility for such material other than to refer to it and incorporate it by reference at the timeof the initial printing of this publication.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002

••

COST CONSIDERATIONS FORSTEEL MOMENT FRAME CONNECTIONS

By Patrick M. Hassett and James J. Putkey

TABLE OF CONTENTSACKNOWLEDGMENTS

TABLE OF CONTENTS Page

1. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

2. CONNECTIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

From FEMA 350 (Prequalified)

1. Welded Unreinforced Flange - Bolted Web (WUF-B) . . . . . . . . . . . . . . . . . . . . . 32. Welded Unreinforced Flange - Welded Web (WUF-W) . . . . . . . . . . . . . . . . . . . 53. Welded Free Flange (FF) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74. Welded Flange Plate (WFP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95. Reduced Beam Section (RBS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116. Bolted Unstiffened End Plate ( B U E P ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137. Bolted Stiffened End Plate (BSEP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158. Bolted Flange Plate (BFP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179. Double Split Tee ( D S T ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

From Previous Steel TIPS

10. Welded Flange Plate - Top Plate on Beam (WFP-Alt. 1) . . . . . . . . . . . . . . . . 2111. Welded Flange Plate - Loose Top Plate (WFP-Alt. 2) . . . . . . . . . . . . . . . . . . . 2312. Double Split Tee - Tees on Beam (DST-Alt. 1 ) . . . . . . . . . . . . . . . . . . . . . . . . . 25

Proprietary

13. Slotted Beam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2714. Bolted Bracket . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2915. Reduced Web . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

3. COST CONSIDERATION SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

4. REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

APPENDIX 1- Pre-Northridge Connection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36APPENDIX 2- Quality Assurance for Prequalified Connections . . . . . . . . . . . . . . . . . . . . . . 37ABOUT THE AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39LIST OF PUBLISHED STEEL TIPS

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel 77PS, August 2002

1. INTRODUCTION

This section sets forth the purpose of the Steel TIPS andgives a history of why the authors selected the variousconnections and their cost considerations.

PURPOSE

This Steel TIPS informs engineers of the various costconsiderations to construct ordinary and special moment frame connections.

Connections. The authors chose to gather information on 15 connections. These connectionsinclude:

Nine prequalified connections addressed in FEMA-350.Three proprietary connections referenced in FEMA-350.

•

Three connections included in previous Steel TIPS.

Limitations. This Steel TIPS does not comment on:

Connection design, including performance during an earthquake.Relative Cost Factors of the different connections.Beam to column web connections.

HISTORY OF CONNECTIONS AND COST CONSIDERATIONS

1986 Steel TIPS. A 1986 Steel TIPS, "Steel Connections, Details and Relative Costs," gaverelative costs of various types of connections—shear, non-moment, and moment. The TIPSauthors used fabrication and erection costs to determine relative costs, but without showing costitems. Connection CF-1 in the TIPS, web bolted-flange butt welded, later known as the "pre-Northridge" connection, became the moment frame connection of choice with the lowest relativecost of 1.0.

Northridge Earthquake. During the Northridge Earthquake, the "pre-Northridge" connectionexperienced brittle fractures. See FEMA-350 for a background on the fractures. The brittlefractures showed a need for welding electrodes with higher notch toughness.

FEMA-350 Recommendations. FEMA-350 gives design recommendations on prequalifiedconnections for ordinary and special moment frames. See AISC "Seismic Provisions forStructural Steel Buildings" for prequalification requirements. Rather than attempt to determinethe lowest cost connection, the authors present cost considerations comparing connectionsshown in FEMA and other currently used connections to the "pre-Northridge" connection. Thisapproach is intended to encourage the use of the variety of connections made available afterthe Northridge earthquake. This information also empowers the engineer to consider thepreferences of local fabricators and erectors when selecting connection types.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 1

••

•••

2. CONNECTIONSThis Section presents the 15 selected connections.

ORGANIZATION

The following 30 pages show and discuss each of the 15 connections by showing the connectiondetail on one page and discussing the cost considerations on the opposite page.

The cost considerations include material, detailing, fabrication, shipping, erection, quality control,and quality assurance. Additionally, some connections include FEMA prequalificationparameters for beam flange thickness. See FEMA-350 for complete prequalification data.

Each main cost consideration item includes sub-items appropriate to the main item. Forexample, fabrication includes sub-items for fit-up and welding.

DEVELOPMENT OF COST CONSIDERATIONS

The authors developed cost considerations based on their experience, and input from SSECfabricators and erectors. Obviously, not all fabricators and erectors agreed with each other.Fabrication and erection methods vary according to the firm's size, equipment, personnel, andlocation. Engineers should consider those variations when designing a connection andreviewing shop drawings.

Cost consideration comments compare connections to the "pre-Northridge" connection shownin Appendix 1. The comment "standard1 indicates the cost item considered has the sameapproximate cost as the "pre-Northridge" connection.

Chapter 3 gives a summary of the cost considerations in tabular format.

PROPRIETARY CONNECTIONS

Because of their nature, the authors needed permission from the patent holders to includeproprietary connections in this TIPS. We included connections of patent holders who gave usthe necessary permission.

EARTHQUAKE PERFORMANCE OF CONNECTIONS

Connections have different seismic performance characteristics. Please refer to FEMA reportsfor details regarding performance characteristics.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 2

2.1 WELDED UNREINFORCED FLANGE - BOLTED WEB(WUF-B)

Notes1. See Figure 3-8 and Note 1 of Figure 3-8 for top and bottom flange weld requirements. QC/QA category

AH/T. Refer to Figure 3-5 for weld access hole detail.2. Bolted shear tab. Use pretensioned A325 or A490 bolts. Weld to column flange with fillet weld both

sides, or with CJP weld, to develop full shear strength of plate. Weld QC/QA Category BL/T.3. See Figure 3-6 for continuity plate and web doubler plate requirements.

Reprinted from FEMA-350

Prequalification Data Considered.Type of frame: Ordinary Moment Frames (OMF) onlyMaximum beam flange thickness: 1 inch

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002

Figure 3-7 Welded Unreinforced Flange - Bolted Web (WUF-B) Connection

3

2.1 WELDED UNREINFORCED FLANGE - BOLTED WEB(WUF-B)

Cost Considerations

Material

Detailing

Shop Fabrication

Detail Parts

Main Parts

Fit-up

Welding

Shipping

Erection

Unloading

Shakeout

Erection

Plumb-up

Bolting

Welding

Fit-up

Preheat

Welding

Standard

Weld access holes require special detailing.

Standard

Weld access holes on beams require special work for cutting and grinding to roughnesswithin 500 micro inches.

Standard

Weld for continuity plates and shear tabs on columns needs notch tough electrode withslower deposition rates.

Standard

Standard

Standard

Standard

Standard

Standard

Standard

Standard

Notch tough electrode has slower deposition rates. Removal of back-up bar, backgouging, and fillet reinforcing is all overhead work and labor intensive.

Sequencing Standard

Quality Control /Quality Assurance

See Appendix 2 for current QA recommended by FEMA-350 and 353.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 4

2.2 WELDED UNREINFORCED FLANGE - WELDED WEB(WUF-W)

Notes1. CJP groove weld at top and bottom flanges. At top flange, either (1) remove weld backing, backgouge,

and add 5/16" minimum fillet weld, or (2) leave backing in place and add 5/16" fillet under backing. Atbottom flange, remove weld backing, backgouge, and add 5/16" minimum fillet weld. Weld: QC/QACategory AH/T.

2. Weld access hole, see Figure 3-5.3. CJP groove weld full length of web between weld access holes. Provide non-fusible weld tabs. Remove

weld tabs after welding and grind end of weld smooth at weld access hole. Weld: QC/QA CategoryBH/T.

4. Shear tab of thickness equal to that of beam web. Shear tab length shall be so as to allow 1/8" overlapwith the weld access hole at top and bottom, and the width shall extend 2" minimum back along thebeam, beyond the end of the weld access hole.

5. Full-depth partial penetration from far side. Weld: QC/QA Category BM/T.6. Fillet weld shear tab to beam web. Weld Size shall be equal to the thickness of the shear tab minus 1/16".

Weld shall extend over the top and bottom one-third of the shear tab height and across the top andbottom. Weld: QC/QA Category BL/L.

7. Erection bolts: number, type, and size selected for erection loads.8. For continuity plates and web doubler plates see Figure 3-6.

Figure 3-8 Welded Unreinforced Flange-Welded Web (WUF-W) Connection

Reprinted from FEMA-350

Prequalification Data ConsideredType of frame: OMF, Special Moment Frame (SMF)Maximum beam flange thickness: OMF -1 ½ inch, SMF -1 inch

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 5

2.2 WELDED UNREINFORCED FLANGE - WELDED WEB(WUF-W)

Cost Considerations

Material

Detailing

Shop Fabrication

Detail Parts

Main Parts

Standard

Weld access holes require special detailing.Web welding requires special detailing to suit erector.

Ends of column shear tabs require angled cuts adding labor to hand made plates.Column shear tabs require bevel preparation for weld to column.

Weld access holes on beams require special work. Fabricated roughness is required to500 micro-inches.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002

Fit-up

Welding

Shipping

Erection

Unloading

Shakeout

Erection

Plumb-up

Bolting

Welding

Fit-up

Preheat

Welding

Sequencing

Quality Control /Quality Assurance

Standard

Weld for continuity plates and shear tabs on columns needs notch tough electrode withslower deposition rates.

Standard

Standard

Standard

Standard

Standard

Standard

Tight fit-up of web to shear tab may require more bolts than determined for erectionloads.Fit-up of web for CJP web weld can be difficult if fabrication is not done correctly.

CJP and fillet welds for web require additional preheat.

Notch tough electrode has slower deposition rates. Removal of back-up bar, backgouging, and fillet reinforcing is all overhead work and labor intensive.Vertical CJP weld of beam web to column requires significant additional difficult welding.Skill level of welders and UT technicians are important factors to field production on theseCJP welds.Non-fusible run-off tabs for web end weld require additional work in a cramped space.Fillet weld of beam web to shear tab requires significant additional welding.

Special sequencing is required when considering preheat, restraint, and cooling of welds.

See Appendix 2 for current QA recommended by FEMA-350 and 353.

6

2.3 WELDED FREE FLANGE (FF)

Notes1. CJP groove weld. Note 1 of Figure 3-8 applies. Weld: QC/QA Category AH/T.2. See design procedure in Section 3.5.3.1, Steps 5 through 8, for web plate size and thickness.3. ½" minimum radius.4. Erection bolts: number, type and size selected for erection loads.5. CJP double-bevel groove weld. Weld: QC/QA Category BH/T.6. Fillet welds size, length, calculated in Section 3.5.3.1, Step 8. Weld: QC/QA Category BH/L.7. For continuity plates and web doubler plates see Figure 3-6.

Reprinted from FEMA-350

Prequalification Data Considered

Type of frame: OMF, SMFMaximum beam flange thickness: OMF -1 1/4 inch, SMF - 3/4 inch

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002

Figure 3-9 Welded Free Flange (FF) Connection

7

2.3 WELDED FREE FLANGE (FF)

Cost Considerations

Material

Detailing

Shop Fabrication

Detail Parts

Main Parts

Fit-up

Welding

Shipping

Erection

Unloading

Shakeout

Erection

Plumb-up

Bolting

Welding

Fit-up

Preheat

Welding

Larger shear tabs required.

Special detailing required for girder web cut-out.

Ends of column shear tabs require angled cuts; adding labor for hand made plates.Column shear tabs require double bevel preparation.

Beam web cut-out requires special work. Cutting in the fillet region of the web-flangeintersection is difficult, especially when hand burned, with grinding required.

CJP weld for shear tab requires some additional fit-up work.

Weld for continuity plates and shear tabs on columns needs notch tough electrode withslower deposition rates.CJP weld for shear tab causes difficult welding distortion control.

Standard, but wider shear tab on column must be watched. Care must be taken to avoidbending ends of beam flanges.

Standard

Standard

Standard

The deep cut-out of the web may present plumb-up problems.

Tight fit-up of web to shear tab may require more bolts than determined for erectionloads.

Standard

Additional preheat required for shear tab fillet welding.

Notch tough electrode has slower deposition rates. Removal of back-up bar, backgouging, and fillet reinforcing is all overhead work and labor intensive.Additional shear tab fillet welding with multiple passes in vertical and overhead positionsis likely.

Sequencing Special sequencing required when considering preheat, restraint, and cooling of welds.

Quality Control /Quality Assurance

See Appendix 2 for current QA recommended by FEMA-350 and 353.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 8

2.4 WELDED FLANGE PLATE (WFP)

Notes1. Flange plate. See Section 3.5.4.1, Steps 1-4, for sizing requirements. Plates shall be fabricated with

rolling direction parallel to the beam.2. CJP groove weld: single or double bevel. Weld in shop or field. When using single-bevel groove weld,

remove backing after welding, back-gouge, and reinforce with 5/16"-minimum fillet weld. When usingdouble bevel weld, back-gouge first weld before welding other side. Weld QC/QA Category AH/T. Ifplates are shop welded to column, care must be exercised in locating and leveling plates, as shimming isnot allowed between the plates and the beam flanges. If plates are field-welded to column afterconnecting to beam, weld access holes of sufficient size for weld backing and welding access shall beprovided.

3. Fillet welds at edges of beam flanges to plate. Size welds according to the procedure in Section 3.5.4.1,Step 5. Welds may be shop or field. Provide weld tabs at end to provide full weld throat thickness to theend of the plate. Remove weld tabs and grind the end of the weld smooth. Use care to avoid grindingmarks on the beam flange. Weld: QC/QA Category BH/L.

4. Fillet weld at end of flange plate to beam flange. Welds may be shop or field. Maintain full weld throatthickness to within 1" of the edge of the flange. Weld: QC/QA Category BH/T.

5. Shear tab of length equal to db-2k-2". Shear tab thickness should match that of beam web.6. Erection bolts: number, type, and size selected for erection loads.7. Full depth-partial penetration from far side. Weld: QC/QA Category BM/T.8. Fillet weld both sides. Fillet on side away from beam web shall be same size as thickness of shear tab.

Fillet on the side of the beam web shall be ½". Weld: QC/QA Category BH/T.9. Fillet weld shear tab to beam web. Weld size shall be equal to the thickness of the shear tab minus 1/16".

Weld: QC/QA Category BH/L.10. For continuity plates and web doubler plates see Figure 3-6. For calculation of continuity plate requirements,

use flange plate properties instead of beam flange properties.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002

Figure 3-11 Welded Flange Plate (WFP) Connection

Reprinted from FEMA-350

Prequalification Data ConsideredType of frame: OMF, SMFMaximum beam flange thickness: OMF - 1 ½ inch, SMF -1 inch

9

2.4 WELDED FLANGE PLATE (WFP)

Cost Considerations

Material

Detailing

Shop Fabrication

Detail Parts

Main Parts

Fit-up

Welding

Shipping

Erection

Unloading

Shakeout

Erection

Plumb-up

Bolting

Welding

Fit-up

Preheat

Welding

Sequencing

Extra material required for flange plates.The authors consider the shop welding of both top and bottom flange plates impracticalbecause of resulting erection tolerances. We consider a shop welded bottom plate andfield welded top plate as the practical option.

Special detailing is required for locating beam web to shear tab holes in relation to thebottom flange plate.

Flange plates require CJP bevel preparation, and shop must track rolling direction.

Beam does not require flange bevel preparation or access holes. Top flange needs acope for back-up bar. Web requires bevel for PJP weld to shear tab.

Shear tab and bottom flange plate require additional fit-up. Bottom flange plate fit-upmust be square and level.

Weld for continuity plates, shear tabs, and flange plates on columns needs notch toughelectrode with slower deposition rates.CJP welds on flange plates need distortion control.

Column shipping takes more trailer space because of protruding flange plates.Protruding flange plates require special care to avoid bending.

Protruding flange plates require special care to avoid bending.

Column flange plates take some additional deck space.

Erection can be impaired if detailing and fabrication do not account for beam tolerancesand if flange plates are not square to column.

Proper sequencing of the top flange plate weld will eliminate problems of bay shrinkagewith resulting benefits to plumb-up.

Tight fit-up of web to shear tab may require more bolts than determined for erectionloads.

The loose top flange plate allows field to set correct root openings.

Less preheat required for fillet welds

Notch tough electrode has slower deposition rates.Removal required of back-up bars and run-off tabs on top flange plates.Fillet welding in lieu of CJP welding is a benefit for the field.Fillet weld of beam web to shear tab and vertical PJP weld of beam web to columnrequire significant welding.Possible gaps at bottom flange of beam to flange plate may require larger fillet welds.

Preheat, restraint, and cooling of web groove weld may require special sequencing.Welders must follow a specified joint construction procedure.

Quality Control /Quality Assurance

See Appendix 2 for current QA recommended by FEMA-350 and 353.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 10

2.5 REDUCED BEAM SECTION (RSB)

Notes1. See Section 3.5.5.1 for calculation of RBS dimensions. See FEMA-353, Recommended Specifications

and Quality Assurance Guidelines for Steel Moment Frame Construction for Seismic Applications, forfabrication details including cutting methods and smoothness requirements.

2. See Figure 3-8, and Note 1 to Figure 3-8, except that weld access hole may be as shown there, or as inAISC LRFD Vol. 1, Fig. C-J1.2, for rolled shapes or groove welded shapes.

3. Web Connection: Erection bolts: number, type, and size selected for erection loads.a. Alternative 1: CJP welded web. Weld QC/QA Category BM/L. Shear tab length is equal to the

distance between the weld access holes plus ¼". Shear tab thickness is as required for erection andthe tab serves as backing for CJP weld (3/8" min. thickness). Shear tab may be cut square, ortapered as shown. Weld of shear tab to column flange is minimum 3/16" fillet on the side of thebeam web, and a fillet sized for erection loads (5/16" minimum) on the side away from the beamweb. No weld tabs are required at the ends of the CJP weld and no welding of the shear tab to thebeam web is required.Weld: QC/QA Category BM/L. Erection bolts are sized for erection loads.

b. Alternative 2: Bolted shear tab. Shear tab and bolts are sized for shear, calculated as in Section 3.2and using the methods of AISC. The shear tab should be welded to the column flange with a CJPgroove weld or fillet of ¾ tpl on both sides. Weld: QC/QA Category BL/T. Bolts shall be ASTMA325 or A490, and shall be fully-tightened.

4. For continuity plates and web doubler plates see Figure 3-6.

Figure 3-12 Reduced Beam Section (RBS) Connection

Reprinted from FEMA-350

Prequalification Data ConsideredType of frame: OMF, SMFMaximum beam flange thickness: 1 3/4 inch

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 11

2.5 REDUCED BEAM SECTION (RBS)

Material

Detailing

Shop Fabrication

Detail Parts

Main Parts

Fit-up

Welding

Shipping

Erection

Unloading

Shakeout

Erection

Plumb-up

Bolting

Welding

Fit-up

Preheat

Welding

Sequencing

Cost Considerations(For welded web or bolted web options)

Reduced section requires a slight increase in beam weight.

Special detailing required for cut-out of flange reduced section and weld access holes.

End cuts on column shear tabs are not mandatory; increased cost if manually cut.Column shear tabs require large fillets, or bevel preparation and CJP if bolted option used.

Weld access holes on beams require special work for cutting and grinding to roughnesswithin 500 micro inches.Automated equipment provides more precise and efficient cutting of reduced beamsections. Reduced section cuts may require grinding. See FEMA-350 for repairrecommendations.If welded web option used, then beam web requires beveled edge.

More fit-up required for bolted web option because of CJP weld.

Weld for continuity plates and shear tabs on columns needs notch tough electrode withslower deposition rates.Bolted web option requires CJP or heavy fillet weld on shear tab.

Standard

Standard

Standard

Standard

Standard

Standard, but tight fit-up of web to shear tab may require more bolts than determined forerection loads.

Standard for bolted option. Welded web option may be more difficult if fabricationtolerances are not controlled.

Standard for bolted option; welded web requires additional preheat.

Notch tough electrode has slower deposition rates. Removal of back-up bar, backgouging, and fillet reinforcing is all overhead work and labor intensive.CJP weld of beam web to column requires significant additional welding.

Special sequencing is required for welded web option when considering preheat,restraint, and cooling of welds.

Quality Control /Quality Assurance

See Appendix 2 for current QA recommended by FEMA-350 and 353.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 12

2.6 BOLTED UNSTIFFENED END PLATE (BUEP)

Notes1. ASTM A36 end plate. For sizing see Section 3.6.1.1.2. CJP groove weld. This weld has special requirements. See FEMA-353, Recommended Specifications and

Quality Assurance Guidelines for Steel Moment Frame Construction for Seismic Applications, forfabrication details. Weld: QC/QA Category AH/T.

3. Fillet weld both sides, or CJP weld; see Section 3.6.1.3 for sizing requirements. See FEMA-353,Recommended Specifications and Quality Assurance Guidelines for Steel Moment Frame Constructionfor Seismic Applications, for fabrication details. Weld: QC/QA Category BM/L.

4. Pretensioned ASTM A325 or A490 bolts. Diameter not to exceed 1-1/2 inch. See Section 3.6.1.1 forsizing requirements.

5. Bolt location is part of the end plate design. See Section 3.6.1.1.6. For continuity plates and web doubler plates, see Figure 3-6. For calculation of panel zone strength, see

Section 3.6.1.1.7. Shim as required. Finger shims shall not be placed with fingers pointing up.

Figure 3-13 Bolted Unstiffened End Plate (BUEP) Connection

Reprinted from FEMA-350

Prequalification Data ConsideredType of frame: OMF, SMFMaximum beam flange thickness: 3/4 inch

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 13

2.6 BOLTED UNSTIFFENED END PLATE (BUEP)

Cost Considerations

Material

Detailing

Shop Fabrication

Detail Parts

Main Parts

Fit-up

Welding

Shipping

Erection

Unloading

Shakeout

Erection

Plumb-up

Bolting

Welding

Quality Control /Quality Assurance

Heavy end plate, shims, and longer doubler plates add significant material cost.

Special detailing is required for end plates and to allow for erection clearances, columndepth over-run tolerances, column flange twist tolerance, and shimming.

End plate holes must be precisely located to match column holes. Automated equipmentprovides more precise and efficient plate cutting and hole drilling.

Hole drilling on column flange must be precise to match end plate holes; best made withautomated fabrication equipment.No weld access holes are required.

Positioning of end plate requires careful fit-up on beam.

CJP of beam flanges to end plate requires additional shop welding.Notch tough electrode has slower deposition rates.No web bolt holes or weld access holes required. Note: The CJP flange weld is madewithout a weld access hole; testing has shown this procedure acceptable.

End plates may require additional cribbing.

Extended end plate causes handling problems.

End plates require additional blocking on deck.

End plates present the problem of fitting the beam between the column flanges resultingin extra erection time including expensive crane time.

Plumb-up is difficult due to fixity of connection and shimming.Column depth tolerance can throw off bay widths.Shimming is required to obtain correct bay width.

The shimming required is time consuming.Bolt fit-up and installation may be a problem if fabrication is imperfect.Misaligned holes may require reaming.Bolt sizes greater than 1 1/8 inch diameter require heavier equipment to fully tension.

No field welding is required.

Welding quality control and quality assurance are shifted from the field to the shop.Shop must perform more careful fabrication with resulting quality control increase.Absence of weld access hole simplifies UT at web intersection.See Appendix 2 for current QA recommended by FEMA-350 and 353.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 14

2.7 BOLTED STIFFENED END PLATE (BSEP)

Notes1. ASTM A36 end plate. For sizing, see Section 3.6.2.1.2. CJP groove weld. This weld has special requirements. See FEMA-353, Recommended Specifications and

Quality Assurance Guidelines for Steel Moment Frame Construction for Seismic Applications, forfabrication details. Weld: QC/QA Category AH/T.

3. Fillet weld both sides, or CJP weld; see Section 3.6.2.4 for sizing requirements. See FEMA-353,Recommended Specifications and Quality Assurance Guidelines for Steel Moment Frame Constructionfor Seismic Applications, for fabrication details. Weld: QC/QA Category BM/L.

4. Pretensioned ASTM A325 or A490 bolts. See Section 3.6.2.1 for sizing requirements.5. Bolt location is part of the end plate design. See Section 3.6.2.1.6. For continuity plates and web doubler plates, see Figure 3-6. For calculation of panel zone strength, see

Section 3.6.2.1.7. Stiffener is shaped as shown. Stiffener thickness shall be the same as that of the beam web.8. Stiffener welds are CJP double-bevel groove welds to both beam flange and end plate. Weld: QC/QA

Category AH/T for weld to endplate. BM/L for weld to beam..9. Shim as required. Finger shims shall not be placed with fingers pointing up.

Figure 3-15 Stiffened End Plate Connection

Reprinted from FEMA-350

Prequalification Data ConsideredType of frame: OMF, SMFMaximum beam flange thickness: 1 inch

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 15

2.7 BOLTED STIFFENED END PLATE (BSEP)

Cost Considerations

Material

Detailing

Shop Fabrication

Detail Parts

Main Parts

Fit-up

Welding

Shipping

Erection

Unloading

Shakeout

Erection

Plumb-up

Bolting

Welding

Quality Control /Quality Assurance

Heavy end plates, stiffeners, shims, and longer doubler plates add significant materialcost.

Special detailing is required to allow for erection clearances, column depth over-runtolerances, column flange twist tolerance, and shimming. End plates and stiffenersrequire additional detailing.

End plate holes must be precisely located to match column holes. Automated equipmentprovides more precise and efficient fabrication.

Hole drilling on column flange must be precise to match end plate holes; best made withautomated fabrication equipment.No weld access holes are required.

Positioning of end plate requires careful fit-up. Stiffener plates require additional fit-upwork.

CJP of beam flanges to end plate requires additional shop welding.Notch tough electrode has slower deposition rates.No web bolt holes or weld access holes required. Note: The CJP flange weld is madewithout a weld access hole; testing has shown this procedure acceptable.Stiffener plate CJP weld to end plate may cause end plate distortion.Multiple positioning of beam requires more rolling of beam due to CJP at top flange,bottom flange, and stiffener plates.

Beams take more trailer space and require more cribbing because of end plates.

Stiffened end plate causes handling problems.

End plates require additional blocking on deck.

End plates present the problem of fitting the beam between the column flanges resultingin extra erection time including expensive crane time.Stiffener plates may distort the end plate, causing additional erection problems.

Plumb-up is difficult due to fixity of connection and shimming.Column depth tolerance can throw off bay widths.Shimming is required to obtain correct bay width.

The shimming required is time consuming.Bolt fit-up and installation may be a problem if fabrication is not nearly perfect.Misaligned holes may require reaming.Bolt sizes greater than 1 1/8 inch diameter require heavier equipment to fully tension.Increased number of holes increases probability of misalignment.

No field welding is required.

Welding quality control and quality assurance are shifted from the field to the shop.Shop must perform more careful fabrication with resulting quality control increase.Absence of Weld access holes simplifies UT at web intersection.See Appendix 2 for current QA recommended by FEMA-350 and 353.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 16

2.8 BOLTED FLANGE PLATE (BFP)

Notes1. Size the flange plate and bolts in accordance with Section 3.6.3.1. Bolts are fully pretensioned ASTM

A325 or A490, designed for bearing. Bolt holes in flange plate are oversize holes. Use standard holes inbeam flange. Washers as required by RCSC, Section 7.

2. CJP groove weld, single or double bevel. Weld in shop or field. When using single-bevel groove weld,remove backing after welding, backgouge, and reinforce with 5/16" minimum fillet weld. When usingdouble bevel weld, backgouge first weld before welding other side. Weld: QC/QA Category AH/T.

3. Shims are permitted between flange plates and flanges.4. Size shear tab and bolts by design procedure in Section 3.6.3.2. Bolt holes in shear tab are short-slotted-

horizontal; holes in web are standard. Weld QC/QA Category BM/L.5. For continuity plates and web doubler plates see Figure 3-6. For calculation of continuity plate requirements,

use flange plate properties as flange properties.

Reprinted from FEMA-350

Prequalification Data ConsideredType of frame: OMF, SMFMaximum beam flange thickness: OMF -1 1/4 inch, SMF - 3/4 inch

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 17

Figure 3-17 Bolted Flange Plate (BFP) Connection

2.8 BOLTED FLANGE PLATE (BFP)

Cost ConsiderationsMaterial

Detailing

Shop Fabrication

Detail Parts

Main Parts

Fit-up

Welding

Flange plates and shims add additional material.The authors consider shop welded, field bolted flange plates as the practical option sinceshimming facilitates the necessary erection tolerances.

Special detailing required to allow for beam depth over-run tolerance, beam flange twisttolerance, and shimming for shop attached flange plates.

Flange plate holes must be precisely located to match beam flange holes.Flange plates require bevel preparation. Shop must track rolling direction.

Hole drilling on beam flanges must be precise. Automated equipment provides moreprecise and efficient fabrication.No weld access holes required on beams.

Flange plate fit-up must be carefully braced square and level and allowance made forweld shrinkage.

Weld for continuity plates, shear tabs, and flange plates on columns needs notch toughelectrode with slower deposition rates.CJP beam flange plate weld to column adds shop welding, but weld is better positionedin shop and protected from weather.Flange plate angular distortion must be controlled during welding.

Shipping Column shipping takes more trailer space because of protruding flange plates.Protruding flange plates require special care to avoid bending.

Erection

Unloading

Shakeout

Erection

Plumb-up

Bolting

Welding

Quality Control /Quality Assurance

Protruding flange plates require special care to avoid bending.

Column flange plates take some additional deck space.

Can go smoothly if flange plates are straight. Sufficient gap must be made betweenflange plates to allow quick erection between columns.

Bolts in flange plates may help plumb-up process by keeping bays from racking.

Required shimming is time consuming.Oversized holes in flange plates and slotted holes in shear tabs will help hole alignment.Bolt sizes greater than 1 1/8 inch diameter require heavier equipment to fully tension.

No field welding.

Weld quality control and quality assurance are shifted from the field to the shop.See Appendix 2 for current QA recommended by FEMA-350 and 353.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 18

2.9 DOUBLE SPLIT TEE (DST)

Notes1. Split Tee: length, width, and thickness determined by design according to Section 3.7.1.2.2. Fully pretensioned ASTM A325 or A490 bolts in standard holes sized for bearing. For sizing, see

Section 3.7.1.2, Step 7.3. Fully pretensioned ASTM A325 or A490 bolts in standard holes sized for bearing. For sizing, see

Section 3.7.1.2, Step 4.4. Shear tab welded to column flange with either CJP weld or two-sided fillet weld. For calculation of

design strength of shear tab, welds, and bolts, see Section 3.7.1.2, Step 14. Weld: QC/QA CategoryBM/L.

5. For continuity plates and web doubler plates see Figure 3-6.

Figure 3-20 Double Split Tee (DST) Connection

Reprinted from FEMA-350

Prequalification Data ConsideredType of frame: OMF, SMF

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 19

2.9 DOUBLE SPLIT TEE (DST)

Material

Detailing

Shop Fabrication

Detail Parts

Main Parts

Fit-up

Welding

Bolting

Shipping

Erection

Unloading

Shakeout

Erection

Plumb-up

Bolting

Welding

Quality Control /Quality Assurance

Cost Considerations

Split tees and shims add additional material.

Special detailing required to allow for beam and column depth over-run tolerance, beamflange twist tolerance, and shimming.

Split tee holes must be precisely located to match beam flange holes and column flangeholes. Automated equipment provides more precise and efficient fabrication.Commonly, tees are cut from W shapes to make WT shapes.

Hole drilling on beam flanges and column flanges must be precise, best made withautomated fabrication equipment.No weld access holes required on beams.

Shear tab is more easily fit-up when fillet welded.

Weld for continuity plates, shear tabs, and doubler plates on columns needs notch toughelectrode with slower deposition rates.

Split tee positioning must be carefully made square and level and allowance made forshim compression.

Beam or column shipping takes more trailer space because of protruding split tees.Protruding split tees require special care to avoid bending.

Protruding split tees require special care to avoid bending.

Split tees on columns or beams need special cribbing.

Can go smoothly if split tees are straight. A bend in WT web will hold up erection due tobeam getting jammed between columns.

Standard.

Required shimming is time consuming.Standard holes specified may require reaming.Bolt sizes greater than 1 1/8 inch diameter require heavier equipment to fully tension.

No field welding.

Weld quality control is shifted from the field to the shop.See Appendix 2 for current QA recommended by FEMA-350 and 353.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 20

Prequalification Data ConsideredForell/Elsesser Engineers qualified this connection by test for a specific column and beamcombination on a specific project.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 21

2.10 WELDED FLANGE PLATE - TOP PLATE ON BEAM(WFP-ALT.1)

This connection is a version of the Welded Flange Plate (WFP) Connection. The top flange plate isshop fillet welded to the beam.

2.10 WELDED FLANGE PLATE - TOP PLATE ON BEAM(WFP-ALT.1)

Material

Detailing

Shop Fabrication

Detail Parts

Main Parts

Fit-up

Welding

Shipping

Erection

Unloading

Shakeout

Erection

Plumb-up

Bolting

Welding

Fit-up

Preheat

Welding

Cost Considerations

Flange plates require extra material.

Locating beam web holes and shear plate holes with bottom flange plate requires specialdetailing.

Column shear tabs require CJP bevel preparation.Flange plates require CJP bevel preparation.Flange plates require cutting to fit flange width. Automated equipment provides moreprecise and efficient fabrication. Shop must track rolling direction.

Beam does not require bevel preparation or access holes.Cope required at top flange.

Bottom flange plate and shear plate require careful fit-up to ensure tolerances are kept.

Weld for continuity plates, shear plate, and bottom flange plate on column needs notchtough electrode with slower deposition rates.CJP weld on shear plate and bottom flange plate need distortion control.

Column shipping takes more trailer space because of protruding flange plates.

Protruding flange plates require special care to avoid bending.

Column flange plates take some additional deck space.

Bottom plate could cause problems for connecting if not fabricated with care.

Standard

Web connection requires only two erection bolts.

May need to clamp bottom flange plate to bottom flange of beam.Field must remove shop fit-up bar on bottom flange plate.

Fillet welds require less preheat.

Notch tough electrode has slower deposition rates.Back-up bar removal and fillet weld reinforcement requires work in overhead position.Fillet welding of bottom flange plate to beam flange requires significant welding, but inhorizontal position.Fillet weld of beam web to shear tab requires significant welding, some in overhead andvertical positions.

Sequencing Special sequence is required when considering preheat, restraint, and cooling of welds.

Quality Control /Quality Assurance

Since top flange plate CJP weld is not impaired by beam web, it is inherently a weld thathas less problems with quality and UT.See Appendix 2 for current QA recommended by FEMA-350 and 353.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 22

ELEVATION

Reprinted from Steel TIPS, see Reference 5

Prequalification Data ConsideredThis connection lacks prequalification.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 23

2.11 WELDED FLANGE PLATE - LOOSE TOP PLATE(WFP-ALT. 2)

This connection is a version of the Welded Flange Plate (WFP) connection. The top flange plate is shippedloose. See Reference 5 for connection origin.

2.11 WELDED FLANGE PLATE - LOOSE TOP PLATE(WFP-ALT.2)

Material

Detailing

Shop Fabrication

Detail Parts

Main Parts

Fit-up

Welding

Shipping

Erection

Unloading

Shakeout

Erection

Plumb-up

Bolting

Welding

Fit-up

Preheat

Welding

Sequencing

Quality Control /Quality Assurance

Cost ConsiderationsFlange plates require extra material.

Locating beam web holes and shear plate holes with bottom flange plate requires specialdetailing.The absence of restrained field welds eliminates the need to provide for weld shrinkage.

Column shear tabs require CJP bevel preparation.Flange plates require CJP bevel preparation.Flange plates require cutting to fit flange width, an operation best suited for automatedfabrication equipment.

Beam does not require bevel preparation or access holes.

Bottom flange plate and shear plate require careful fit-up to ensure tolerances are kept.

Weld for continuity plates, shear plate, and bottom flange plate on column needs notchtough electrode with slower deposition rates.CJP weld on shear plate and bottom flange plate need distortion control.

Column shipping takes more trailer space because of protruding flange plates.

Protruding flange plates require special care to avoid bending.

Column flange plates take some additional deck space.

If not properly fabricated, bottom plate could cause problems aligning beam web holeswith column shear tab holes.

Tightening of web bolts before welding sets column bay spacing.Welding will not effect column spacings with resulting benefits to plumb-up.

Web connection requires only three erection bolts.

Top and bottom flange plates may require clamping to beam flanges.

Fillet welds require less preheat.

Notch tough electrode has slower deposition rates.Back-up bar removal and fillet weld reinforcement requires work in overhead position.Fillet welding of top and bottom flange plates to beam flanges requires significantwelding, half in overhead position.Fillet weld of beam web to shear tab requires significant welding, some in overhead andvertical positions.Possible gaps at bottom flange of beam to flange plate may require larger fillet welds.

Welders must make CJP weld on top flange plate before fillet welding plate to beam.No other joint, connection, or bay sequencing is required.

See Appendix 2 for current QA recommended by FEMA-350 and 353.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 24

2.12 DOUBLE SPLIT TEE - TEES ON BEAM (DST-ALT.1)

This connection is a variation of the Double Split Tee (DST) connection. The split tees are shop filletwelded to the beam flange.

Reprinted from Steel TIPS, see Reference 9

Prequalification Data ConsideredProfessor Popov tested this connection for a specific column and beam size. See Reference 9.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 25

2.12 DOUBLE SPLIT TEE - TEES ON BEAM (DST-ALT.1)

Cost Considerations

Material

Detailing

Shop Fabrication

Detail Parts

Main Parts

Fit-up

Welding

Bolting

Shipping

Erection

Unloading

Shakeout

Erection

Plumb-up

Bolting

Welding

Quality Control /Quality Assurance

Split tees, longer doubler plates, larger shear tabs, and shims add additional material.Shims are required between WT flanges and column flange.

Special detailing required to allow for beam and column depth over-run tolerance andshimming.

Split tee holes must be precisely located to match column flange holes.Commonly, tees are cut from W shapes to make WT shapes.

Hole drilling on column flanges must be precise, an operation best suited forautomated fabrication equipment.No weld access holes required on beams.Beams require no flange bevel preparation.

Split tee fit-up must be carefully made square and level and allowance made for shimcompression.Fillet welded shear tab requires careful fit-up to match beam web holes.

Weld for continuity plates, shear tabs, and doubler plates on columns and tees tobeam flanges needs notch tough electrode with slower deposition rates.

The only shop bolting is the four bolts for each tee.

Beam shipping takes more trailer space because of protruding split tees.Protruding split tees require special care to avoid bending.

Protruding split tees require special care to avoid bending.

Beam split tees take some additional cribbing to properly stack on deck prior toerecting.

Shop welded split tees may cause erection problems because beams need to beentered sideways.

Standard

Required shimming between split tee flanges and column flanges is time consuming.1 1/4 inch diameter bolts require heavier equipment to fully tension.

No field welding.

Weld quality control is shifted from the field to the shop, and CJP welds arereplaced with fillet welds.See Appendix 2 for current QA recommended by FEMA-350 and 353.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 26

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 27

2.13 SLOTTED BEAM (PROPRIETARY)

Prequalification Data Considered.Type of frame: OMF and SMFThe patent holder, SSDA, has qualified this connection for various beam/column combinations,including columns greater than W14.

Reprinted from "Slotted Web connection Manual." See Reference 10.

SSDA BEAM SLOT CONNECTION- Prequalified by ICBO (ER -5861)

U.S. Patent Nos. 5,680,738 & 6,237,303

2.13 SLOTTED BEAM (PROPRIETARY)

Cost Considerations

Material

Detailing

Shop Fabrication

Detail Parts

Main Parts

Fit-up

Welding

Shipping

Erection

Unloading

Shakeout

Erection

Plumb-up

Bolting

Welding

Fit-up

Preheat

Welding

Standard, but there is an added cost for using the proprietary system.

Weld access holes and beam web slots require special detailing.Web welding requires special detailing depending on beam sizes.

Standard

Weld access holes and beam slots require special work. Automated equipmentprovides more precise and efficient fabrication.

Standard

Welds for continuity plates, if those plates are required by the engineer, and sheartabs on columns need notch tough electrode with slower deposition rates.

Standard

Standard

Standard

Standard

Due to web flexibility, some erectors leave a portion of slot temporarily uncut tofacilitate plumbing. After the flange and web welds are completed, the remainder ofslot is cut. Short slotted holes in shear tab, if used, require more plumb-up work.

Standard

Standard

Standard

Notch tough electrode has slower deposition rates.Removal of back-up bar, back gouging, and fillet reinforcing is all overhead work andlabor intensive; however, the top flange back up bar does not require removal.CJP weld of beam web to column requires significant additional welding.Fillet weld of beam web to shear tab requires significant welding.

Sequencing Erector must follow patent holder's specified connection construction procedure.

Quality Control /Quality Assurance

Fabricator must submit shop drawings to SSDA for approval.See FEMA-350 and 353 for current QA recommendations.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 28

2.14 BOLTED BRACKET (PROPRIETARY)

Patented cast steel brackets, supposedly available on the marketplace, make this connectionproprietary. However, the authors could not locate the patent holder for such brackets. The authorsadded fabricated brackets to this TIPS because fabricated brackets are within the public domain.

Figure 3-25 Bolted Bracket Connection

Reprinted from FEMA-350

Prequalification Data ConsideredNo known data.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 29

2.14 BOLTED BRACKET (PROPRIETARY)

Cost Considerations

Material

Detailing

Shop Fabrication

Detail Parts

Main Parts

Fit-up

Welding

Bolting

Shipping

Erection

Unloading

Shakeout

Erection

Plumb-up

Bolting

Welding

Quality Control /Quality Assurance

Brackets add additional material.

Additional detailing is required for brackets.Fit-up issues must be identified and detailed to suit.Consider the beam depth and out of square tolerances. Oversized holes should help.

Brackets require additional fabrication.

Flanges of column and beams require drilling that must be precise. Automatedequipment provides more precise and efficient fabrication.No weld access holes required on beams.

Bolt bottom bracket to column with fully tensioned bolts. Consider whether to attachtop bracket or ship it loose.

Welds for continuity plates on columns and stiffeners on brackets need notch toughelectrode with slower deposition rates.

Bolt bottom bracket and furnish bolts for top bracket.

Must allow for brackets.

Brackets require special care.May need to handle loose brackets.

Brackets take some additional deck space and cribbing; more blocking needed forshakeout.

Connectors must take care to bolt and pin bottom flange to prevent beam fromtipping. Erector may need to consider adding a web plate.

Standard size holes in bottom bracket and bottom beam flange set correct bayspacing. No weld shrinkage to consider.

Oversize holes in top bracket facilitate hole alignment.Bolt sizes greater than 1 1/8 inch diameter require heavier equipment to fully tension.

No field welding.

Bolting quality control replaces welding quality control in field.See FEMA-350 and 353 for current QA recommendations.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 30

2.15 REDUCED WEB (PROPRIETARY)

Reduced Web Section configurations: (a) dual opening and (b) single opening.

Reprinted with permission from Professor Aschheim. See Reference 12.

Prequalification Data Considered.This connection is patented. Professor Mark Aschheim is the inventor and ProgrammaticStructures Inc., owned by Professor Aschheim, is the assignee. Professor Aschheim has testedvarious combinations of opening geometry and beam depth in combination with W14 columnsunder quasi-static reversed cyclic loading conforming to the SAC loading protocol.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 31

2.15 REDUCED WEB (PROPRIETARY)

Material

Detailing

Shop Fabrication

Detail Parts

Main Parts

Fit-up

Welding

Shipping

Erection

UnloadingShakeoutErectionPlumb-upBolting

Welding

Fit-up

Preheat

Welding

Sequencing

Quality Control /Quality Assurance

Cost Considerations(For bolted web [WUF-B] or welded web [WUF-W])

Standard

Weld access holes and web openings require special detailing.Welded web requires special detailing to suit erector.

StandardFor welded web:

Ends of column shear tabs require angled cuts adding labor to hand made plates.Column shear tabs require bevel preparation for weld to column.

Weld access holes require special work for cutting and grinding to roughness within500 micro inches.Web openings on beams require additional work.

Standard

Weld for continuity plates and shear tabs on columns needs notch tough electrodewith slower deposition rates.

Standard

StandardStandardStandardStandardStandard

Standard for bolted web option.For welded web:

Tight fit-up of web to shear tab may require more bolts than determined forerection loads.Fit-up of web for CJP web weld can be difficult if fabrication is not correct.

Standard for bolted web.CJP and fillet welds for welded web require additional preheat.

Notch tough electrode has slower deposition rates. Removal of back-up bar, backgouging, and fillet reinforcing is all overhead work and labor intensive.For welded web:

Vertical CJP weld of beam web to column requires significant additional difficultwelding.Skill level of welders and UT technicians are important factors to field productionon these CJP welds.Non-fusible run-off tabs for web end weld require additional work in a crampedspace.Fillet weld of beam web to shear tab requires significant additional welding.

Standard for bolted web.Special sequencing is required for welded web when considering preheat, restraint,and cooling of welds.

See FEMA-350 and 353 for current QA recommendations.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 32

3 COST CONSIDERATION SUMMARY

A GENERAL NOTE ON COST CONSIDERATIONS REGARDING WELDINGQUALITY CONTROL AND QUALITY ASSURANCE

Brittle fractures experienced in the Northridge earthquake have increased the intensity of weldinginspection with a corresponding increase in the cost of welded connections.

A review of Appendix 2 shows FEMA-353 recommendations: Complete joint penetration (CJP)groove welds require more costly ultrasonic testing (UT), and fillet welds require less costlymagnetic particle testing (MT). Therefore, fabricators and erectors normally prefer fillet weldedjoints over groove welded joints. Additionally, UT testing brings up the following issues:

The skill and training of the UT technician.The UT method used.The skill of the welder.The welder's methods and techniques.

•

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 33

•••

CO

TASK

MATERIALDETAILINGSHOP Detail Parts

Main PartsFit-upWelding

SHIPPINGERECTION

QA/QC

0dsI

mi

UnloadingShake-outErectionPlumb-upBoltingWeld Fit-upWeld PreheatWeldingWeld Seq.

sssisis

sssssssisi

ssiisis

sssssii

miii

No cost, task eliminated

ssiisis

sssissiiii

iiisi

mii

siidsiiiii

sismisis

sssssii

misi

iiisi

mii

si

mii

mi0000s

Decreased cost from Pre Northridge (significant)

imimismimii

si

mii

mi0000i

iiii

mimii

iiidmi0000i

imimiiiss

ssisi0000i

Same cost as Pre Northridge (or just a small increase or decrease)Increase in cost from Pre Northridge (significant)Major increase in cost from pre Northridge

imiisiis

iiissssiii

iiidiis

Iisdsdiiii

ii

midiii

siismi0000i

sisisss

sssisiiiii

iisssss

sisdi0000s

sisisis

sssssssisi

siiisis

sssssii

miii

Cost C

onsiderations for Steel M

oment Fram

e Connections,

Hassett and

Putkey, S

teel TIPS

, August 2002

CONNECTION COST COMPARISON SUMMARY

Section Ref: 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 2.10 2.11 2.12 2.13 2.14 2.15 2.15

Conn. Abbrev. WU

F-B

WU

F-W

FF WF

P

RB

S(W

EL

D W

EB

)

BU

EP

BS

EP

BFP

DS

T

WF

P-A

LT

.1

WF

P-A

LT

.2

DS

T-A

LT

.1

SL

OT

TE

D

BR

AC

KE

T

RE

D.

WE

B(W

UF

-B)

RE

D.

WE

B(W

UF

-W)

4. REFERENCES

1. AISC, (1992), Manual of Steel Construction: Volume IIConnections, ASD/LRFD, First ed., American Institute of Steel Construction, Chicago.

2. AISC, (1994), Manual of Steel Construction: Load and Resistance Factor Design for Structural SteelBuildings, 2nd Edition, American Institute of Steel Construction, Chicago.

3. AISC, (April 15,1977), Seismic Provisions for Structural Steel Buildings, American Institute of SteelConstruction, Chicago.

4. AISC, (November 10, 2000), Seismic Provisions for Structural Steel Buildings, (1997) SupplementNo.2, American Institute of Steel Construction, Chicago.

5. Collin, A.L., and Putkey, J.J., (1999), "Welded Moment Frame Connections With Minimal ResidualStress," Steel TIPS, Structural Steel Educational Council, Moraga.

6. FEMA-350, (2000), Recommended Seismic Design Criteria for New Steel Moment-Frame Buildings,prepared by the SAC Joint Venture for the Federal Emergency Management Agency, Washington,DC.

7. FEMA-353, (2000), Recommended Specifications and Quality Assurance Guidelines for SteelMoment-Frame Construction for Seismic Applications, prepared by the SAC Joint Venture for theFederal Emergency Management Agency, Washington, DC.

8. Forell/Elsesser, (1995), "Steel Construction Update: Successful Test of Welded Steel Beam - ColumnMoment Connection," Forell/Elsesser Engineers, Inc., Structural Engineers, San Francisco.

9. Popov, E.P and Takhirov, S.M., (2001), "Large Seismic Steel Beam-to-Column Connections," SteelTIPS, Structural Steel Educational Council, Moraga.

10. SSDA, (2002), Slotted Web™ Connection Manual, Seismic Structural Design Associates, Inc.,Mission Viejo.

11. Steel Committee, (1986), "steel connections/details and relative costs," The Steel Committee ofCalifornia, Walnut Creek and El Monte.

12. Halterman, A., and M.A. Aschheim (2002), "Reduced Web Section Beams: Experimental Findingsand Design Implications," Proceedings of the 7th US National Conference of Earthquake Engineering,Boston, Massachusetts, July 21-25.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 35

APPENDIX 1 - "PRE-NORTHRIDGE" CONNECTION

For this category of connection, the beam-to-column moment connection CF-1 is the base Rela-tive Cost Index 1.00 connection, with a singleshear plate being fillet welded to the columnflange. Beam flanges are fully welded to thecolumn flange, providing a very ductile and eco-nomical moment connection. Attaching the sheartab to the column with a full penetration weldrather than a double fillet weld increases the rela-tive cost 6%.

Reprinted from Steel TIPS, 1986. See Reference 11.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 36

AP

PE

ND

IX 2 - Q

UA

LIT

Y A

SSUR

AN

CE

FO

R P

RE

QU

AL

IFIE

DC

ON

NE

CT

ION

S

FEMA WELD INSPECTION RECOMMENDATIONS

Reference:FEMA 353 Part IITABLE 5-3, 5-4

Cost C

onsiderations for Steel M

oment Fram

e Connections,

Hassett and

Putkey, S

teel TIP

S, A

ugust 2002 37

The following table summarizes requirements outlined by FEMA 353 and referenced in FEMA 350 PREQUALIFIED connection details:

CONNECTION WELD_________QA/OC CATEGORY CATEGORY INSPECTION_______________NOTESTABLE 5-3 TABLE 5-4

2.1 WUF-B

2.2 WUF-W

2.3 FF

2.4 WFP

2.5 RBS-WW

RBS-BW

Flange weldsShear Tab weld

Flange weldsShear Tab weldWeb End weldWeb to Shear Tab weld

Flange weldsShear Tab weldWeb End weldWeb to Shear Tab weld

Flange Plate butt weldsFlange Plate side fillet weldsFlange Plate end fillet weldsShear Tab weldWeb End weldWeb to Shear Tab weld

Flange weldsShear Tab weldWeb End weld

Flange weldsShear Tab weld

AH/TBL/T

AH/TBM/TBH/TBL/L

AHTBH/TBH/LBH/L

AH/TBH/LBH/TBH/TBM/TBH/L

AH/TBM/LBM/L

AH/TBL/T

13

1213

1111

111121

122

13

MT,UT 100% CJP'sUT 10% CJP's MT10% FILLETS

MT,UT 100% CJP'sMT 25% PJP's, FILLETSMT,UT 100% CJP'sMT 10% FILLETS

MT,UT 100% CJP'sMT,UT 100% CJP'sMT 25% FILLETSMT 25% FILLETS

MT,UT 100% CJP'sMT 25% FILLETSMT 25% FILLETSMT 25% FILLETSMT 25% PJP'sMT 25% FILLETS

MT.UT 100% CJP'sMT 25% FILLETSMT,UT 100% CJP's

MT.UT 100% CJP'sUT 10% CJP's MT10% FILLETS

NO REDUCTIONSMT 6" SPOT RANDOM

NO REDUCTIONSFULL LENGTHREDUCTION APPLIESMT 6" SPOT RANDOM

NO REDUCTIONSREDUCTION APPLIESPARTIAL LENGTHPARTIAL LENGTH

NO REDUCTIONSPARTIAL LENGTHFULL LENGTHFULL LENGTHFULL LENGTHPARTIAL LENGTH

NO REDUCTIONSPARTIAL LENGTHPARTIAL LENGTH

NO REDUCTIONSMT 6" SPOT RANDOM

2.6 BUEP

2.7 BSEP

2.8 BFP

2.9 DST

ALL

ALL

AH/TBM/L

AH/TBM/LBM/LAH/T

AH/TBM/L

12

1221

12

BM/TBL/L

BL/L

MT,UT 100% CJP'sMT 25% FILLETS

MT,UT 100% CJP'sMT 25% FILLETSMT,UT 100% CJP'sMT,UT 100% CJP's

MT,UT 100% CJP'sMT 25% FILLETS

MT, UT 100% CJP'sMT 10% PJP's, FILLETS

MT 10% PJP's, FILLETS

NO REDUCTIONSPARTIAL LENGTH

NO REDUCTIONSPARTIAL LENGTHPARTIAL LENGTHNO REDUCTIONS

NO REDUCTIONSPARTIAL LENGTH

REDUCTION APPLIESMT 6" SPOT RANDOM

MT 6" SPOT RANDOM

NOTE:

REDUCTION:PARTIAL LENGTH:

REDUCE INSPECTION TO 25% IF REJECTION RATE IS LESS THAN 5% AFTER 40 WELDS FOR A GIVEN WELDERFOR WELDS OVER 24 INCHES LONG, TEST 6 INCHES ON EACH END AND 6 INCHES ALONG THE LENGTH AT STARTS& STOPS

Cost C

onsiderations for Steel M

oment Fram

e Connections,

Hassett and

Putkey, S

teel TIPS

, August 2002

38

Flange weldsWeb weld

Flange weldsWeb weldStiffener weld to beamStiffener weld to end plate

Flange Plate butt weldsShear Tab weld

NO WELDING

CONT'Y PLATESFlange WeldsWeb WeldDOUBLER PLATEAll welds

23

3

About the Authors

Patrick M. Hassett received his Bachelor's degree in Civil Engineering in 1983, and a Master'sdegree in Structural Engineering and Structural Mechanics in 1985 from The University ofCalifornia at Berkeley. He is a licensed Civil and Structural Engineer in California and is alsolicensed in Illinois and Missouri. He has fifteen years experience in the fabrication and erectionof major structural steel construction projects. He is a member of SEAONC, and ASCE, is pastchairman of the Structural Steel Educational Council, has served on the SAC Joint VentureProject, and currently serves on the AISC Seismic Design committee.

He is currently running his own consulting firm in Castro Valley, California, primarily serving theengineering needs of steel fabricators and erectors. Among his recent projects is the connectiondesign and erection engineering on the 54-story Torre Mayor Tower in Mexico City. Morerecently, he engineered the steel erection procedure on the Walt Disney Concert Hall in LosAngeles.

James J. Putkey is a Consulting Civil Engineer in Moraga, California. He received a Bachelorof Civil Engineering degree from the University of Santa Clara in 1954. After two years in theU.S. Army, 19 years with the Erection Department of Bethlehem Steel Corporation—PacificCoast Division, and seven years with the University of California—Office of the President, hestarted his own consulting business. He has provided consulting services to owners,contractors, attorneys, and steel erectors for the past 20 years.

Mr. Putkey is now "Semi-Retired." He serves as a hearing officer for the University of California-Office of the President, and occasionally writes or helps write Steel TIPS.

Cost Considerations for Steel Moment Frame Connections, Hassett and Putkey, Steel TIPS, August 2002 39

STRUCTURAL STEEL EDUCATIONAL COUNCIL

TECHNICAL INFORMATION & PRODUCT SERVICE

APRIL 1995

Structural Details to IncreaseDuctility of Connections

By: Omer W. Blodgett, P.E.Senior Design ConsultantThe Lincoln Electric Company

INTRODUCTION

Materials used in steel structures are increas-ingly becoming thicker and heavier. A greaterchance of cracking during welding of beamsto columns, for example, has resulted due toincreased thickness of material. With weldshrinkage restrained in the thickness, width,and length, triaxial stresses develop that mayinhibit the ability of steel to exhibit ductility.This paper will attempt to explain why thesecracks may occur, and what can be done toprevent them, by expanding on informationpresented in the AISC Supplement No. 1(LRFD) or Chapter J 9th Ed. AISC Manual.

Ostress

psi

I I I I I I

£ strain in/in

Figure 1

FIELD RESULTS

I learned about the stress-strain curve (Figure1) while taking "Strength of Materials" alongwith laboratory work at the University of Min-nesota. It took me a long time before I real-ized that this applied only to simple tensilespecimens in the laboratory.

During World War II while I was working in ashipyard, a docked, all-welded tanker, theSchenectady, suddenly broke in two. At thetime, we had no answer as to what could havecaused such a catastrophic failure. We passedit off as perhaps a poor grade of steel or poorworkmanship, and kept on welding our ships.A short time later, we received a bulletin fromThe Lincoln Electric Company in which it wasstated that ductility values come from simpletensile specimens which are free to neck down.The bulletin pointed out that if the same platehad many transverse stiffeners welded to it,the ability to neck down would be greatly re-stricted, and the plate would fail with less ap-parent ductility.

DEFINING DUCTILITY

In Figure 2, Mohr's Circle of Stress has beendrawn, showing a tensile stress of 10 ksi up tothe ultimate of 70 ksi (numbered from 1 to 7).

The corresponding maximum shear stress is atthe top of each circle. For convenience, eachpoint of shear (illustrated as a solid dot) is movedhorizontally until it lies directly above the corre-sponding tensile stress (depicted as an opendot). Notice that these points form a straight line,representing a simple tensile specimen. Fromthis line, it is possible to read off the maximumshear stress for a given tensile stress. This isthe basic f igure used by Professor Gensamer,as shown in Figure 3.

Gensamer introduced the concept of graphicallyillustrating the maximum shear-stress theory offailure. In Figure (4), the horizontal axis repre-sents the tensile stress (o), and the vertical axisrepresents the shear stress (-[). The critical ten-sile stress would be the ultimate tensile strength,but exceeding this value causes immediate fail-

crit icat shear ? Z•c/stress •r o

7 tensilestress(ksi)

30'

20'

10

Figure 2 ,'•I 0

/ ia

: : t I I I II 2 3 /4 5 6 7

tensile stress (ksi)

Figure 3

40 II

--30 .•, ,,,•th ..· • I •., I

v, crlticat shear stress t • ":1

· , • n i •.•.,•"• •¢- ; '• 1•. " / ,•,<'• - ! •i ,-

, ,

10 2'0 30 40 50 60 70applied tensile stress (ksi)

Shear Tensile Relationship for a Simple Tensile Specimen

Figure 4

ure with little energy absorbed, that is, brittle fail-ure. Exceeding the critical shear stress causesslip or ductile failure. In a simple tensile speci-men, the resulting shear.stress is one half ofthe applied tensile stress. This means movingup along a line having a 26.56° slope. This slopeis not dependent upon the type or strength ofsteel used. When this specimen reaches the-yield strength (ay), the corresponding shem•stress is at its critical value (•:CR)' This meansthe critical shear (•:CR) is equal to one half of theusual yield strength of the material in a simpletensile test. Above the critical shear value, plas-tic deformation takes place, with the specimenslipping along millions of 45 ° slip planes. Dur-ing this time, there is some work hardening ofthe material. Finally, the critical tensile stress isreached, and failure occurs.

4sot.-. •,' • I

'•=1/2o

•-- L = W q

4 s o - • , f I

i v:, i ®I f ' •' f I

©

I L - W I

necked down tensilo specimen

Ductility of Steel

Figure 5

2 Steel Tips April 1995

In Figure 5a, the member is subjected to a ten-sile stress (o) under the yield strength (ay). Thisresults in elastic strain and is recoverable whenthe stress is removed. Notice also in Figure 5athat a shear occurs which has a maximum valueof •:--1/20 on a plane at 45°, with the axis of theapplied tensile stress. If the applied stress (o) isincreased to a value of (0¥), the resulting shearstress exceeds its critical value-tcR=l/2ov , then apermanent slip occurs on planes at 45°, asshown in Figure 5b and 5c.

This is plastic strain and, if continued, will causethe specimen to neck down (Figure 5d). As thecross-sectional area continues to becomesmaller, the tensile stress finally exceeds thecritical normal stress (tensile strength) and themember fails.

All of this can be seen in the stress-strain curveof Figure 6. Region (a) below the yield strengthcovers the elastic strain portion. Region (c) cov-ers the plastic strain portion with the membernecking down. Point (d) is tensile failure. In thestress-strain curve of Figure 6, region (a) is allelastic strain. The resulting shear stress (•) isunder the critical value -OcR=l/2 ay, so no plasticstrain takes place.

In region (c) the resulting shear stress exceedsthe critical value and plastic strain takes place

with more and more necking down. The ductilityof the simple tensile stress specimen occursbecause there is a shear stress component fromthe particular load condition and, more impor-tantly, it exceeds the critical value by a consid-erable amount. Let us see if we can find out whythis test specimen is ductile; then we can checkthe ductility of other loaded members or details.

The ductility of a simple tensile specimen oc-curs because there are two shear stresse (%-3)and ('•2-3) resulting from the applied tensile stress(o3), as shown in Figure 7a and 7b. Notice thatwhen the stress (%) reaches its critical valuefor failure (70 ksi in this example), the two shearstresses have already exceeded their criticalvalue of 20 ksi. There are two shear stressesbecause there are two circles: circle (1-3) andcircle (2-3). The third circle, (1-2), has no ra-dius, and hence no shear stress, since it is apoint.

=zer

tension

g

w 'ri '3 el

r--TT7--- , _ .%!0 ' 2 •0'3

Figure 7A

-- 6aT•

e 3Of4t.= 20

lo! - - I ' I - -

.oi .02 .03' i I I •'

.G4 .OS .Q6 .107 ,.(}6Total $tra;n E Tn/in

: ; ,I '.09 .10 .11 .t2

,G$ rGt. e 1-3

Ptqstic , •movement' • / ' I

m4ke;• Sl)•g;men I • I

r ° w e O . • t = zen=

i3(l-i)

aircte 2-3

Z-3)

I .Thismqvernent •c.i.3)ioir lthedir•ion Of (•3

PtasticmYJvement

makes s p4K:imenthinner

This movement •'a(z.alis ;n the Idirection af 0'3 I

tatar ptost;¢ strain ]n d;rectlon of

•JZ •3 (I-3) '[' •'3 C 2-3} Ifram fram•t-3 'rz-3

wiU tend to reduce the residual stres.• ((T3)

Figure 6 Figure 7B

Steel April 1995 3

Any value of shear for 1;1. 3 and 1;2-3 above thecritical 20 ksi will cause plastic strain. Notice inFigure 7a that both circle (1-3) and circle (2-3)cause plastic strain $3(1-3) and £3(2-3)' Therefore(o3) will be: •3 = E3(1-3) + s3(2-3).

Since E3(1.3) ' - £3(2-3)' we then have: s3 = 2s3¢.3),which will tend to reduce residual stressescaused by welding.

If the specimen is pulled to failure, o3 will reachits critical value, or tensile strength; see Figure8. By this time, the two shear stresses are abovethe critical value and plastic strain or movementwill have taken place. Notice that the total plas-tic strain consists of two values: C3(1.3) and £3(2-3)'The movement •3 acts in the direction of the

40;

t..-

crlticot shear • •\• I!

lr'cr_, s toad tine It.'•'} r e p r e s e n t s I

{ • O ' I II

; : .* I I I I10 20 30 40 50 60 70

normqt stress 0'3

Figure 8

I 2 3 4normal elastic t o t a l plasticstress strain strain strain

0 3 £e •T Ep

10 .00033 .0003315 .00050 .0005020 .00067 .0006725 .00083 .0008330 .00100 .0010035 .00117 .0011740 .00133 .0013345 .0015 .00236 .0008650 .0017 .00485 .0031555 .0018 .01200 .0102060 .0020 .03185 .0298565 .0022 .08234 .0801470 .0023 .20229 .20000

Table 1

stress o3 and would tend to reduce any residualstress. This member should behave in a ductilemanner. Plastic behavior takes place from 03 = •*40 ksi up to 70 ksi, and is caused by two differ-

ent plastic strains E3(1.3) and £3(2-3)'

Table 1 lists the data from a typical stress-straincurve for structural steel. Total elastic plus plas-tic strain is listed in Column 3. The elastic strain,calculated from s = °/E, is listed in Column 2. Bysubtracting the elastic strain from the corre-sponding total rain, we obtain the plastic strain,shown in Column 4.

RESIDUAL STRESSES ISOLATED

Figure 9 illustrates that two important residualstresses exist in the weld's termination zone. Thebutt joint in the flange has a residual stress lon-gitudinal to the length of the flange (o3), as wellas a stress transverse to the flange (%). Longi-tudinal stress is tensile along the center line ofthe flange where the weld access hole termi-nates. It can be compared to tightening a steel • '

weld (accesshole

•roG¥,ew e l d

° o4 •_3

Figure 9

4 Steel Tips April 1995

30 '•'•I u : criti•t shear st•s, . . . . . . . . . -

· u t t' 2 _ :

10 2O 30 &O SO 60 70 80

Figure 10

cable lengthwise in the center in tension, withcompression spread out on both sides. Thetransverse stress (Ol) is tensile in the weld zone,including a portion of the adjacent plate, goingthrough zero, and then compression, beyond theadjacent plate. This transverse stress (o•) is alsosimilar to tightening a steel cable.

RESIDUAL STRESSES APPLIED

These residual stresses may be applied to aweld detail having a narrow weld access hole,as shown in Figure 10. This hole terminates ata point where (0•) and (03) are in tension. Sincethe web at the edge of the weld access holeoffers some restraint against movement in thethrough thickness direction of the flange plate,stress in the (02) direction may have an appre-ciable value. All of the circles will be small. Nei-ther (T2.3) nor (T•.3) will probably ever reach thecritical shear stress value, and plastic strain orductility will not occur, as the right hand of Fig-ure 10 illustrates.

If the weld access hole can be made wider, asrecommended by AISC Specification, Ninth Edi-tion, so that it terminates in a zone where thetransverse residual stress (0•) is compressive(see Figure 11), then a more favorable stresscondition will result in greater ductility in the (03)direction. In this case, shear stress (-c•.3) will be

high as shown on Mohr's Circle of Stress, andthe critical shear value will be reached at a muchlower tensile stress or load value. This will pro-duce more ductility in the (o3) direction, greatlyreducing the chance of a transverse crack in theflange at the termination of the weld access hole.

EXAMPLE #1

Consider the unrestrained section, similar to asimple tensile specimen, shown in Figure 12.When there is no applied stress (o•) in thethrough thickness direction or (o2) across thewidth, these values are zero. This will producethe largest of Mohr's circles, and the greatestvalue of shear (•;2.3) and ('•-3)' In both cases,these shear stresses are equal to one half ofthe applied tensile stress (o3). These two shear-tensile lines are drawn in the lower portion ofthe figure. Although there are two lines, whichwould indicate good ductility, there is a differ-ence between the two. One line represents neck-lng down through the thickness, and the otherrepresents necking down across the width. Al-though the unit strains are the same in this case,the strain acting across the width would resultin greater overall movement or elongation overthe length of the specimen.

'o, . .'F 0 / . I cr;tiCOt .sh•_ Str•_•

- --(k•i) 2 f - ; ? - - - : - -•- -•.,--"2d'k,; -

Figure 11

Apri11995 5

In the case of the restrained section shown inFigure 13, (here L=W/4), the angle of maximumshear stress lies along an angle of o - 76° andthe resulting shear value is •: -- .23 03. The twoshear values (%-3) and ('¢2-3) produce two shear-tensile lines. The lower line acting across thewidth does not produce enough shear to exceedthe critical value, hence no plastic yielding. Theupper line indicating good plastic yielding, how-ever, acts only through the thickness and theoverall movement would be less than the ex-ample on the right. To get a better picture of thisbehavior, the stress-strain curve shown in Fig-ure 14 has been created for the two details.

' , - : : r I t - I . .

I I I I

TO

,,• sa

? o41t•

$ 0

.01 .02 .03 .04 .05 .OG .07 •14 .09 JO .!1 .12 J ) .14 .lis Jll J? J!) .2Qunit strO)n in/in

Figure 14

EXAMPLE #2

There has been some discussion about the weldconnecting the beam flange to the column flangeas being brittle. Referring to Figure 15, the ma-terial at point (A), whether it be weld metal orbase metal it cannot exhibit the ductility of a

J_

necking dcTw• thru thickness· ]

necking • ocross wk• neckb•J down thru tl•cknesS· 2

necking down across widthlrZ.3

Ial

40'

• 30.

10'

I

· . '•,.,ec•' ' •;b

10 20 30 40 SO 60 70tensite stress (ksi)

40¸

v

2o

10

i.

I

Hti• shear stress _ •', . . . . . . . - -- - • . . . . . . . . . . •l

qr'cr:2O ks, • ( • Y . . . . . . . . . . •i '•

/ •!•

10 20 30 40 5• GO 70tensite stress (ksD

Figure 12 Figure 13

6 Steel April 1995

simple tension test. Ductility can only take placeif the material can slip in shear along numerousslip planes. Four conditions are required for duc-tility:

1. There must be a shear stress (•) componentresulting from the given load condition.

2. This shear stress must exceed its criticalvalue by a reasonable amount. The more itexceeds this value, the greater will be theresulting ductility.

3. The plastic shear strain resulting from thisshear stress must act in the direction whichwill relieve the particular stress which cancause cracking.

4. There must be sufficient unrestrained lengthof the member to permit "necking down."

If conditions (1) and (2) are not met, there willbe no ductility and no yield point. The stress willsimply build up to the ultimate tensile strengthwith little or no plastic energy absorbed. We callthis condition a brittle failure.

Figure 15 shows two regions in question:

Figure 15

I

Point (A) at the weld joining the beam flange tothe face of the column flange. Here there is re-straint against strain (movement) across thewidth of the beam flange (•) as well as throughthe thickness of the beam flange (s2)-

Point (B) is along the length of the beam flangeaway from the connecting weld. There is no re-straint across the width of the flange or throughits thickness.

Figure 16 shows the three equations for straingiven in most strength of material texts, shown

O2

(•1 O3

1•3 = "E (O3-•.[%-[[,[(•1)

1•2= • (-•%+%-•o•)

1s,= -• (-•%-•%+o•)

or it can be shown that

E [•E3+•2+(1-•)•,]o•= (l+tz) (1-21z)

E [•3+(1-!Z)•2+[Lt•]02= (1+•) (1-2•)

E [(1-tZ)•3+•2+p.•]03= (1+[•) (1-21Z)

Figure 16

steel Tips April 1995 7

- - - r • I1' IIr I ... • , , • : , . - • . ; - • - , • v, • +

02=0

ai,

!

!

!L.

• - °

ol=O

o3=30ksi

· £3=+.001

•.'

E [tz%+•ze2+(1-•)e l]%= (1 +lz) (1-2tz)

30000o,- (1 +.3) (1-.6) [.3(+.001 ) +.3(-.O003)+.7(-.O003)]=Zero

in the flange. By using Poisson's ratio of iz=0.3for steel the following strains are found for asimple tensi le specimen when stressed too3=30ksi.

£3 = +.001

s2 =-.0003

s• =-.0003

Using these strains in the three formulas for re-sisting stresses we find:

o• = Zero

02 = Zero

o3 = 30 ksi

o2=% =Zero

E [(1-p,)£3+$S:+!,te,]°3= (1 +It) (1-21Z)

30000°3:- (1.3) (1-.6) [.7(.001)+.3(-.0003)+.3(-.0003)]=30.0 ksi

ac1.3=35

:

Figure 17

0.•=70

in upper box. For our use, these have been con-verted into corresponding equations for stress,shown in lower box.

Figure 17 is an element of the beam flange fromFigure 15 point (B). There is no restraint(%+o2=0) against the 30 ksi longitudinal stress

This is plotted as Mohr's circle of stress in adotted circle. The larger solid line circle is for astress of 70ksi or ultimate tensile stress. Theresulting maximum shear stresses (%-3) and(-c2.3) are the radii of these two circles or 35 ksi.The ratio of shear to tensile stress is 0.5. Figure18 plots this as line (B). Notice at a yield point of55 ksi, the critical shear value is 1/2 of this or27.5 ksi. When this critical shear stress isreached, plastic straining or movement takesplace and ductile behavior will result up to theultimate tensile strength, here 70 ksi. Figure 19shows a predicated stress-strain curve indicat-ing ample ductility.

Figure 20 shows an element from Point (A) (Fig-ure 15) at the junction of the beam and columnflange. Whether we consider weld metal or thematerial in the column or beam makes little dif-ference because this region is highly restrained.Suppose we assume:

e3 = +.001 (as before)

•2 = Zero '•. (but now highly restrainedE1 = Zero J with little strain)

From the given equations, we find the followingstresses:

8 Steel Tips April 1995

30 ·27.5

. . . h e

m 20fJ•

tllID

I 0

critical shear stress 'c=1/2 o • b o y.Y•'t.•J,

";'•'•'•, • iv , i ,

10 20 30 40 50 55 60 70

Figure 18

% = 17.31 ksi ) Increase to ( = 30.0 ksi

(72 : 17.31 ksi • ultimate tensile I, = 30.0 ksi

(73 = 40.38 ksi strength = 70.0 ksi

The lower portion of the sheet is a plot of Mohr'scircle of stress. The maximum stresses are:

-c•. 3 = 1:2.3 = 20 ksi

The ratio of shear to tensile stress is 0.286. InFigure 18, this condition is plotted as line (A).Notice it never exceeds the value of the criticalshear stress (27.5 ksi); therefore, there will beno plastic strain or movement, and it will behaveas a brittle material. Figure 19 shows a predi-cated stress-strain curve going upward as astraight line (elastic) until the ultimate tensilestress is reached in a brittle manner with noenergy absorbed plastically.

Would it help if the strength and ductility of theweld metal or base metal were changed? SeeFigure 21. The top figure is for lower strength,more ductile steel, tensile strength of 60 ksi anda yield strength of 40 ksi. The lower figure is fora higher strength, lower ductile steel, tensilestrength of 70 ksi and a yield strength of 55 ksi.Notice in the case of no restraint (B) that thelower strength material will result in more ductil-ity. However, in the real world where there is

80

70

60

50(/3

40(/3

30

20

10

.05 .1 ,15strain in/in

Figure 19

.2 .25 .3 .35

Steel Tips April 1995 9

t

v

restraint (A), the lower strength material doesnot provide any help against cracking. Neithermaterial will provide any ductility. It might be ar-gued thatthe higher strength material (lower fig-ure) would be stronger. It still will perform in abrittle manner if over stressed.

Assuming we have good workmanship with nodefects or stress raisers, the real success of thisconnection will depend upon getting the adja-

( 7 1 =

0'2

l!ii

ii ·ilI

(J1

E [g£3+g•2+(1-•L)•,](1+•) (1-2•)

- • 03i

,,' E3=+.001

30,000°1- (1+.3)(1-.6)

o2=0•=17.31 ksi

O 3 =

30,000°3= (1.3) (1-.6)

[.3(+.001)+Zero+Zero]=17.31 ksi or 30.00 ksi

E [(1-ILL)E3+ -(1+•) (1-2•)

- - [(1-.3)(.001)+Zero+Zero]= 40.38 ksi or 70.00 ksi

1• 1.3=20

/ , / " - ' " • O 3 = 7 0

Figure 20

30

20(/)(/)

'• 10

ID

Steel ITS=60 ksi IYP=40 ksi I.

· = i ,10 20 30 40 50 60 70

tensile stress (o) ksi

SteelTS=70 ksiYP=55 ksi

30 .'• 27.5.

• 20

O'J

IDU'J

1;=1/2 o¥=27.5 ksi

• •' , , , i !10 20 30 40 50 55 60 70

tensile stress (o) ksi

Figure 21

cent beam to plastically deflect before this criti-cal section cracks.

C O N C L U S I O NThe way in which a designer selects structuraldetails under particular load conditions greatlyinfluences whether the condition providesenough shear stress component so that the criti-cal shear value may be exceeded first, produc-ing sufficient plastic movement before the criti-cal normal stress value is exceeded. This willresult in a ductile detail and minimize the'chances of cracking.

10 Steel Tips April 1995

I

REFERENCES

AISC Supplement No. 2, January 1, 1989. Tothe Specification for the Design, Fabrication &Erection of Structural Steel for Buildings.

Bjorhovde, Brozzetti, Alpsten and Tall. "ResidualStresses in Thick Welded Plates," AWS Weld-ing Journal, August 1972.

Blodgett, Omer W. Weight of Weld Metal, TheJames F. Lincoln Arc Welding Foundation, Bul-letin D417, April 1978.

Estuar and Tall. "Experimental Investigation ofWelded Built-Up Columns," AWS Welding Jour-nal, April 1963.

Gayles and Willis. "Factors Affecting ResidualStresses in Welds," AWS Welding Journal, Au-gust 1940.

Gensamer, Maxwell. "Strength of Metals UnderCombined Stresses," American Society of Met-als, 1941.

Parker, Earl R., Brittle Behavior of EngineeringStructures, John Wiley and Sons, Inc., 1957.

Shanley, F.R., "Plastic Strain--Combined Load-ing,'' Strength of Materials, McGraw-Hill BookCo., 1957; Chapter 11.

Steel Tips April 1995 11

STRUCTURAL STEEL EDUCATIONAL COUNCIL

TECHNICAL INFORMATION & PRODUCT SERVICE

AUGUST 1997

Dynamic Tension Tests of

Simulated Moment Resisting

Frame Weld Joints

byE.J. Kaufmann

ATLSS Engineering Research CenterLehigh University

TABLE OF CONTENTS

I°

II.

III.

INTRODUCTION

TEST SPECIMENS

MECHANICAL PROPERTIES

Page

1

2

4

IV. TEST PROCEDURE

V. TEST RESULTS 7

VI. SUMMARY 8

VII. REFERENCES 9

APPENDIX A - Test Results

APPENDIX B - Weld Cost Comparisons

Index of Steel Tips PublicationsThe following is a list of available Steel Tips. Copies will be sent upon request. Some are invery limited quantity.

· Seismic Design of Special Concentrically Braced Frames· Seismic Design of Bolted Steel Moment Resisting Frames· Structural Details to Increase Ductility of Connections· Slotted Bolted Connection Energy Dissipaters· Use of Steel in the Seismic Retrofit of Historic Oakland City Hall· Heavy Structural Shapes in Tension· Economical Use of Cambered Steel Beams· Value Engineering & Steel Economy· What Design Engineers Can Do to Reduce Fabrication Costs· Charts for Strong Column Weak Girder Design of Steel Frames· Seismic Strengthening with Steel Slotted Bolt Connections· Seismic Design of Steel Column-Tree Moment-Resisting Frames· Dynamic Tension Tests of Simulated Resisting Frame Weld Joints

I. INTRODUCTION

Under the SAC Phase I program a pilot project was conducted to develop and evaluate arelatively simple and inexpensive test specimen for studying moment frame weld jointperformance (Ref. 1). A specimen was designed to simulate the behavior of a single beamflange-to-column flange weld joint which could be tested in a large capacity universal testmachine in tension under dynamic loading rates similar to earthquake loadings. The testspecimen concept is illustrated in Figure 1.

The results of the pilot test program showed that weld joints fabricated using electrodes withhigher notch toughness than the E70T-4 electrode in common use prior to the Northridgeearthquake, such as E7018, E70TG-K2, and E71T-8, performed satisfactorily in the test. Thiswas also in conjunction with improved detailing including removal of weld backing and weldtabs and adherence to D I.I welding code procedural requirements. The results indicated thatbrittle fractures initiating in the weld metal, as occurred in ETOT-4 welded connections, could beavoided when weld metal with a minimum CVN impact toughness requirement of 20 ft-lbs @-20F was used. Although only axial tension loads were applied in the test the results closelyparalleled the performance of similarly fabricated weld joints in full-size connection tests (Ref.2) and appeared to provide a viable means of assessing weld metal toughness requirements formoment frame applications.

To expand the test database to include other currently available flux-cored electrodes theStructural Steel Educational Council of the California Field Ironworkers Administration Trustsponsored additional testing, reported herein, to evaluate the performance of weld joints weldedwith other electrode types as well as duplicate tests performed in the pilot study. Eight testspecimens were fabricated by a commercial fabricator in California using three currently availableflux-cored electrodes (E70T-6, E70T-7, and E71T-8) and one shielded metal arc electrode(E7018). The fabricated specimens were then shipped to the ATLSS Engineering ResearchCenter at Lehigh University for testing.

Figure 1 Simulated Beam-Column Tension Specimen

II. TEST SPECIMENS

A sketch of the simulated beam flange-to-column flange weld joint test specimen is shownin Figure 2. The column element consists of an 8 in. length of a W14x176 wide flange shape(A572 Gr. 50) with one flange removed. A l"x 6" plate (A36) is groove welded to the columnflange face to simulate the beam flange-to-column flange connection. A slotted pull plate iswelded to the column web to permit the the assemblage to be gripped in a universal test machineand loaded in tension at static or dynamic loading rates. A simulated coped beam web plate istack welded to the beam plate to introduce welding restrictions similar to welding the bottomflange of a moment connection. The web plate was removed after welding to facilitateinstallation of test instrumentation.

Duplicate specimens were prepared using each of four electrode types. Table 1 gives a summary ofthe welding procedure parameters. In all eight specimens the weld tabs and backing were removedafter welding and a reinforcing fillet weld was applied to the weld root. All simulated beam flangeto column joints were fabricated to a 3/8" root and a 30° included angle.

BothSides 5/8 ....

40"

< I

5/8 '• Cf/

12"

•'""--'->-< Tack

•_ 16"

I • • N • 8 , , ix <CJP

W 1 4 x 176 • 6"x1"A572 Gr. 50 (A36/

(one flange removed)

Figure 2 Simulated Be Flange-to-Column Flange Test Specimen Design.

<

<I

m

B-Z

tm

r-

©

p..

,_ • '- co o

E• z o_ o_ z•'•-' IJJ W IJJ UJ

UQ.

- IE•.E[ o o < o I o < o o < o o <

- E

II.

i

cg • < co < to < o eg <

,•_ • ,_ '- d• Z co co

-- I

u u ,- ,- ,- Z • Z ,- ,- ,.• cj c••'• .,- C•

I

II

0 c'") • -i. ..• cO u'• cO u'•

*"=' esi c'• ("qi

o • c• • o • e• 4 e• e• c5 •6 t-q (,4 04 e- e'• • Oq •--

Z

= _.•_ to to eg otO u• 0 0 I.• (•q 0

o P G •- P o o

--• w w w ww

co

o

o.

8'13(D

0

0-

O•C'0Q.I

II

c•c

Co

-oc

E

o

c

II

3

IH. MECHANICAL PROPERTIES

The mechanical properties of the beam flange plate and column shape were determined aftertesting. Unyielded material located at the end of the grip length of the beam plate was used tofabricate standard 0.505 in. dia. tensile specimens. Standard 0.505 in. dia. specimens were alsofabricated from the W14 x 176 column flange at the ASTM A673 test location. Table 2 givesa summary of the base material properties. The stress-strain behavior of the two materials isshown in Figure 3.

Table 3 gives AWS required and typical mechanical properties for the filler metals used.Mechanical properties of E70T-4 filler metal is also included for comparison. E70T-6, E71T-8and E7018 filler metals have a required Charpy V-notch (CVN) impact toughness requirementof 20 ft-lbs @ -20F. E70T-4 filler metal does not have a notch toughness requirement andtypically provides 5-15 ft~lbs @ +70F. E70T-7 filler metal, like E70T-4, also has no AWSminimum toughness requirement. Procedure qualification tests using E70T-7 weld metal hasindicated a notch toughness intermediate to E70T-4 and the higher toughness filler metals. Thesetests provided an average toughness of 8 ft-lbs @ OF. CVN test data for the various filler metalsis shown in Figure 4.

TABLE 2 MECHANICAL PROPERTIES

Y.S. T.S. Elong.(2") R.A.(ksi) (ksi) (%) (%)

W14 x 176 (Column) 56.6m 75.2m 38.6¢) 77.1¢)ASTM A572 Gr. 50

Beam Flange Plate 42.9 73.5 28.4 59.3ASTM A36

I) Standard I-lange locationIINI-

,'• ............................................................... , • - - A 5 7 2 • . • - ( G o • } - ,

'2m 5li-

t.

tiiiiiiiiiiiiiiiiiiiiiiiiiiiiiill . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21 . . . . . . . . . . . . . . * . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

i, 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . :!!!!!!!!!!!!!!!!!!!!!!!::::::: ............................................(,, , . ,

0 Il. 1 0.2 0.3 0.4

Strain

Figure 3 Material Stress-Strain Behavior.

4

TABLE 3WELD METAL PROPERTIES

AWS Required Typical

Y.S. U . T . S . Elong. CVN Y.S. U.T.S. Elong. CVNksi ksi % fi-lbs ksi ksi % fi-lbs

E70T-4 60 min. 72 min. '22 min. .m 60-70 80-95 15-25 5-15@ +70F

E70TG-K2 58 min. 70-90 22 min. -(• 70-75 85-90 2 5 - 3 0 20-40@-20F

E71T-8 60 min. 72 min. 22 min. 20 min. 65-75 70-90 2 5 - 3 0 20-70@-20F @-20F

E70T-6 60 min. 72 min. 22 min. 20 min. 65-75 70-90 2 5 - 3 0 25-75@-20F @-20F

E70T-7 60 min. 72 min. 22 min. -¢) 60-65 80-90 22-26 5-10@ OF

E7018 58 min. 70 min. 22 min. 20 min. 65-75 75-85 2 5 - 3 0 90-120@-20F @-20F

1. No Requirement2. No Requirement, will meet 20 ft-lbs @-20F3. From manufacturer or laboratory tests

O9

m

>•(.5

U.IZU.I

LLI

OO3

-95200

[ ]

160 -- •,

O

120 --X

80 --

40 --

0-' 40

TEMPERATURE, °C

-55 -15 25 65

I I I IE70T-4

E70TG-K2 ·

E71T-8 ·

E7018E7OT-7

O

O Oo o 4•.

O O Z•

· 8 o •

· [ i a I g B I-80 -20 40 100 160

TEMPERATURE, °F

105

I

I220

Figure 4 CVN test data for various filler metals.

240

200

160

120

80

40

0

Q•

::30

>.•(.5

ILlzILl

ILl

O(/3

95.D047

5

IV. TEST PROCEDURE

Specimens were tested in a PC-controlled 600 kip capacity universal test machine modified topermit dynamic load rates to be applied to the test specimen. With modification, a maximumcrosshead displacement rate of 0.15 inches/sec could be achieved. Dynamic tests were conductedin displacement control at the maximum crosshead displacement rate. In addition to recordingcrosshead displacements, a 2 in. displacement range linear variable differential transformer(LVDT) was also installed to measure weld joint displacements over the ungripped length of thebeam flange (approx. 8 inches) relative to the column flange face. Figure 5 shows aninstrumented test specimen installed in the test machine. Load, crosshead displacement, andLVDT displacement data were recorded with a PC data acquisition system.

Test specimens were loaded to failure in a single tension load cycle applied in two ramp ratesegments. Initially specimens were loaded at a crosshead displacement rate of 0.05 inches/secto a load of 60 kips (-10 ksi in beam flange plate) to seat the grips after which the displacementrate was increased to 0.15 inches/sec to failure. A typical LVDT displacement-time plot for adynamic test is shown in Figure 6. Over the 8 in. gauge length the displacement rate in thevicinity of the weld joint corresponds to a strain rate of -0.02 sec't. This strain rate correspondedto about 1 sec. loading through the elastic range. In comparison, strain rates for static loadingare typically of the order of 0.001-0.0001 sec't. After testing the specimens were examinedvisually for evidence of cracking or to determine the fracture origin, mode of fracture, and crackpath.

Figure 5 Test specimen installed in 600 kip test machine.

6

"'-7'. 2

· ,.•,. 1.8

(-. 1.6Q )

1.4 _

CD 1.2O

GO

ch o6

0.4

0.2

• 1 o

F I t I i 1

i i

i i I i i / I I ' !

F t f 1 i

· ! , ¢ ' • I ,I I • t I i i

I i , • ii I I ! i

0 5 10 15 20 25 3O

Time (sec)

Figure 6 Typical LVDT displacement vs. time plot.

V. TEST RESULTS

A summary of the results of the eight tests is given in Appendix B. With the exception of thespecimens welded with E70T-7 filler metal, all other test specimens behaved similarly. Duplicatespecimens welded with E7018, E71T-8, and E70T-6 filler metals failed by ductile tensile failureof the beam flange plate. No visual evidence of weld joint cracking was detected in any of thesetest specimens. Absence of whitewash flaking on the weld metal surface and column flange facearea also indicated that little or no inelastic deformations developed in these areas.

The two specimens welded with ETOT-7 filler metal showed mixed performance. Test SpecimenNo. 5 failed brittlely in the beam flange plate after significant yielding occurred in the plate.Examination of the fracture indicated that localized ductile tearing developed at the weld toe ofthe beam flange plate prior to initiation of brittle fracture (see Test No. 5 photographs inAppendix B). The tearing appeared to follow the weld fusion line although it was not clearwhether the tear propagated in weld metal or in the heat-affected-zone (HAZ). The duplicate testspecimen (Test No. 6) failed by ductile tearing of the beam flange plate, however, evidence ofsub-critical tearing at the same weld toe location and also in the adjacent base material was alsoobserved (see Test No. 6 photographs in Appendix B).

The cause of the weld toe tearing is not entirely clear, however, examination of the fracture incross-section in Test No. 5 also revealed significant weld toe undercut and a steep transition ofthe top reinforcement weld bead which was not observed in the other test specimens. Theundercut in Test No. 5 was measured to be 0.08" in depth which just exceeds the 1/16" maximumpermitted by Di.1. No undercut was measured in Test No. 6 although a similar steep transitionof the weld reinforcement also existed. The weld toe undercut observed in Test Specimen No.

7

5 may have influenced tearing initiation at the weld toe. It is also noteworthy that the E70T-7specimens were welded with a higher heat input than any of the other specimens (88 Kj/in vs.28-56 Kj/in) which may have resulted in a softer HAZ than in the other specimens. Ductiletearing at the weld toe at beam flange tips has also been observed in cyclically loaded full-scaleconnection tests after extensive plastic deformation of the beam flange has occurred (Ref. 2).

The test results support the current SAC recommendation for weld metal used in critical jointshaving a minimum CVN impact toughness of 20 ft-lbs @ OF (Ref. 3). All test specimens weldedwith filler metals which exceeded this requirement (ie. 20 ft-lbs @ -20F) performed well underintermediate strain rate loading. Although the ETOT-7 specimens did not satisfy the 20 ft-lb @OF recommendation (Avg. 8 ft-lbs @ OF) and also exhibited brittle behavior in one test, there wasno clear indication that weld metal fracture was causal to the failure. Additional test data on thefracture toughness and weld joint performance of this weld metal would be helpful in definingminimum weld metal toughness requirements.

VI. SUMMARY

1. Eight simulated beam flange-to-column flange weld joint test specimens werefabricated using three currently available flux-cored electrodes (E70T-6, E70T-7, E71T-8)and one shielded metal arc electrode (E7018). Duplicate specimens welded with fillermetals having a minimum CVN impact toughness requirement of 20 ft-lbs @ -20F (E70T-6, E71T-8, and E7018) performed satisfactorily under dynamic loading conditions.Duplicate specimens fabricated using an E70T-7 electrode with lower notch toughness (8ft-lbs @ OF) also performed satisfactorily although premature brittle fracture developedin one test presumably due to excessive weld toe undercut. The test results provideadditional confu'mation that brittle fracture in moment frame weld joints can besuppressed through adequate levels of weld metal toughness in conjunction with improvedweld joint detailing (ie. removing weld tabs and weld backing).

2. The test results support the current SAC recommendation for weld metal used incritical joints having a minimum CVN impact toughness of 20 ft-lbs @ OF.

8

REFERENCES

1. Kaufmann, E.J., Fisher, J.W., "A Study of the Effects of Material and Welding Factors onMoment Frame Weld Joint Performance Using a Small-Scale Tension Specimen", SACTechnical Report 95-08, 1995.

2. Xue, Ming, Kaufmann, E.J., Lu, Lc-Wu, Fisher, J.W., "Fracture and Ductility of MomentConnections Under Dynamic Loading", Proceedings ASCE Structures Congress, Portland,Oregon, 1997.

3. Interim Guidelines: Evaluation, Repair, Modification and Design of Welded Steel MomentFrame Structures, FEMA 267, 1995.

9

APPENDIX A - Test Results

l0

Test No.: !Test Date: 8/27/96Weld Electrode: E7018

Test Description: Specimen weldedwith 1/8"¢ E7018. Weld tabs andbacking removed. Weld root re-inforcing fillet added. UTacceptable.

Test Result: No weld joint cracking.Specimen failed by ductile tensilefailure of the beam flange plate.

500

450

400

350

·• 300

"'•'250

G• 2000..J 15o

100

50

Test #1

F I i i,"" N

" ' I

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5

Crosshead Displacement (in)

11

Test No.: 2Test Date: 8/28/96Weld Electrode: E7018

Test Description: Duplicate ofTest gl. Specimen weldedwith 1/8"q• E7018. Weld tabs andbacking removed. Weld root re-inforcing fillet added. UT ....acceptable.

Test Result: No weld joint cracking.Specimen failed by ductile tensilefailure of the beam flange plate.

T e s t # 2

50O

45O

4OO

•350

· 300

"-" 250"0Cd 2000

__J 150

100

50

0

I ! i

h/ ¢ .,

, , ,

t

/ .... l0 0.5 1 1.5 2 2.5 3 3.5 4 4.5

C r o s s h e a d Displacement (in.)

12

Test No.: 3Test Date: 8/29/96Weld Electrode: E71T-8

Test Description: Specimen weldedwith 0.072"¢ E71T-8. Weld tabs andbacking removed. Weld root re-inforcing fillet added. UTacceptable.

Test Result: No weld joint cracking.Specimen failed by ductile tensilefailure of the beam flange plate.

Test #3

50O

450

4OO

• ' 350

300

v 2 5 0"0

2000

-..1 150

lO0

50

0

" ; "' Li

i i i

jrJ

/-.

iJ

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5

Crosshead Displacement (in.)

13

Test No.: 4Test Date: 8/29/96Weld Electrode: E71T-8

Test Description: Duplicate ofTest #3. Specimen weldedwith 0.072"• E71T-8. Weld tabs andbacking removed. Weld root re-inforcing fillet added. UTacceptable.

Test Result: No weld joint cracking.Specimen failed by ductile tensilefailure of the beam flange plate.

Test #4

500

450

400

•' 350Q.

-• 300

"• 250'ID('d 2000.._1 150

lO0

50

0

[. , . • m m''''-'• b..

S- f

! ,

iP

- J II t

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5

Crosshead Displacement (in.)

14

Test No.: 5Test Date: 8/29/96Weld Electrode: E70T-7

Test Description: Specimen weldedwith 7/64"¢ E70T-7. Weld tabs andbacking removed. Weld root re-inforcing fillet added. UTacceptable.

Test Result: Specimen failed bybrittle fracture of the beam flangeplate. Fracture initiated at the edgeof the beam flange plate from a localizedductile tear at the weld toe fromundercut. Fracture occurred at a load nerothe ultimate tensile strength of the beamplate. 0.93" LVDT displacement atfracture. No weld metal or column flangefracture detected.

Test #5

500

450

400

'•' 35O

-300

250%3(13 200o

...J 15o

100

50

f

J/ '

0 0.5 1 1.5 2 2.5 3 3.5 4

LVDT Displacement (in)

15

Test No.: 5Test Date: 8/29/96Weld Electrode: E70T-7

Top) Beam flange plate fracture surface Bottom) Enlarged view ofweld toe ductile tear

16

Test No.: 5Test Date: 8/29/96Weld Electrode: E70T-7

Top) Re-assembled cross-section of the fracture Bottom) Enlarged viewof the weld toe crack initiation location. Note the weld toe undercut.

17

Test No.: 6Test Date: 8/29/96Weld Electrode: E70T-7

Test Description: Duplicate ofTest #5. Specimen weldedwith 7/64"q• E70T-7. Weld tabs andbacking removed. Weld root re-inforcing fillet added. UTacceptable.

Test Result: Specimen failed byductile tensile failure of thebeam flange plate. Sub-criticaltears developed at the edge ofthe beam flange plate at the weldtoe (same location as Test #5)and in the base material adjacentto the weld toe. No weld metal orcolumn flange cracking detected.

Test #6

5OO

450

400

'G'350•L

-• 300

'• ' 250-O03 2O00

.._1 150

100

50

0

0.5 1 1.5 2 2.5 3 3.5

Crosshead Displacement (in.)4 4.5

18

Test No.: 6Test Date: 8/29/96Weld Electrode: E70T-7

Top) Sub-critical tearing of the beam plate at the weld toe andin adjacent base material. Bottom) Enlarged view of tears.

19

Test No.: 7Test Date: 8/30/96Weld Electrode: E70T-6

Test Description: Specimen weldedwith 3/32"¢ E70T-6. Weld tabs andbacking removed. Weld root re-inforcing fillet added.

Test Result: No weld joint cracking.Specimen failed by ductile tensilefailure of the beam flange plate.

Test #7

5OO

45O

4O0

'•' 350Q..• 300

•" 250'rJCd 200O--.I 150

100

50

0

] I - - L _ _

, / / •[

/ , II0 0.5 1 1.5 2 2.5 3 3.5 4 4.5

Crosshead Displacement (in.)

20

Test No.: 8Test Date: 8/30/96Weld Electrode: E70T-6

Test Description: Duplicate ofTest #7. Specimen weldedwith 3/32"4 E70T-6. Weld tabs andbacking removed. Weld root re-inforcing fdlet added.

Test Result: No weld joint cracking.Specimen failed by ductile tensilefailure of the beam flange plate.

Test #8

5OO

450

400

•'350

·• 300

250'0(13 200O

150

100

50

0.5 1 1.5 2 2.5 3 3.5

Crosshead Displacement (in.)4 4.5

21

APPENDIX B - Weld Cost Comparisons

22

COST COMPARISON

In order to provide the reader with a more complete picture, the Structural Steel Education Council(SSEC) has complied a cost comparison of the electrodes utilized in the Lehigh University tests. E 70T-4electrode which was used in earlier studies was requested by SAC. The 70T-4 electrode in the 0.120diameter was included because most estimating programs utilized that electrode as a basis for calculatingcomplete penetration costs in the fiat or horizontal position.

The cost comparison factors were determined by reviewing the cost data supplied by 3 of the erectormembers from the council and reviewed by the remaining members. They took the following factors intoconsideration when calculating those costs:

1. Cost of the Electrode

2. Labor to install the weld

3. Cost of equipment required to weld

It should be noted that the inefficiency costs of using additional welders in order to maintain a reason-able schedule was not included in the cost data, nor were the additional training costs associated with theuse off those electrodes not normally used for this application.

Cost Comparison Lehigh University Test Specimens

Manufacturer Process Manufacturer AWS AWS Notch Diameter Cost FactorDesignation Specifications Classifications Tough

Lincoln F C A W NR232 5.20 E71T8 Yes 0.072 3.1Lincoln F C A W NR305 5.20 E7OT6 Yes 3/32 2.0Lincoln F C A W NR311 ni 5.29 E70TG-K2 Yes 7/64 2.4Lincoln SMAW LH70 5.1 E7018 Yes 5/32 7.3Lincoln FCAW NS3M 5.20 E7OT4 No 0.120 1.0Lincoln F C A W NR311 5.20 E7OT7 No 7/64 1.4

COST FACTOR ASSUMPTION:1. Mid- 1997 California Labor and Electrode prices.2. Costs are based on field deposition of weld metal utilizing the AWS D1.1 parameters for volts,

amps; electrodes stick out and travel speed shown on the attached procedure qualificationrecords for each of the electrodes shown.

3. Cost of inspection not included.4. Based on welding under normal field conditions in the flat position.

23

I'E.. ••110N & -•• Sc"I=IVlC'E

REPORTNO.

UCB/EERC-92/10JULY 1992 EARTHQUAKE ENGINEERINGRESEARCHCENTER

SLOTTEDBOLTEDCONNECTIONENERGYDISSIPATERS

(WITHANAPRIL,1993ADDENDUMOFSOMERECENTRESULTS)

by

CARL E. GRIGORIANTZONG-SHUOHYANGEGOR P. POPOV

Report to National Science Foundation

Abstract

Slotted Bolted Connections (SBCs) are modified bolted connections designed to dissipateenergy through friction during rectilinear tension and compression loading cycles. Exper-imental results on two types of SBCs are reported. In one type, friction occurs betweenclean mill scale steel surfaces; in the other, friction is between clean mil1 scale steel andbrass surfaces. The behavior of connections with brass on steel frictional surfaces is foundto be more uniform and simpler to model analytically than that with steel on steel surfaces.These connections maintain essentially constant slip force, and unlike those with steel onsteel surfaces, require minimal overstrength of the system in design. The frictional mecha-nisms giving rise to the observed behavior are explained. As an example of application a onestory diagonally braced frame was designed and its behavior determined for four differentearthquakes. Experimental results are presented for the fabricated SBC for this frame sub-jected consecutively to the four displacement histories derived from these earthquakes. Theagreement between the analytical and experimental results is found to be excellent. Becauseof the intrinsic simplicity of the SBCs and their very low cost, their use in seismic designand retrofit applications appears to be very promising.

This 7•ps publication is a re-print of a Univ. of California, Berkeley,

Earthquake Engineering Research Center Report No. UCB/EERC-92/10

and includes an April 1993 addendum.

Introduction

Various types of energy dissipating devices, utilizing friction as means of energy dissipation,have been tested and studied by researchers [4, 6, 7]. Two of the common features of thesedevices have been that their manufacture requires precision work or exotic materials andthat their installation demands specialized training. Consequently, the additional expensein using such devices has prevented their wide acceptance in engineering practice. Thedevelopment of the Slotted Bolted Connections (SBCs) as energy dissipators represents anattempt to overcome the abovementioned shortcomings of these systems. SBCs, as presentedin this paper, require only slight modification of standard construction practice, and requirematerials that are widely available commercially.

In this paper a Slotted Bolted Connection (SBC), see Figure 1, refers to a bolted connec-tion where the elongated holes or slots in the main connecting plate, in which the bolts areseated, are parallel to the line of loading. In addition a Belleville washer [8] is placed underthe nut. Two types of SBC specimen are discussed in this paper, one with brass insert platesand one without. Upon tightening of the bolts, the main plate is "sandwiched" directlybetween either the brass insert plates or the outer steel plates. The holes in the brass insertplates and in the steel outer plates are of standard size. When the tensile or compressiveforce applied to the connection exceeds the frictional forces developed between the frictionalsurfaces, the main plate slips relative to either the brass insert plates in the case of the firsttype specimen or the outer steel plates in the case of the second. This process is repeatedwith slip in the opposite direction upon reversal of the direction of force application. Energyis dissipated by means of friction between the sliding surfaces. Application of cyclic loads ofmagnitude greater than the slip force results in approximately rectangular hysteresis loops.The earliest investigations of SBCs as energy dissipators date back to 1976 when a series ofexperiments were carried out at San Jose State University (SJSU) [1] on specimens similarin concept to those presented here. The term SBC used here is adopted from the reportby T. F. Fitzgerald, et al. [3]. A number of other researchers have also investigated similardevices [2, 5].

Specimens and Experimental Results

To date, over forty SBC specimens of various bolt sizes, configurations and surface conditionshave been tested at the University of California at Berkeley (UCB). Experimental resultsfor specimens presented in this paper are representative of the salient SBC characteristicsencountered throughout testing. Presented here are two specimens which are identical inevery aspect with the exception that one includes shim like brass insert plates with a holepattern matching that of the outer steel plates. Figures 1 and 2 show, respectively, thedetails of an SBC connection and the overall view of a typical assembled test specimen.

3

Both specimens are of A36 steel. The steel surfaces were cleaned to clean mill scalecondition. The brass plates were of the widely available half hard cartridge brass variety(UNS-260). The test specimens were prepared by a local structural steel fabricator so asto simulate industry standards. Holes and slots in the steel plates were punched, and theedges were deburred. The two specimens described in this section are two bolt specimens.The bolts used were ½ inch diameter, 3• inches long A325 bolts. The Belleville washers usedwere 8-EH-112 Solon compression washers. One such washer with a hardened washer on topwas placed under each nut. Belleville washers are initially cone shaped annular disk springswhich flatten when compressed. Earlier studies of SBCs [1] have shown that without theuse of Belleville washers, and under large cyclic displacements, there is an almost immediateloss of bolt tension resulting in quick degeneration of the slip force. With the inclusion ofBelleville washers, both turn of the nut and torque wrench methods of developing minimumbolt tension (70% of minimum tensile strength [llD become inapplicable. To achieve thedesired initial bolt tension, Direct Tension Indicator (DTI) washers were placed under eachbolt head. DTIs are specially produced washers with protrusions pressed out of the flatsurface. As the bolt is tightened, the compressive force exerted on the DTI flattens theprotrusions and reduces the gaps between the flat portions of the DTI and the head of thebolt. The gaps can easily be measured with a supplied feeler gage. When the feeler gagefails to enter a specified number of gaps, the desired load in the bolt has been reached. DTIsused here were designed to indicate a bolt tension in the range of 12 to 14 kips.

The specimens, described above, were placed within an MTS loading frame as shownin Figure 3. The ram was capable of applying forces of 300 kips statically and 250 kipsdynamically, with a maximum displacement stroke of 6 inches. Both displacement andforce control were possible through a controller unit, and a function generator enabled theservorarn to produce preprogrammed load or displacement histories. All testing was doneunder displacement control. Axial load and displacements in the specimen were measuredthrough a load cell built into the MTS loading frame and a Linearly Variable DisplacementTransducer (LVDT) built into the servoram. Axial force and displacement were monitoredand recorded using a Data Acquisition System in conjunction with an IBM PC-AT computer.In addition, an X-Y plotter recorded load-displacement curves on paper for immediate visualobservation of results.

Figures 4 and 5 show the applied displacement histories, force responses and the resultinghysteresis loops for the two selected tests. Figure 4, representing the case of friction betweenlike clean mill scale steel surfaces, shows the main shortcoming of SBCs with friction betweensteel surfaces. As seen in the force response diagram, there is an almost immediate increasein the slip force followed by a quick drop to a magnitude several times less than the peakslip force. Although this behavior has not been observed in all tests of SBCs with frictionbetween like steel surfaces, it has been present, to various extents, in the majority of cases.In tests with specimens where the mill scale steel surfaces were polished by wire brushingand those in which the surfaces were roughened and the mill scale removed by sand blasting,

this behavior not only did not disappear but was actually intensified. The occurrence of thisbehavior in SBCs where friction occurs between steel surfaces renders such SBCs inefficient,at best, and impractical, at worst, as energy dissipators. Figure 5 represents the case of aSBC test with friction between clean mill scale steel and brass surfaces. As seen in Figure5, the use of brass insert plates significantly reduces the variations in slip force magnitudeobserved in SBCs where friction occurs between steel plates, almost completely eliminatingthis undesirable behavior.

Discussion of Experimental Results

A discussion of experimental results involving friction must necessarily involve conceptsof Tribology. Tribology is the body of science dealing specifically with friction, wear andlubrication. Terminology is a matter controversy in this field. The Tribological terminologyused here is adopted from E. Rabinowicz's classic book "Friction and Wear of Materials"[9]. Friction is defined as "resistance to motion which exists when a solid object is movedtangentially with respect to the surface of another which it touches." Wear is defined as the"removal of material from solid surfaces as a result of mechanical action." Of the several typesof wear discussed in Tribology literature, the two most relevant to the present discussion areadhesive wear and abrasive wear. Adhesive wear occurs when "two smooth bodies are slidover each other, and fragments are pulled off one surface to the other." These fragmentsmay later return to the original surface or form into loose wear particles. Abrasive wearoccurs when "a rough hard surface, or a soft surface containing hard particles, slides ona softer surface and ploughs a series of grooves in it." The material from the groovesgenerally forms into loose wear particles. Adhesive wear is almost universally present in allfrictional phenomena, and it is the authors' belief that it, in conjunction with some abrasivewear, is the main mechanism of wear in the SBCs tested. In general, no one explanationcan satisfactorily account for observed frictional behavior as many different mechanismsare involved in friction and wear processes, some simultaneous, some sequential and ofteninteracting with each other. Presented here is a qualitative explanation of the experimentallyobserved SBC behavior based on the above mentioned Tribological notions and experimentalobservations. The explanation given here applies to both SBCs where friction occurs betweenlike steels and where friction occurs between steel and brass. It is believed that as slidingis begun, wear particles are formed due to adhesive wear between the sliding surfaces. Thisresults in outward displacement of the outer plates in the direction of the bolt axes. This inturn results in an increase in the bolt tension force and therefore an increase in the normalforce between the sliding surfaces. As frictional force is directly proportional to normalforce, this increase in the normal force is observed as an increase in the slip force. Withcontinued sliding, a portion of the loose wear particles fall out of the connection, as observedexperimentally, while the rest are either reabsorbed or act as abrasive particles contributing

5

to abrasive wear. In Tribological terminology, the phenomenon that occurs here can be,simplistically, described as adhesive wear giving rise to wear particles which then causeadditional abrasive wear. That abrasive wear occurs despite the smoothness of the originalsurfaces is evidenced by the appearance of sliding surfaces observed after the completion ofexperiments and upon the dismantling of the specimens. In the case of friction between likeclean mill scale steel surfaces, both surfaces can be described as severely scratched. Whilein the case of friction between clean mill scale steel on brass, only the brass surface appearsas scratched while the steel surface appears undamaged but with smears of brass. Scratchedsurfaces are a typical consequence of abrasive wear. The fall out and reabsorption of wearparticles has the effect of reducing the bolt tension force as the outer plates now displaceinward. This results in a reduction of normal force and is observed as a drop in the slipforce. That the outer plates displace outward and then inward simultaneous with rise anddrop in the slip force has been confirmed by measurements of the displacements of the outerplates along the axes of the bolts.

The above mentioned behavior, i.e. initial increase in slip force followed by a drop,observed in both Figures 4 and 5, although clearly far more poignantly evident in Figure 4,is directly attributed to the wear mechanisms mentioned above. The difference in behaviorbetween the two types of specimens is solely due to the choice of the use of brass as africtional surface, as the other two parameters known to influence adhesive wear, namelyinitial normal force and total travel distance, were identical for the two presented specimens.This choice was made precisely with the reduction of wear in mind. Brass is a commonchoice as a material frictionally compatible with low and medium carbon steels, and is oftenused in moderate cost applications where it is desired to reduce adhesive wear [9].

Application and Verification of Assumptions

As an illustration of the utility of SBCs as energy dissipators, consider the example structureshown in Figure 6. A SBC with a slip force of 60 kips connects the diagonal brace to the mainstructure. Analysis of the structure was performed using the DANS [10] computer program.Newmark's step-by-step integration method was used. The structure was assumed to behaveas a shear structure, and the SBC was assumed to behave as an elastic-perfectly-plasticconnection. Viscous damping was assumed to be 2%. Responses due to four accelerationhistories were calculated. The acceleration histories were as follows: the 1971 Pacoima Damearthquake S16E, the 1952 Taft earthquake N21E with magnification factor of 5, the 1940E1 Centro earthquake S00E with magnification factor of 2 and the 1987 Whittier earthquakeN00E, at Sylmar, with magnification factor of 40. Figures 7, 8, 9 and 10 show groundacceleration histories, structure displacement responses and energy diagrams for each appliedhistory. The columns remain elastic at all times and the SBC prevents the buckling oryielding of the diagonal brace. An examination of the energy diagrams reveals that on the

6

average close to 85% of the total input energy is dissipated by the SBC.To verify the validity of the assumption of elastic-perfectly-plastic behavior for SBCs with

brass insert plates and to observe the response of such an SBC to displacement histories morerealistically representing response to actual earthquakes, an SBC specimen was designed toslip at 60 kips. Based on previous results from tests of specimens with two • inch diameterA325 bolts, a test specimen with eight • inch A325 bolts was fabricated. The specimen wassubjected to SBC slip displacement responses derived from the above mentioned analyses.The four SBC slip displacement response histories were applied consecutively, in the orderof acceleration histories mentioned above, to this specimen. Figures 11, 12, 13 and 14show SBC sup displacement response histories and analytical and experimental hysteresisdiagrams for each acceleration history. It is seen that the target slip force of 60 kips isattained almost perfectly in response to the first displacement history. As expected, the slipforce drops, although not significantly, for the next three applied displacement histories. Therectangular shape of the hysteresis loops, coupled with the reasonably constant slip force,indicates that the assumption of elastic-perfectly-plastic behavior for SBCs with brass insertplates is a valid one.

Concluding Remarks

Both SBC types have been shown capable of dissipating significant quantities of energy asjudged by the areas enclosed by the experimentally arrived at hysteresis loops. Slip force inSBCs where friction occurs between like steel plates has been shown to vary significantly.The peak slip force for such SBCs occurs almost immediately and may be several timesthe magnitude of the mean slip force. As such, for this type of SBC to dissipate energythroughout the course of ground excitation, either the members supporting the SBC mustbe designed with excessively large safety factors or the SBC itself must be under-designed.On the other hand, in SBCs where because of the brass insert plates friction occurs betweenbrass and steel, slip force has been shown to remain relatively constant over the range ofinterest. It has also been shown that such SBCs behave in nearly perfect elastic-perfectly-plastic manner. In view of these results, it is evident that SBCs with steel on brass frictionalsurfaces possess significant advantages in terms of efficiency as energy dissipators and easeof modelling. As such, and with low material and fabrication cost, these SBCs exhibit greatpotential as an alternative choice for energy dissipation in seismic design and retrofit ofstructures.

AcknowledgementsThe authors are grateful for the support of the National Science Foundation under GrantBCS-9016781 enabling the pursuit of the described research. The continued encouragementsof Henry Lagorio and S. C. Liu of NSF are particularly appreciated.

Thanks are also due to Bill MacCracken our electronics engineer who has been involved inevery phase of testing over three years and whose assistance with testing and data acquisitionequipment operation has been invaluable.

Machine shop specialists Mark Troxler, Jeff Higginbotham and Doug Zulaica are also thankedfor their assistance.

The opinions expressed in this paper are those of the writers and do not necessarily reflectthe views of the sponsor.

References

[1]

[2]

Venuti, W.J., "Energy Absorption of High Strength Bolted Connections," Test Report,Structural Steel Educational Council, California, May, 1976.

Pall, A.S. and Marsh, C., "Energy Dissipation in Panelized Buildings Using LimitedSlip Bolted Joints," Proceedings, AICAP-CED conference, Vol. 3, Rome, Italy, May,1979.

[3]

[4]

[6]

[7]

Fitzgerald, T.F., Anagnos, T., Goodson, M., Zsutti, T., "Slotted Bolted Connectionsin Aseismic Design of Concentrically Braced Connections," Earthquake Spectra, Vol. 5,No. 2, 1989.

Pall, A.S., Verganalakis, V. and Marsh, C., "Response of Friction Damped BracedFrames," J. Struct. Div., ASCE, 108(6), 1313-1323, 1987.

Roik, K., Dorka, U. and Dechent, P., "Vibration Control of Structures Under Earth-quake Loading by Three Stage Friction Grip Elements," Earthquake Engineering andStructural Dynamics, Vol. 16, 501-521, 1988.

Constantinou, M.C., Reinhorn, A.M., Mokha, A. and Watson, R., "Displacement Con-trol Devices for Base Isolated Bridges," Earthquake Spectra, Vol. 7, No. 2, 1991.

Aiken, I.D. and Kelly, J.M., "Earthquake Simulator Testing and Analytical Studies ofTwo Energy Absorbing Systems for Multistory Structures," Report No. UCB/EERC-90/03, University of California, Berkeley, October, 1990.

[8] Timoshenko, S., Strength of Materials, Vol. 2, Van Nostrand Co., New York, NY,1934.

[9]

[10]

[11]

Rabinowicz, E., Friction and Wear of Materials, John Wiley and Sons, Inc., NewYork, NY, 1965.

Yang, T.S., "DANS, A Computer Program for the Dynamic Analysis of Nonlinear ShearBuildings," CE£99 Project, University of California, Berkeley, 1991.

Kulak, L.K., Fisher J.W. and Struik, J.H.A., Guide to Design Criteria for Boltedand Riveted Joints, 2nd. Ed. John Wiley and Sons, Inc., New York, 1987.

9

1/8" TH. BRASS PU

MAIN PLATE ...............

.... 1/2" DIA. A325 BOLT. 3-1/2" LONG

,. HARDENED FLAT WASHER

8-EH-112 SOLON

COMPRESSION WASHER

NUT

DIRECT TENSION INDICATOR (DTI) ...... :

UNDER HEAD

_ JTER PLATES

'- 9/16"x3-1/2" LONG SLOT

Figure 1

1/8" TH. BRASS INSERT PLATES.'::::::•.'•

, i ! ? ,! ,

................................ I , . . I I ,

.......... HARDENED WASHERB-EH- 112 SOLON

COMPRESSION WASHERS

UNDER NUT

I

· ' .......... DIRECT TENSION INDICATOR WASHER(DTI), UNDER HEAD

........... 1/2" DIA. A525 BOLT, 5 - 1 / 2 " LONG,'.' ALL PLATES ARE 5/8" TH. Al6 BAR STOCK

- , II o o O o O o i = . . . . , ooooo ................................ o o o ° o I I o o oo o

° ° • o°° I i ' • l ' o o oi 0 0 0 0 0

i.,.

"" WELD

l•']['•'gure 2 g/1 s"x3-M2" LONG SLOT

10

LOAD CELL

ADJUSTABLE GRIPS

SBC TEST SPECIMEN

! II !! I

1

I I I I

I ' , ', I I . . . .

. . . . . . . . . . I ' . . . . . . . ! '''• . . . . .

I

MTS TESTING FRAMELVDT

MTS SERIES 252 SERVORAM

Figure 3

11

STEEL ON S'I'T•T. IMPOSED D I S P L A • STEEL ON BRASS IMPOSED2• i ; 2 i i • i

...... i . . . . . . . . . . - : i· 0.5

0

-1

-1.5 i i :i :i i ...........

-2

40

0 20 40 60 80 100 120 (SF_•"ONDS)

STEEL ON STF• FORCB Rr•sm•s£

,_.20r•

g

-20

-40

140 0 20 40 60 80 100 120 140 (SECONDS)

STEEL ON BRASS FORCE RESPONSE

i i ' ' i

. . . . . . . . . . . . . . . . . . . . . . . . . i . . . . . . . . . . . . . . . . . . . . . . . . . • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i . . . . . . . . . . . . . . . . . . . . . . . . i . . . . . . . . . . . . . . . . . . . . . . . . .

2O

0

-20

4O

40

,.,2O

•o-20

20 40 60 8O 100 120TIME (SECONDS)

STEEL ON s'r•.h':L HYSTERESIS DIAGRAM

140

-2

I I I I I I I

4.5 4 ..e.5 o o.5 1 J.s 2DISPLACEMENT (INCHES)

40

20

-20

-40

-2

20 40 60 80 100 120 140TrME (SECONDS)

STEEL ON BRASS HYSTERESIS DIAGRAM

............................................ [ ...................... i ...............................................................................................................

-1.5 -1 -0.5 0 0.5 1 1.5 2DISPLAC•Mta,rr (INCHF_.S)

Figure 4 Figure 5Note: 1 Inch = 25.4 mm, 1 Kip = 4.45 kN.

12

150 kipsi ,

Rigid deck

W12 !

,6 ft

, • • W12X72 W12X190

42 ft •,o

F i g u r e 6

13

AOUPl I•.RATION Iii,STORY: I*PACOIMA ACCI• I•.RATION HISTORY: 5*TAFT

$Z 0.5 0.5 ·O

0 0 . . . . . . . . .

4

1 . • ] a [ ] , • , [ • ! 1 . 5 m , , _ , ] • • • [ , [

0 2 4 6 8 10 12 14 16 18 20 = 0 2" 4 6 8 10 12 14 16 18 20Tnvm (srcol•rDS) • (SSCONDS)

1.5 D NSE, 1.5 STRUCTURE DISPLACEMENT RESPONSE

. . _ _ . . _ . . . 5

0

................................................................................................... o . . . . . . . . . . . . . . . . . . . . . . . . . . . . .-0.5

4

-1.5 ' . . . . . . . . . ] -1.5 . . . . . . . . . .

9O0

80O

E100

0

0 2 4 6 8 10 12 14 16 18 20TIME (S•'X)•S)

ENERGY DIAGRAM

,. , V'•o• D T

[

0 2 4 6 8 10 12 14 16 18 20 ( S • S )

0 2 4 6 8 10 12 14 16 18 20 (sF•co•s)

ENERGY DIAGRAM

, 2 r Friction D ned I

0 2 4 6 8 10 12 14 16 18 20 (sr=co•s)

Figure 7 Figure 8

Note: 1 Inch -- 25.4 mm, 1 Kip •- 4.45 kN.

1.5

1

$

0

-0.5

-1

ACC•r I•.RATION HISTORY 2*EL CENTRO

-1.5

14

I I I I I I I I ,

0 2 4 6 8 10 12 14 16 18 20TIME (SF.•Z'OlVDS)

STRUCnmE m S P L • • •mS•NS£1.5

!..i

1.5

1

0.5

0

-.0.5

-1

-1.5

-1.5 . . . . . . . . . .0 2 4 6 8 10 12 14 16 18 20

(SF. CO•S)

ENERGY DIAG45O

. . . . . . •,•t F.n•4•

Friction Dnm *xl

i

I I I I I I I

,40O

5O

00 2 4 6 8 10 12 14 16 18 20

Tn• (SECONDS)

100

ACCI•.•[OtTION HISTORY: 40*WH1TflER

I I I J I I I I I

0 2 4 6 8 10 12 14 16 18 20Tn• (SF•-'ONDS)

STRUCTURE DISPLACEMENT RESPONSE

1

O.5

0 : •'•;- :;- : ' % ' ; ; • • ........................

-O.5

-1

= 1 . - • ' ' ' J ' ' L , , ,

0 2 4 6 8 10 12 14 16 18 20TIME (SECONDS)

ENERGY DIAGRAM

/ visco• Damp•

0 2 d 6 8 10 12 14 16 18 20Tn• (SECONnS)

Figure 9 Figure 10Note: 1 Inch = 25.4 mm, 1 Kip = 4.45 kN.

1.5

1

o •

-1

-1.50

80

60

40

• • o

ua 0

• -20

-40

-60

-80

8 0

60

4O

20% . J

0

O -20

-4O

-60

1 5

SBC SLIP DUE I*PACOIMA

{ T { T ! T IJ I I

I I I {

' - - - q - - - - + .... '{- .......... 4- ' ' ' ' .... + .... {-----F---{--•I { I I ' ' ' 'i I i /I { I I I I t I

. . . . . ' . - - k - . - _ I t , J i l {J { i i {

I r i i l B I l I I t I! I I I t n I I I ! I

" " ' " " ....... - t . . . . . -• . . . . +---------•--------{ . . . .

I I I I I t J 1/ - ' - 4 - - i ' ' {'•q----%---+- F - ' '{

! ! t ' I { i I! I I I I I

I { I I I I I l . . . . -{ . . . . -i. . . . . -- . . . . . . . . . . . . . . • . . . . 4. . . . . J - - - - - - - - - L - • - - - { _ •

{ ! i I { I I i {

i { I , J I ', ,I J J I

, L , t i I i [ [ J2 4 6 8 10 12 14 16 18 20

TIME (SECONDS)

ANALYTICAL HYSTERESIS DIAGRAM

i T T Ji I I l

........ . . . . . . . . . . . . . . . . i J1IJ

. . . . . . . · . . . . . .

1

j ' { [i { II I , i I

--0.5 0 0.5 1 1.5DISPLACEMENT (INCHES)

EXPERIMENTAL HYSTERESIS DIAGRAM

IIt

II{

-1.5 -1

- 8 0

' !

t i t

. . . . . . . . . . { . . . . . . q -

---t . . . . . . _, ul , . . . . . . . . . . . . . .

i i

SBC ;LIP DUE 5*TAFT1.3 I I I • • ! I

I I I I i I i II • ! tI I I I [ I I IL ! I ! • I

1 r ' ' • r - " ' T i tJ t ! I i ! {J [ I i i } ! II I I I i i I/ ! { I • t i I

0-• l .... : I : • d/ i i I t i t { k {

I iI I , A l i i I/ I I I ] ! I

I I i ' i0 , i . . . . . i . . . . . . , / ' -

{ •1 J I t i, U , i * , ,t w ! I I i i t

i I I • { { I I I 4., - i . . . . . . . . . . . . . . . . . . .i . . . . . . .I ! I i It I I t i Ii I I I • I

! t j i • I , I I-1.5 , , I , , I ,

0 2 4 6 8 10 12 14 16 18 20TIME (SECONDS)

ANALYTICAL HYSTERESIS DIAGRAM, , {

i ii ! I

• , , , t I

-1 -0.5 0 0.5 1 1.5DISPLACEMENT (INCHES)

EXPERIMENTAL HYSTERESIS DIAGRAM

80

60

40

20

0

-20

-40

- 6 0

-80-1.5

80

6O

4O

20

0

-20

-4O

-6O

-80-1.5

j ' , -

[ -

-1.5 -1 -0.5 0 0 . 5 1.5 -1 -0.5 0 0 . 5 1 1.5DISPLACEMENT (INCHES) DISPLACEMENT (INCHES)

F i g u r e 1 1 F i g u r e 1 2

Note: 1 Inch = 25.4 mm, 1 Kip = 4.45 kN.

1

- 1

-1.5e

SBU SLIP DUE 2*EL CENTRO' i : ' ! !' '

I { I t • J t ! II - - - - • . t 1 ............. • .............. - ........... .t . . . . . . . t . . . . . ? - - - " J . . . . -I

_ i [ • • i [ i i -

i J i i i J Ji • i i i i i f

............ 4 ........ • . . . . . . . • ............. i ................. i ............ • . . . . . . . • . . . . . . . i . . . . 4 . . . . . .

J { { i • J J ! |

........ 4 ........... .l . . . . . . . . . 4 ........... i .................. i .............. 4. . . . . . . • . . . . . j . . . . . . . ' . . . . .

t ! 1

I I I I J I I I

2 4 6 8 1 0 1 2 1 4 1 6 1 8 2 0 (SECONDS)

ANALYTICAL HYSTERESIS DIAGRAM8O

6O

4O

-2o-4O

16

8O

6O

4O

-4O

-6O

-8O

1='

0 - '

t

- 1 . :

. . . . . . . . . . . . . . . . . . . . . . r . . . . . . . . . . . . . . . . . . . . . . .

-6o ...................... i . . . . . . . . . . . . . . . . . . . ......................* !

' " 8 0 J I , , i

-1.5 -1 -0.5 0 0.5 I 1.5

E X P E R I I • A L HYSTERESIS DIAGRAM, ,, , , { {

j { i i { i J { {

,. I , I I I I

-1.s -• .o.s o o.s 1 1.sDISPLACEMENT (INCHF_.S)

SBC SLIP DUE 40*wHrrrlER

' ' ' i J i !, { i i i J I: • } { .......... { ................. { ........... , . •J . . . . •i - - r - i .......

{ i I • [ i {{ { , { { { • }

i l l { i l J { J m d{ { • { } i {

. . . . . . . . . . . t ............................ -1 •---} .................... .-T----"

' J i! ==

i, , , , , , , , i • i

0 2 4 6 8 1 0 1 2 1 4 1 6 1 8 2 0TIME (SF, C O • S )

ANAJ.,Y33CAJ., HYSTERESIS DIAGRAM

4 0 . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . • . . . . . . . . . . . . . . . . . . . . . . . . . .

-80 '-1• -1 -0.5 0 0_5 1 1-5

D I S P L A C E M E N T ( m C t m S )

EXPERIMENTPJ, HYSTERESIS DIAGRAM8 0 ! '" • i •

6 0 . . . . . . . . . . ]--- .............. • ........................ • . . . . . . ? ............. F ...................

J

4020 {

0

-20

..4O

-60 •

..8O-1.5 -1 -0.5 0 0.5 1 1.5

r r (mOW_S)

F i g u r e 1 3 F i g u r e 1 4

N o t e : 1 Inch = 25.4 m m , 1 K i p = 4.45 kN.

17

Addendum

April, 1993

Shake Table Experiments

Within six months of the publication of the original body of this document, a test structureincorporating twelve SBCs was tested on the shake table at UCBs Earthquake SimulatorLaboratory. For sake of completeness, a sampling of the experimental results gained fromthis testing program is presented in this addendum.

The three story one bay steel test structure, depicted in Figure lA, supported 30,000lbs. per floor and stood over 20 ft. high. The lateral force resisting system of the structureconsisted of two moment resisting frames and Chevron braces connected with SBCs at eachlevel and on each of the two frames of the structure. The design slip loads for the SBCs weredetermined by computer simulation of the structure's response to various seismic inputs. Thedesign called for slip forces of 15 kips for SBCs at the first level and 7.5 kips at the secondand third levels. This requirement was accommodated by using two bolt SBCs identicalto that shown in Figures 1 and 2 of this report and leaving one of the two bolts loose inthe second and third levels. This arrangement was chosen so that slip loads in the secondlevel could be doubled by tightening of the second bolt midway through the testing programto experiment with an altogether different design with a different structural response. Theentire bracing system, including the SBCs, was fabricated commercially using standard shoptolerances and practice at a local steel fabrication shop.

Figure 2A shows the shake table acceleration history for one of the tests conducted onthe structure. Figures 3A and 4A show hysteresis diagrams for the six SBCs on each ofthe two frames in the structure in response to the above table acceleration history. Thediagrams are arranged such that they represent hysteresis diagrams for SBCs at the first,second and third levels at the bottom, middle and top of the figures respectively. The datapresented in these figures is raw, unfiltered with no adjustment to account for zero shiftsin instrumentation. It is seen that the curves are similar in character to those obtainedfor SBCs tested in the MTS testing frame. The slip loads are within reasonable range ofthe design requirements. It must be noted that as the second and third level SBCs wereeffectively single bolt connections, variations in slip loads axe expected to be greater thanthose in the two bolt connections at the first level. With a larger number of bolts, as wouldbe the case in a real structure, the degree of variation between slip loads for SBCs with same

18

number and size of bolts is expected to be smaller due to averaging of errors. It is also notedthe these curves verify again the validity of the elastic-perfectly-plastic characterization ofthe behavior of SBCs. A sense for the effectiveness of SBC may be gained from Figure5A. In this figure, the top curve represents the absolute input energy of the structure basedon integration of the measured base shear force of the structure with respect to the tabledisplacements. The curve immediately below this curve represents the sum of the total energydissipated by the SBCs, based on calculation of areas enclosed by the hysteresis curves, andstrain and kinetic energies. It is seen that at the end of the record, where kinetic and strainenergies vanish, nearly 75 % of the input energy is dissipated by the SBCs. The figure alsoindicates the relative magnitude of energy dissipated at each story and the magnitude ofenergy dissipated by each individual SBC, each layer below the "Total Dissipated" curverepresenting the contribution of one SBC.

In summary, the results obtained from the shake table testing of the structure withSBCs, a glimpse of which has been presented above, appear to verify at once the practicalityof implementation of SBCs into realistic structures and their effectiveness. A tremendouswealth of data has been generated from these experiments and the process of data reductionis currently in progress. Furthermore, analytical studies complementing and motivated bythe experimental efforts are being conducted by the authors with the aim of establishingdesign guidelines for use of SBCs in real structures.

19

Figure IA

TABLEACCELERATIONHISTORY1985CHILEEQ, LLOLLEOSIGNAL,AMPLIFIED TO PTA=.88G

=;

I

5

I I

10 15

TIME(SECONDS)

Figure 2A

!

2O

I

25

HYSTERESIS OF BRACE SF_3A . . . . .

• o

• . •

-15

-0.3 -0.2 -0.1 0 0.1 0.2 0.3DISPLACEMENT 0NCHF3)

• i i ....................................................... ii15 .................................................... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

•o ....................................................... i ............................ •i ............................................................ i .......................

- l o ! i i

-20 i i , i i ;

-0.3 -0.2 -0.1 0 0.1 0.2 0.3DISPLACEMENT 0NCT-IES)

HYSTERESIS OF BRACE SE1A

20

15

10

5

0

-5

-10

-15

-20

20

HYSTERESIS OF BRACE NE3Ai

-0.3 -0.2 -0.1 0 0.1 0.2 0.3

2O

15

10

5

0

-5

-10

-15

-20-0.3

D•SPt. A • (m•)

HYS'rERF_3IS OF BRACE NE2A

-0.2 -0.1 0 0.1 0.2 0.3 r r ( m a r e s )

HYSTERF_SlS OF BRACE NE1A20 20

15 ......................... i ...................................................................................... 'i ................................................ 15 ................... -' ....................... i ............................ = .......... : ....... ---• . . . . . . . .

i i! iii iii ........................... ...... ..................... !i!ii i ! . . . . . . . . . . . . . . . . . . . . . . . . •i ............................. i! ............... . , ' ? - - ............... i . . . . . . . . . . . . . .-15 -15 .................... i

-20 ' ' -20-0.3 -0.2 -0.1 0 0.1 0.2 0.3 -0.3 -0.2 -0.1 0 0.1 0.2 0.3

r)xst, LACEM•rr (n,•ctms) DXSPLnCEMt•-T (mctms)

F i g u r e 3 A

Note: 1 Inch = 25.4 mm, 1 Kip = 4.45 kN.

OF BRAC• SW3A

20 J i 'i

15 i •

10 ......................................................... • ...........................................{

-10 ..................... • ................................................................................................................. J .....................{

-20 { i { J i

0.3 0.2 oa 0 -0.1 -o.2 -o_aDrSPLACEMEm' (nqclms)

}rvgmRSSiS OF BRACE SW2A2O

15

10

• o

•-•. • o ] ] . • } Z ] Z i i i i i i 'i

t •-20 ' ' ' { {

0.3 0.2 0.1 0 -0.1 -0.2 -0.3 0 N • )

OF BRACE SW1A

' [ 5 ...................................................................................................................................................

•; ..... :i .................1 0 .............................

• 5 ...... ........................................

0 .. . . . . . i ............

- , , • ' • • - i • i i i • i j' ' ' • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

-10 .................................

. 2 O I I l

0.3 0 .2 0.1 -0 .2 -0.30 -0.1DISPLACE• (n,i•)

21

HYSTERESIS OF BRAC•E NW3A2O

15 • - - . 4 . . . . . '• ................... -: .................................................... i . . . . . . .

1 0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

-5

- - : i i : i !iiii .. :ilil i i . i i i:- .iii i :.i :ii:ii.i:iiii ii:i:i<i::i:ii:i:i:-15

' 2 0 '

0.3 0 .2 0 .1 0 -O. 1 -0.2 -0.3DISPLACEMENT ( I N a m S )

HYSTERESIS OF BRACE NW2A2 0 T , , f ,

- --

i,i i : i:i iii i:i"::i:i- 2 0 '

0.3 0 .2 0 .1 0 -O. 1 -0.2 -0.3

2O

15

l0

5

0

-5

-10

-15

-200.3

ACEmmT

HYSTERESIS oF BRACE)•WA

................................................... - ......................... t .............................. ? .................................................

I [ I t {

0.2 0.1 0 -o. 1 -0.2 -0.3DISPt. A C E m • r r • )

F i g u r e 4 A

Note: 1 Inch -- 25.4 mm, 1 Kip = 4.45 kN.

ENERGY INPUT AND DISSIPATION i•STOltW. S1985 CHil.V. EQ, LLO• SIGNAl. A M P • TO l•A•

22

l

l

BY SDCa

• •. ....................................

. . _ t'" r .;*

...... • - • - - ,

&4/d

I f J

10 IS 20

A T 3

AT•2BYM

AT 1

t

2S 3O

Figure 5A

EARTHQUAKE ENGINEERING RESEARCH CENTER REPORT SERIES

EERC reports are available from the National Information Service for Earthquake En$ineerin•NISEE) and from the National Technical InformationService(NTIS}. Numbers in parentheses are Accession Numbers assigned by the National Technical Information Service:. these are followed by a i•iee code.Contact NTIS. 5285 Port Royal Road, Springfield Virginia. 22161 for more information. Reports without Accession Numbers • not available from NTISat the time of printing. For a current complete list of EERC reports (from EERC 67-1) and availablity information, please contacl University of Califonmt,EERC, NISEE. 1301 South 46th Street. Richmond, California 94804.

UCB/EERC-90/16 'Sensitivity of Long-Period Response Spectra to System Initial Conditions,' by Bla.•uez. R., Ventunt, C. and Kelly, J'.M., 1990.

UCB/EERC-90/17 ' Behavior of Peak Values and Spectral Ordinates of Near-Source Strong Ground-Motion over a Dense Array,' by Niazi, M., June 1990.

UCB/EERC-90/18 'Material of Elastomers used in Earthquake Base Isolation,' by Papoulia, K.D. and Kelly, J.M., 1990.

UCB/EERC-90/19 'Cyclic Behavior of Steel Top-and-Bottom Plate Moment Connections,' by HarrioR, J.D. and Astaneh-Asl. A.. August 1990, (PB9 ! 229260/AS)A05.

UCB/EERC-90/20 'Seismic Response Evaluation of an Instrumented Six Story Steel Building,' by Shem J.-H. and Astaneh-Asl, A., December 1990, (PB91229 294/AS)A04.

UCB/EERC-90/21 'Observations and Implications of Tests on the Cypress Street Viaduct Test Structure,' by BolIo, M., Mahin, S.A.. Moehle, J.P.,Stephen. R.M. and Qi, X.. December 1990.

UCB/EERC-91/01 'Experimental Evaluation of Nitinol for Eneri• Dissipation in Structures,' by Nims, D.K., Sasaki, fl=K. and Kelly, J.M., 1991.

UCB/EERC.91/02 'Displacement Design Approach for Reinforced Concrete Structures Subjected to Earthquakes,' by Qi, X. and Moehle, J.P., January1991.

UCB/EERC-91/03 'A Lons-Period Isolation System Using Low-Modulus Hijh-Damping Isolators for Nuclear Facilities at Soft-Soil Sites,' by Kelly, J.M.,March 1991.

UCB/EERC-g[/04 'Dynamic and Failure Characteristics of Bridgestone Isolation Bearing,' by Kelly, J.M., April 1991.

UCB/EERC-91/05 'Base Sliding Response of Concrete Gravity Dams to Earthquakes.' by Chopra, A.K. and Zhan$, L., May 1991.

UCn/EERCo91/06 'Computation of Spatially Varying Ground Motion and Foundation-Rock Impedance Matrices for Seismic Analysis of Arch Dams," byZhang, L. and Chopra, A.K., May 1991.

UCB/EERC-91/07 'Estimation of Seismic Source Processes Using Strong Motion Array Data,' by Chiou, S.-J., July 1991.

UCB/EERC-91/08 'A Response Spectrum Method for Multiple-Support Seismic EJtcitations,' by Der A. and Neuenhofer, A., AulPiSt 1991.

UCB/EERC-91/09 'A Preliminary Study on Energy Dissipating Cladding-to-Frame Connection,' by Cohen, J.M. and PoweU, O.H.. September 1991.

UCB/EERC-91/10 'Evaluation of Seismic Performance of a Ten-Story RC Building During the Whittier Narrows Earthquake,' by Miranda, E. and Bergtero, V.V., October 1991.

UCB/EERC-91/I 1 'Seismic Performance of an Instrumented Six Story Steel Building.' by Anderson, J.C. amd Bertero, V.V, November 1991.

UCB/EERC-91/12 'Performance of Improved Ground During the Loma Prieta Earthquake,' by Mitchell, J.K. and Wentz, Jr., F.J., October 1991.

UCB/EERC-91/13 'Shaking Table - Structure Interaction,' by Rinawi, A.M. and Clouf, R.W, October 1991.

UCB/E£RC-91/14 'Cyclic Response of RC Beam-Column Knee Joints: Test and Retrofit,' by Mazzoni, S.. Moehle, J.P. and Thewalt, C.R., October 1991.

UCB/EERC-91/I$ 'Design Guidelines for Ductility and Drift Limits: Review of State-of-the-Practice and State-of-the-Art isa Ductility and Drift-BasedEarthquake-Resistant Design of Buildings,' by Bertero. V,V., Anderson, J.C., K.rawin.kler, H., Miranda, E. and The CUREa and TheKajima Research Teams,, July 1991.

UCB/EERC-91/16 'Evaluation of the Seismic Performance ora Thirty-Story RC Buildinl,' by Anderson. J.C., Miranda, E., Bertero. V.V. and The KajimaProject Research Team., July 1991.

UCB/EERC-91/17 'A Fiber Beam-Column Element for Seismic Response Analysis of Reinforced Concrete Structures,' by Taucer, F., Spacone, E. andFilippou, F.C.. December 1991.

UCB/EERC-91/18 'Investi•tion of the Seismic Response of a Lightly-Damped Torsionally-Coupled Building.' by Boroschek. R. and Mahin, S.A.,December 1991.

'Studies of a 49-Story Instrumented Steel Structure Shaken dui'inS the l. oma Prieta Earthquake,' by Bonowitz, D., Chert. C.-C. andAstaneh-AsL A., February 1992.

UCB/EERC-92/02 'Response of the Dumbarton Bridge in the Loma Prieta Earthquake,' by Fenves, G.L.. Filippou. F.C. and Sze. D.T., January 1992.

UC•EERC-92/03 'Models for Nonlinear Earthquake Analysis of Brick Masonry Buildinls,' by Menili, Y.,'McNiven, H.D. and Tanrikulu, A.K., March1992.

UCB/EERC-92/04 'Shear StreniBh and Deformability of RC Bridge Columns Subjected to Inelastic Cyclic Displacements,' by Aschheim. M. and Moehle.J.P., March 1992.

UCB/EERC-92/05 'Parameter Study of Joint Openin$ Effects on Earthquake Response of Arch Dams,' by Fenves, G.L, Mojitahedi, S. and Reimer, R.,April 1992.

UCB/EERC,92/06 "Seismic Behavior and l•sign of Semi-Rigid Steel FTamcs,' by Nadet, M.N., and Astancb-Asl, A., May 1992.

UCB/EERC-g2/07 "A Beam Element for Seismic Damage Analysis,' by Spacone, E., Ciampi, V. and FUippou, F.C., August 1 • .

UCI•EERC-•2/O8 'Nonlinear Static and Dynamic Analysis of Reinforced Concrete Subasscmblages," by Filippou, F.C., D'Ambrisi, and lsan, ,it.,

August 1992

UCB/EERC-92/10 "Slotled Bolted Connection Energy Dissipators," by Grigorian, C.E., Yang, T.-S. and Popov, E.P., July 1•)2.

UCB/EERC-88/14

UCB/EERC-88/15

UCB/EERC-88/16

UCB/EERC-88/17

UCB/EERC-88/18

UCB/EERC-88/19

UCB/EERC-88/20

UCB/EERC-89/0 I

UCB/EERC-89/02

UCB/EERC-89/03

UCB/EERC-89/04

UCB/EERC-89/05

UCB/EERC-89/06

UCB/EERC-$9/07

UCB/EERC-89/08

UCB/EERC-89/09

UCB/EERC-89/10

UCB/EERC-89/I I

UCB/EERC-89/12

UCIFEERC.89/l 3

UCB/EERC-89/14

UCB/EERC-89/15

UCB/EERC-89/16

UCB/EERC-90/01

UCB/EERC-90/02

UCB/EERC-90/03

UCB/EERC.90/04

UCB/EERC-90/0$

UCB/EERC-90/06

UCB/EERC-90/0?

UCB/EERC-90/08

UCB/EERC-90/09

UCB/EERC-90/I 0

UCB/EERC-90/i 1

UCB/EERC-90/i 2

'An Experimental Study of' the Behavior of Dual Steel Systems.' by Whittaker. A.S., UanlL C.-M. and Berteto. V.V,, September 1918.(PB91 212 712)AI6.

'Dynamic Moduli and Damping Ratios for Cohesive Soils,' by Sun, J.g.. Golesorkhi. R. and Seed, H.B., August 1918, (PB91 210922)A04.

'Reinforced Concrete Flat Plates Under Lateral Load: An Experimental Study Including Biaxial Effects,' by Pan, ,aL and Moehle, J.P.,October 1988, (PB91 210 856)AI3,

'Earthquake Engineering Research at Berkeley - 1988.' by EERC, November 1988, (PB91 210 864)A!0.

'Use of Energy as a Design Criterion in Earthquake-Resistant Design,' by Uanlg C.-M. and Benero, V.V., November 19811, (PB91 210906/AS)A04.

'Steel Beam-Column Joints in Seismic Moment Resisting Frames,' by Tsai, K.-C. and Popov, E.P., November 19118, (PBgl 21798a/AS)A20.

'Base Isolation in Japan, 1988,' by Kelly, J.M., December 1988, (PB91 212 449)A08.

'Behavior of Long Links in Eccentrically Braced Frames,' by Enlelhardt, M.D. and Popov, E.P., January 1989, (PB92 143 OS6)Aig.

'Earthquake Simulator Testing of Steel Plate Added Damping and Stiffness Elements,' by Whittaker, A., Bertero, V.V., Alonso, J. amdThompson, C., January 1989, (PB91 229 252/AS)Al0.

'Implications of Site Effects in the Mexico City Earthquake of Sept. 19, 1985 for Earthquake-Resistant Design Criteria in the San Fran-cisco Bay Area of California,' by Seed, II.B. and Sun, J.l.. March 1989, (PBgl 229 369/AS)A07.

'Earthquake Analysis and Response of Intake-Outlet Towers,' by GoyaL A. and Chopra, A.IC. July 1989, (PB91 229 286/AS)AIg.

'The 1985 Chile Earthquake: An Evaluation of Structural Requirements for Bearing Wall Buildinp,' by Wallace, J.W. and Moehle,J.P., July 1989, (PB91 218 O08/AS)AI3.

'Effects of Spatial Variation of Ground Motions on Large Multiply-Supported Structures,' by HaG, H., July 1989, (PB91 22916 l/AS)A08.

'EADAP - Enhanced Arch Dam Analysis Program: Users's Manual.' by Ghanaat, Y. and Clough. R.W., AulluSt 1989, CPB91 212$22)A06.

'Seismic Performance of Steel Moment Frames Plastically Designed by Least Squares Stress Fields,' by Obi. lC, and Mahin, SA.,August 1989, (PBgl 212 597)A05.

'Feasibility and Performance Studies on [mprovin8 the Earthquake Resistance of New and Existing Buildin8s Using the Friction Pendu-lum System,' by Zayas. V., Low. S.. Mahin, S.A. and Bozzo, L , July 1989, {PB92 143 064)A!4.

'Measurement and Elimination of Membrane Compliance Effects in Undrained Triaxial TestinlL' by Nicholson. P.O., Seed, R.B. andAnwar. II.. September 1989. (PB92 139 641/AS)Al3.

'Static Tilt Behavior of Unanchored Cylindrical Tanks,' by Lau, D.T. and Clough, R.W., September 1989, (PB92 143 049)AI0.

'ADAP-88: A Computer Proip=am for Nonlinear Earthquake Analysis of Concrete A•h Dams,' by Fenves, G.L, Mojtahedi, S. and Rei-ruer, R.B., September 1989, (PB92 139 674/AS)A07.

'Mechanics of Low Shape Factor Elastomeric Seismic Isolation Bearings,' by Aiken, I.D., Kelly, $.M. and Tajirian, F.F., November1989, (PB92 139 732/AS)A09.

'Preliminary Report on the Seismological and Engineering Aspects of the October 17, 1989 Santa Cruz oma Prieta) Earthquake,' byEERC, October 1989, (PB92 139 682/AS)AC)4.

'Experimental Studies of a Single Story Steel Structure Tested with Fixed, Semi-Ri8id and Flexible Connections,' by Nader, M.N. andAstaneh.Asl, A, August 1989, (PB91 229 21 I/AS)Al0.

'Collapse of the Cypress Street Viaduct as a Result of the l,oma Prieta Earthquake,' by Nims, D.IC, Miranda, £., Aiken, I.D., Whit-taker, A.S. and Bertero. V.V., November 1989, (PB91 217 93S/AS)A05.

'Mechanics of High-Shape Factor Elastomeric Seismic Isolation Bearings,' by Kelly, J.M., Aiken, I.D. and Tajirian, F.F, March 1990.

'Javid's Paradox: The Influence of Preform on the Modes of Vibrating Beams,' by Kelly, J.M., Sackman, J.L. and Javid, A., May 1990,(PB91 217 943/AS)A03.

'Earthquake Simulator Testing and Analytical Studies of Two Energy-Absorbing Systems for Muitistory S t r u , a u • ' by Aiken, I.D. andKelly, J.M., October 1990.

'Damage to the San Francisco-Oakland Bay Bridle Durinl the October 17, 1989 Earthquake,' by Astaneh-Asl, A., June 1990.

'Preliminary Report on the Principal Geotechnieal Aspecu of the October IT, 1989 l.oma Prieta Earthquake,' by Seed, ]LB., Dicken-son. S.E., Riemer, M.F., Bray, J.D., Sitar, N., Mitchell, J.IC, Idriss, I.M., K•yen, R.E., Kropp, A., Harder, Jr. and Power, M.S.,April 1990.

'Models of Critical Regions in Reinforced Concrete Frames Under Seismic Excitations,' by Zul6qar, N. and Filippou, F.C., May 1990.

'A Unified Earthquake-Resistant Design Method for Steel Frames Using ARMA Models,' by I, Conte, J.P., Mahhng S.A. andPister, K.S,, June 1990.

'Soil Conditions and Earthquake Hazard Mitil•tion in the Marina District of San Francisco,' by Mitchell J.IC, Masood, T• Kayen,R.E. and Seed, R.B., May 1990.

'Influence of the Earthquake Ground Motion Process and Structural Properties on Response Characteristics of Simple S • ' byConic, J.P., Pister, K.S. and Mahin, S.A., July 1990.

'Experimental Testing of the Resilient-Friction Base Isolation System,' by ClarlL P.W. and Kelly, J.M., July 1990, (PB92 143 OT2)A011.

'Seismic Hazard Analysis: Improved Models, Uncertainties and Sensitivities,' by Araya, R. and Der Kiureghiam A., Mat• 1988.

'Effects of Torsion on the Linear and Nonlinear Seismic Response of Structures," by Sedarat, H. and Berteto, V.V., September 19119.

STRUCTURAL STEEL EDUCATIONAL COUNCIL

TECHNICAL INFORMATION & PRODUCT SERVICE

MARCH 1998

Compatibilityof

Mixed Weld Metal

By

Alvaro L. Collinand

James J. Putkey

Acknowledgments

The Authors wish to thank Pat Hassett, Rudy Hofer, Dave McEuen, and JamieWinans of the Structura l Steel Educat ion Council, and Roger Ferch of the HerrickCorporat ion for thei r review and comments.

The information presented in this publication has been prepared in accordance with recognizedengineering principles and is for general information only. While it is believed to be accurate,this information should not be used or relied upon for any specific application withoutcompetent professional examination and verif ication of its accuracy, suitability, andapplicability by a licensed professional engineer, designer, or architect. The publication of thematerial contained herein is not intended as a representation or warranty on the part of theStructural Steel Education Council or of any other person named herein that this informationis suitable for any general or particular use or of freedom from infringement of any patent orpatents. Anyone making use of this information assumes all liability arising from such use.

Caution must be exercised when relying upon other specifications and codes developed byother bodies and incorporated by reference herein since such material may be modified oramended from time to time subsequent to the printing of this edition. Structural Steel EducationCouncil bears no responsibility for such material other than to refer to it and incorporate it byreference at the time of the initial publication of this edition.

Index of Steel TIPS Publications

The following is a list of available Steel TIPS. Copies will be sent upon request. Some are invery limited quantity.

· Seismic Design of Special Concentrically Braced Frames

· Seismic Design of Eccentrically Braced Frames

· Seismic Design of Column Tree Moment Resisting Frames

· Dynamic Tension Tests of Simulated Moment Resisting Frame Welded Joints

· Reference Guide for Structural Steel Welding Practices

· Seismic Strengthening with Steel Slotted Bolt Connections

· Slotted Bolted Connection Energy Dissipaters

· Heavy Structural Shapes in Tension Applications

j l

INTRODUCTION

PURPOSE

The purpose of this Steel TIPS is to provide structural designers, fabricators, anderectors with the history, use, and compatibility of mixed weld metals forstructural steel applications.

ORGANIZATION AND CONTENT

To accomplish the purpose, the authors have organized this Steel TIPS into thefollowing categories:

Historical background.AWS and AISC requirements.Effect of the Northridge Earthquake.Combanations of mixed weld metal.

The authors do not recommend or approve any particular combination of mixedweld metal Instead, they set forth combinations used and possible combinations.Please remember, the Engineer has the right to approve any combination of mixedweld metal.

CONTENTS

ACKNOWLEDGMENTS

CONTENTS

INTRODUCTION

1. HISTORICAL BACKGROUND / 1

2. AMERICAN WELDING SOCIETY (AWS) REQUIREMENTS / 3

3. AISC REQUIREMENTS / 5

4. EFFECT OF 1994 NORTHRIDGE EARTHQUAKE / 8

5. IMPACT TESTING BEFORE NORTHRIDGE / 9

6. MIXED WELD METAL COMBINATIONS / 10

7. COMMENTARY ON IMPACT REQUIREMENTS AND IMPACT TESTING / 17

REFERENCES / 21

1. HISTORICAL BACKGROUND

For many years fabricators and erectors haveused, as common practice, electrodes withdifferent specifications and classifications in thesame weld (mixed weld metals). Mixed weldmetal results from:

Different root pass and fill-in passelectrodes.Weld repair work to both shop and fieldwelds.The more recent practice of making fieldwelds over shop welds.

Compatibility is a basic requirement of mixed weld metals. The differentelectrodes and the mixed weld metal must have, as a minimum, matching yieldstrength, tensile strength, and impact properties--if impact properties arespecified.

In this SteelTIPS, the authors will limit their discussion to two welding processes:

Shielded Metal Arc Welding (SMAW).Flux Cored Arc Welding (FCAW).

MIXING ELECTRODES.

In the Same Weld Process. First, fabricators and erectors commonly used differentmanual coated "stick" electrodes (SMAW) in the same weld. Then, when f lux coredelectrodes (FCAW) came on the market in the mid 1950's, they started usingdifferent FCAW electrodes in the same weld. They used the different electrodes:

To take advantage of one electrode's penetration capabilities for root passesand the other electrode's capabilities for fill-in passes.For weld repair work.

In Different Weld Processes. With the availability of self shielding f lux coredelectrodes, fabricators and erectors began using electrodes from the two differentwelding processes in the same weld. They used:

SMAW electrodes in the root passes for good penetration and Iow hydrogenproperties, and FCAW electrodes in the fill-in passes for high depositionrates.SMAW over FCAW for weld repair work.

FCAW With Outside Gas Shielding. With the availability of flux cored electrodeswith outside gas shielding in the mid 1960's, fabricators and erectors started usingtwo FCAW processes in the same weld:

Self-shielding from the flux in the core of the wire (FCAW-ss).Gas shielding from an outside source (FCAW-g).

Fabricators generally used:

FCAW-g in the root passes for good welding characteristics.FCAW-ss electrodes in fill-in passes for good deposition rates.

Outside gas shielding was generally limited to shop fabrication because erectorshad difficulty protecting the gas from winds encountered in field erection.

Interestingly, fabricators and erectors used Gas Metal Arc Welding (GMAW)equipment to make the first Flux Cored Arc Welds. GMAW equipment used solidwire electrodes with inert gas shielding. Users called this GMAW welding process"Dual Shielding." Now common practice is to call flux cored electrodes withoutside gas shielding (FCAW-g) as "Dual Shielding."

Since Northridge. Then the 1994 Northridge Earthquake occurred. Since theEarthquake, the volume of mixed weld metal has proliferated greatly, and with justabout any combination of SMAW and FCAW electrodes. Combinations resultedfrom both damage repair work and from new welds. Damage repair work involvedwelding one classification or process over another classification or process. Newdesign details set up the situation where erectors used one process to weld overa different shop process.

IMPACT REQUIREMENTS.

At first, the construction industry paid little regard to impact requirements of theelectrodes. Later, various Advisory Task Groups formed to investigate theEarthquake damage (e.g., AISC, AWS, SAC Joint Venture, SEAOC, LA City andCounty, and other Code Agencies) called for impact requirements in electrodesused in seismic designs. The various code agencies are putting these impactrequirements into their codes.

2. AMERICAN WELDING SOCIETY(AWS) REQUIREMENTS

The AWS Structural Welding Code--Steel (AWS t - - - - - • - • •Code) covers the welding requirements for • • 1welded steel structures. [1] This Code does not i \/ • Idirectly address the use of mixed weld metals or ' I ( • Iimpact requirements for welds. However, the • IAWS Code:

· Implies the use of mixed weld metals in Section 3.3.

Sets forth impact testing requirements in Annex III if the contract drawingsor specifications require impact testing.

MIXING WELD METALS

Base Metal/Filler Metal Combinations. Although the AWS Code does not directlyaddress the use of mixed weld metals, it does imply their use. Close attention tothe untitled table in Section 3.3, "Base Metal/Filler Metal Combinations," page 41,shows:

"Any steel to itself or any steel to another in the same group" can be weldedby "Any filler metal listed in the same group."

"Any steel in one group to any steel in another" can be welded by "Any fillermetal listed for a lower strength group. [SMAW electrodes shall be the Iowhydrogen classification]" (E7015, E7016, and E7018). [2]

Implications. Section 3.3 and Table 3.1 on page 42 show that many base metalsin asteel group can be welded to each other by different electrodes; therefore, theAWS Code allows mixed weld metal in the same weld based on strengthrelationships.

Additional Requirements by Advisory Task Groups. Besides compatible yieldstrength and tensile strength requirements required by the AWS Code, the AdvisoryTask Groups recommended that electrodes involved in mixed weld metal weldsmust have compatible impact requirements Further, the Advisory Task Groupsrecommended testing and evaluation of the mixed weld metal combinations.

IMPACT

Not Addressed in AWS Code. The AWS Code does not address impactrequirements; that is the Engineer's responsibility. However, the AWS Code doesaddress impact testing requirements in Annex III.

IMPACT TESTING

Application of Annex III of the AWS Code. Annex III of the AWS Code addressesimpact testing. The title of Annex III is "Requirements for Impact Testing" with asubtitle "Mandatory Information." A further comment under the subtitle states:

(This Annex is a part of ANSI/AWS D1.1-96, Structural WeldingCode--Steel and includes mandatory requirements for use in thisstandard.) [3]

However, Section III1.1 states:

The impact test requirements and test procedures in this Annex shallapply only when specified in the contract drawings or specifications inaccordance with 5.26.5(3)[d] and 4.1.1.3, and Table 3.1 of this code. [4]

Thus, the decision to call for impact testing requirements is left to the discretionof the designer or engineer responsible for the contract drawings or specifications.

Use of Test Results. Annex Section III1.2 states in part:

. . . The energy values determined are qualitative comparisons on aselected specimen and although frequently specified as an acceptancecriterion, they cannot be used directly as energy figures that wouldserve for engineering calculations [e.g., failure analysis calculations]. [5]

Scatter. A great scatter is normal in Charpy V-Notch test results. The AWS Codeprovides a limited discussion of scatter in Annex Ill; however, it references otherpublications that thoroughly discuss fracture toughness--including scatter. AnnexIII, Table II1-1, calls for three specimens for each test location, with an optional fivespecimens per test location. When using five specimens, Note 2 in the Tableapplies and states in part, "The highest and lowest values are then discarded tominimize some of the scatter normally associated with Charpy testing of welds andHAZ. [Emphasis added.]" [6] HAZ denotes the portion of the base metal whosemechanical properties or microstructure has been altered by the heat of weldingand quenching effect of the base metal. See Article 7, "Commentary on ImpactRequirements and Testing," for comments on impact testing and scatter.

4

3. AISCREQUIREMENTS

The AISC Manual of Steel Construction:AIIowableStress Design(AiSC Manual) addressesmixing weld metals and impact requirements forboth the base metal and weld metals as follows:

Mixed Weld Metal. PART 5-Specifications and Codes, Specification forStructural Steel Buildings--Allowable Stress Design and Plastic Design (AISCSpecifications), discusses mixing weld metals.

Impact. PART 1-Dimensions and Properties, briefly discusses impact. PART5 -Specifications and Codes, addresses limited impact requirements. [7]

MIXING WELD METALS

AISC Specifications. Refer to Chapter J-Connections, Joints and Fasteners. SectionJ2.6., Mixed Weld Metal, states:

6. Mixed Weld Metal

When notch-toughness is specified, the process consumables for allweld metal, tack welds, root pass and subsequent passes,deposited in a joint shall be compatible to assure notch-toughcomposite weld metal. [8]

The following AISC Specification Commentary illustrates a lack of compatibilitybetween process consumables (electrodes), and reinforces the Advisory TaskGroups' recommendation that users evaluate mixed weld metals by testing.

AISC Specifications Commentary. Section C-J2.6., Mixed Weld Metal, states:

6. Mixed Weld Metal

Instances have been reported in which tack welds deposited usinga self-shielded process with aluminum deoxidizers (which by itselfprovided notch-tough weld metal) were subsequently covered byweld passes using a submerged arc process (which by itselfprovided notch-tough weld metal) resulted in composite weld metalwith Iow notch-toughness (Terashima and Hart, 1984; Kotecki andMoll, 1970; and Kotecki and Moll, 1972). [9]

IMPACT

PART I Dimensions and Properties. Pages 1-4, 1-5, and 1-6 have very good,concise write-ups on Brittle Fracture, Lamellar Tearing, and Jumbo Shapes andHeavy Welded Built-up Sections. However, PART 1 barely touches on notch-toughness (im pact). The last paragraph under the subtopic "Selecting a Steel" doesmention notch toughness.

PART 5 Specifications and Codes. The Specifications and correspondingCommentary address impact in Sections A3.1.c. Heavy Shapes, A4.2. Impact, andA.4.5. Other Forces. Engineers seldom request impact requirements for basemetal, except for heavy shapes. See below.

Section A3.1.c. Heavy Shapes specifies impact requirements for the followingmembers when subject to primary tensile stresses due to tension or flexure ifspliced using full penetrauon welds:

ASTM A6 Groups 4 and 5 rolled shapes.Built-up members with plates exceeding 2 in. thick

For this use the contract documents shall specify, "... the steel shall be specifiedin the contract documents to be supplied with Charpy V-Notch testing inaccordance withASTMA6, Supplementary Requirement S5. The impact test shallmeet a minimum average value of 20 ft-lbs, absorbed energy at +70°F . . . . " [10]

When using mixed weld metal in Groups 4 and 5 rolled heavy shapes, the designeror engineer, fabricator, and erector, should be familiar with impact requirementsand precautions addressed in:

Section A3.1.c.The Section's Commentary.All sectaons listed In Section A3 1 c on page 5-126, Includingcorresponding commentary sections

The Section also states in part:

The above supplementary toughness requirements shall also beconsidered for welded full-penetration joints other than splices in heavyrolled and built-up members subject to primary tensile stresses. [11]

The Specification Commentary discusses "considered."

Section A4.2. Impact, does not call for Charpy V-Notch toughness testing. TheSection states:

2. Impact

For structures carrying live loads* which induce impact, theassumed live load shall be increased sufficiently to provide forsame.If not otherwise specified, the increase shall be not less than:

[See pages 5-29 for listings ranging from 10% - 100%][The * footnote is not included in the above quote] [12]

Note: The listed percentages increase the live loads to compensate for loads thatinduce impact.

Section A4.5. Other Forces, states:

5. Other Forces

Structures in localities subject to earthquakes, hurricanes and otherextraordinary conditions shall be designed with due regard for suchconditions. [13]

1 1 1 EFFECT OF 1994NORTHRIDGEEARTHQUAKE

As Mentioned in Article 1, "HistoricalBackground," fabricators and erectors had usedelectrodes with different specifications andclassifications for many years before theNorthridge Earthquake.

MIXED WELD METALS

1 .20°F

Northridge Damage Repairs. The use of mixed weld metals increased following theEarthquake, mainly because damage repairs involved gouging out for weld cracks,and rewelding with an electrode other than the electrode used in the original weld.Erectors most commonly used--and continue to use--SMAW E701 8 Iow hydrogen,manual electrodes and certain FCAW electrodes to repair damaged weld jointsmade with FCAW, E70T-4 flux cored electrodes.

New Welds. Engineers started following the Advisory Task Group'srecommendation of requiring impact tests for mixed weld metals in new welds.

IMPACT REQUIREMENTS

Northridge Damage Repairs. During initial damage repairs, engineers paid littleattention to the impact requirements of the original weld metal and the repairelectrode because:

Expediency of the repairs precluded investigative testing.

AWS Code and the AISC Specifications did not have any impact requirementsfor base metals--except for limited requirements by AISC for Steel Groups 4and 5 jumbo shapes and certain built-up members. (See Specification A3.1 .Creviewed in Article 3, "AISC Requirements").

SMAW and FCAW repair electrodes had good impact requirements--usually20 ft. lbs. at-20°F.

As repair work progressed, the Investigating Advisory Task Groups made theirrecommendations available. Engineers expressed concern about therecommendations regarding impact requirements of the original weld metal andthe repair weld metal. However, engineers apparently took no action regardingrecommendations on the original weld metal, but did follow recommendations onrepair weld metal.

New Welds. Later, the Advisory Task Groups recommended that if engineerswanted impact requirements for mixed weld metal, each electrode used to makethe weld had to meet those impact requirements. As a result, engineers nowfrequently request impact requirements for both SMAW and FCAW electrodes.

8

11 IMPACT REQUIREMENTSAND IMPACT TESTINGBEFORE NORTHRIDGE

Most of the structural steel construction beforethe 1994 Northridge Earthquake had very littleimpact requirements because:

Applicable editions of the AWS Code hadno impact requirements.Applicable editions of the AISC Manual hadlimited or no impact requirements.

AWS IMPACT REQUIREMENTS

%

In AWS Code. Various AWS Codes governed the construction of structuressubjected to the Northridge Earthquake. The 1988, 1990, and 1994 AWS Codesincluded Appendix Ills that are almost identical to Annex III in the 1996 AWS Code.Therefore, the present AWS impact requirements limited to impacttesting--reviewed in Article 2, "AWS Requirements"--applied to structures builtunder the earlier named codes. Before 1988 the AWS Code lacked even thoseimpact testing requirements.

AISC IMPACT REQUIREMENTS

In AISC Manual. Various AISC Manuals governed construction of structuressubjected to the Northridge Earthquake. The 1980 AISC Manual included 1978Specifications sim ilar to the 1989 AISC Manual Therefore, the present AISC im pactrequirements--reviewed in Article 2, "AISC Requirements"--applied to structuresbuilt from 1978to 1994. Before 1978, the AISC Manual did not address impactrequirements.

6. MIXED WELD METALCOMBINATIONS

The use of mixed weld metal combinations canbe divided into the following three main periodsbased on the different conditions for thecombinations during the periods:

1. Pre-Northridge Earthquake Practice.2. Urgent Northridge Earthquake Damage Repairs.3. Post-Northridge Earthquake Practice.

The following "General Criteria," "Conditions" listed under a Period, and criteriawithin the tables, decided the electrode combinations used for each period.

GENERAL CRITERIA

The following criteria pertain to electrodes listed in Period 1, 2, and 3 tables:

Root weld passes require good penetration and good fusion to base metals.Fill-in weld passes require good fusion to base metals and to other passes.The AWS Welding Handbook and electrode manufacturers' bulletins containelectrode information and specifications. [14]FCAW-ss and FCAW-g electrodes make good fill-in passes because of theirhigher deposition rates. However, outside gas shielded electrodes createproblems in field welding because the shielding gas must be protected fromthe wind.The Engineer may require approval of mixed weld metal including electrodecombinations previously qualified by test.

The following AWS Specifications give root pass, fill-in pass, and impactrequirements for electrodes listed in Period 1, 2, and 3 tables:

AWS AS. 1 Specification for Carbon Steel Electrodes for Shielded Metal ArcWelding. [ 15 ]AWS AS. 5 Specification for Low Alloy Steel Electrodes for Shielded Metal ArcWelding. [16]AWS A5.20 Specification for Carbon Steel Electrodes for Flux Cored ArcWelding. [17]AWS A5.29 Specification for Low Alloy Steel Electrodes for Flux Cored ArcWelding. [18]

10

PERIOD 1. PRE-NORTHRIDGE EARTHQUAKE PRACTICE

Conditions. Based on the type of weld, Pre-Northridge Earthquake Practice can bedivided into two categories; 1) Original Weld, and 2) Weld Repair. Weld RepairCombinations probably made up a much larger volume of mixed weld metal thanOriginal Weld Combinations.

Original Welds. Combinations in Table 1 show SMAW and FCAW fill-in passelectrodes welded over SMAW root pass electrodes. Typically the root passelectrodes were selected for good penetration and good fusion. The fill-in passelectrodes were selected for good fusion to base metals and to other passes.Engineers, fabricators, and erectors paid very little attention to electrode impactrequirements; however, they reported no problems with the mixed weldcombinations.

TABLE 1 ORIGINAL WELD COMBINATIONS

ROOT PASS ELECTRODES

AWS CLASSIFICATION

GROUP 1 - KNOWN

FILL-IN PASS ELECTRODES

AWS CLASSIFICATION

COMBINATIONS

Impact Requirements = 20 ft. lbs. at-20°F.

SMAW ElectrodesE701 5E701 6E701 8

Impact requirements = 20 ft. lbs. at-20°F, unless noted.

SMAW ElectrodeE7028

FCAW ElectrodesE70T-4 (No Impact)E7XT-7 (No Impact)E71T-8E70TG-K2

GROUP 2 - POSSIBLE COMBINATIONS

Impact requirements = 20 ft. lbs. at-20°F to -1 50°F.

SMAW ElectrodesE701 5E701 6E701 8E7048E801 5E801 6E801 8

Impact requirements = 20 ft. lbs. at-20°F to -1 00°F.

FCAW ElectrodesE7XT- 1 E70T4- K2E7XT- 5 E70TS-A 1E70T-6 E71T8-NilE71T-8 E71T8-Ni2E7XT-9 E71T8-K6E7XT-12 E8XT-Nil, Ni2E61T8-K6 E80T-Nil, Ni2, K2

Notes: 1. Erectors and fabricators could have welded any Fill-in pass electrodeover any root pass electrode.

11

Weld Repairs. Combinations in Table 2 show SMAW and FCAW weld repairelectrodes welded over SMAW and FCAW original weld electrodes. Fabricators anderectors used a multitude of electrode combinations to make weld repairs to shopand field welds. Repair items included'

Undercut.Cracks, porosity, incomplete fusion, and slag inclusion.Undersized welds.Lamellar tearing (gouging into adjacent weld metal required).

Fabricators and erectors could have made the original weld with SMAW, FCAW,GMAW, or SAW electrodes. SMAW and FCAW electrodes usually made up the repairelectrodes.

TABLE 2 WELD REPAIR COMBINATIONS

ORIGINAL WELD ELECTRODES WELD REPAIR ELECTRODES

AWS CLASSIFICATION AWS CLASSIFICATION

Impact requirements = 20 ft. lbs. at-20°F, unless noted.

SMAW ElectrodesE701 5E7016E7018E7028

FCAW ElectrodesE70T-4 (No Impact)E7XT-7 (No Impact)E7XT-8E70TG-K2

Impact requirements = 20 ft. lbs. at-20°F, unless noted.

SMAW ElectrodesE701 5E701 6E701 8

FCAW ElectrodesE70T-4 (No Impact)E7XT-7 (No Impact)E7XT-8E70TG- K2

Notes: 1.

.

Electrodes listed under "Original Electrodes" are taken from"Group 1-Known Combinations" in Table 1.

Erectors and fabricators could have welded any repair electrodeover any existing weld electrode.

12

PERIOD 2. URGENT NORTHRIDGE EARTHQUAKE DAMAGE REPAIRS

Conditions. Combinations in Table 3 show SMAW repair weld electrodes weldedover SMAW and FCAW electrodes in the existing weld. FCAW electrodes werereadily available for damage repairs. They were very popular, high depositelectrodes. However, for the emergency damage repairs immediately following theearthquake, erectors commonly used SMAW Iow hydrogen electrodes welded overa gouged out joint made with SMAW or FCAW electrodes.

Most of the weld joint damage consisted of a lack of fusion of the weld metal tothe base metal. Erectors repaired this type of damage by:

Back-gouging the joint to clean base metal and clean weld metal.Grinding to clean up.Rewelding fill-in and build-up with E70XX electrodes.

Again, engineers and erectors paid little attention to electrode impactrequirements of the resulting mixed weld metal joint, although they suspectedmost original welds were made with FCAW E70T-4 electrodes that had no impactrequirements. Engineers accepted most of the combinations in Table 3, withacceptance based on normal welding procedures.

TABLE 3URGENT EARTHQUAKE DAMAGE REPAIR COMBINATIONS

EXISTING WELD ELECTRODES REPAIR WELD ELECTRODES

AWS CLASSIFICATION AWS CLASSIFICATION

Impact requirements = 20 ft. lbs. at-20°F, unless noted.

SMAW ElectrodesE701 5E701 6E701 8E7028

FCAW ElectrodesE70T-4 (No Impact)E7XT-7 (no Impact)E7XT-8E70TG-K2

Impact requirements = 20 ft. lbs. at-20°F.

SMAW ElectrodesE701 5E7016E701 8

Notes: 1. Erectors could have welded any repair electrode over any existingweld electrode.

13

PERIOD 3. POST-NORTHRIDGE EARTHQUAKE PRACTICE

Conditions. Based on the type of welding, Post-Northridge Earthquake Practice canbe subdivided into two categories:

Later Northridge Earthquake damage repairs.New welds.

Later Northridge Earthquake Damage Repairs. After completing the urgentNorthridge Earthquake damage repairs, erectors started making "later" damagerepairs. Erectors made these later damage repairs under criteria based on reportsby the Advisory Task Groups. The Advisory Task Groups unanimouslyrecommended that all electrodes in aweld metal combination shall have matchingphysical properties (e.g., yield strength, tensile strength, and elongation) andcompatible impact Requirements--usually 20 ft. lbs. at -20°F. Combinations inTable 4 show SMAW or FCAW repair electrodes welded over SMAW or FCAWelectrodes in the existing weld.

TABLE 4 LATER DAMAGE REPAIR WELD COMBINATIONS

EXISTING WELD ELECTRODES REPAIR WELD ELECTRODES

AWS CLASSIFICATION AWS CLASSIFICATION

Impact requirements = 20 ft. lbs at-20°F, unless noted.

SMAW ElectrodesE701 5E701 6E701 8E7028

FCAW ElectrodesE70T-4 (No Impact)E70T-7 (No Impact)E71T-8E70TG- K2

Impact requirements = 20 ft. lbs. at-20°F.

SMAWElectrodesE7015E7016E7018

FCAW ElectrodesE71T-8E70TG -K2

Notes: 1. Erectors could have welded any repair electrode over any existingweld electrode.

14

NewWelds. Criteria developed by the Advisory Task Groups and new joint designsdeveloped by engineers have increased the use of mixed weld combinations. Thefollowing situations may require mixed weld metal:

1. Back-up bar removal with subsequent fill-in and build-up.2. Beam flange weld to column flange with a shop welded cover plate acting as

a back-up bar.3. Beam web weld to column flange with a shop welded shear plate acting as a

back-up bar.4. Column splice weld over box column shop weld.

Test Program Combinations. The James F. Lincoln Arc Welding Foundation isconducting tests on compatibility of various electrode combinations. See theFoundation's publication "Fabricators' and Erectors' Guide to Welded SteelConstruction" for a discussion of mixing weld metal and for test results. [1 9] Table5.3 in the Guide gives intermixing recommendations. Most combinations listed anthe Table meet the Advisory Task Groups' impact requirements, although someindividual electrodes have no specified impact requirements.

Currently Used Combinations. The authors have helped develop or have learnedof various combinations of mixed weld metal. Table 5 shows some currently usedcombinations of FCAW electrodes welded over SMAW electrodes or other FCAWelectrodes.

15

TABLE 5 NEW WELD COMBINATIONS

ROOT PASS OR FILL-IN PASS APPLICATIONSBASE WELD ELECTRODES OF

ELECTRODES WELD COMBINATIONS

AWS CLASSIFICATION

E701 8 E71T-8 Beam flange to column flange

E70T-6 E71T-8 Make column splice smooth

E70T-6 E71T-8 Overlay from back-up bar removal

E70T-7 E71T-8 Make column splice smooth(No impact)

E70T-1 E71T-8 Column splice weld over shop weld onbox column

E70T-1 E71T-8 Beam flange to column flange weld overshop weld on cover plate

E70T-1 E71T-8 Beam web to column flange weld overshop weld on shear plate

E70T-1 E70T-6 Column splice weld over shop weld onbox column

E70T-1 E70T-6 Beam flange to column flange weld overshop weld on cover plate

I

ETOT-1 E70TG-K2 Column splice weld over shop weld onbox column

E70TG-K2 E70T-6 Beam bottom flange to column flange

E70TG-K2 E71T-8 Beam bottom flange to column flange

E70T-1 E70TG-K2 over Test, Joint B-U4a-GFE71T8-Nil

E71T-8 E70T-6 Test, Joint B-U2a-F

Notes: 1. The fill-in pass electrodes are welded over root pass electrodes orbase weld electrodes.

2. Each row shows a specific weld combination.3. Electrode impact requirements vary from 20 ft. lbs. to 45 ft. lbs. at

-20°F, unless noted.4. New weld repairs also use the combinations shown in Table 5.5. Fabricating shops use the ETOT-I classification (FCAW-g) electrode.6. The Engineer may require approval of any mixed weld metal

combination.

16

7. COMMENTARY ONIMPACT REQUIREMENTSAND IMPACT TESTING , , ,,7

This Article gives the authors' personal views on 'L•' {Iimpact requirements and impact testing. The /7/text in this Article 7 assumes the Engineer /•z/.c.•___•specifies impact requirements and impact '%.testing requirements.

IMPACT REQUIREMENTS

Welding Electrodes. AWSSpecificationsAWSAS.1,AS.5, A5.20, andA$.29 giveimpact requirements for commonly used welding electrodes. See "General Criteria"in Article 6, "Mixed Weld Combinations," for complete Specification titles.Approximately 60 to 75 percent of the specified electrodes have impactrequirements of 20 ft. lbs. at -20°F. So specifying SMAW and FCAW weldingelectrodes with proper impact requirements is not a problem. We recommend theEngineer specify the electrode impact requirements in the project Specifications,just like the Engineer specifies the grade of steel for the project.

Base Metal. Most structural steels in the AISC Specifications have no impactrequirements. An exception is ASTM A6 Steel Groups 4 and 5 hot-rolled shapesand welded sections made of plate with a minimum thickness of 2 in. Theseshapes and sections need impact requirements of 20 ft. lbs. at 70°F, under certaintension stresses and with complete joint penetration welded splices (See Article 3,"AISC Requirements"). These are not strict impact requirements, especially whenmost SMAW and FCAW electrodes have impact properties of 20 ft lbs at -20°F.

AWS Welding Handbook, Figure 12.18 on page 400, "Typical Transition Curve forMild Steel Plate," shows absorbed energy values. [14] For temperate zones,especially if the structural steel frame is enclosed, mild steel and Iow alloy highstrength steels have absorbed energy values of 20 ft. lbs. to 40 ft. lbs. at 25°F to750F.

However, for very Iow temperature zones, (e.g., the North Slope in Alaska, Parts ofCanada, and the Rocky Mountains), the Handbook recommends the Engineerspecify a minimum impact requirement. We will always be indebted to the CharpyV-Notch impact test for pointing out in World War II how cold water temperaturescaused brittle fracture on ships. Recent discoveries and a review of eyewitnessaccounts now confirm the passenger ship TITANIC experienced brittle fracturefailure when colliding with the iceberg and when sinking.

IMPACT TESTING

Charpy V-Notch Test. The Steel Industry extensively uses the Charpy V-Notch

17

Impact Test on its steel products--including weld metal. Test specimens are smallbars--machined, ground, and notched, usually 10mm x 10mm x 55mm in length(0.394 in. x 0.394 in. x 2.165 in.). A specially designed testing machine supportsthe specimens in the horizontal position. A pendulum force strikes and breaks thespecimen with asingle blow, with the pendulum force striking on the side oppositethe notch. The testing machine measures and records the energy absorbed inbreaking the test specimen.

Other Standard Methods. Besides CharpyV-Notch testing, ASTM A370-92 StandardTest Methods and Definitions for Mechanical Testing of Steel Products and ASTME23-88 Standard Test Methods for Notched Bar Impact Tes ting of Metallic Materialsaddress the following other methods of impact testing:

The Izod V-Notch Test (broken in vertical cantilever action).The Drop Weight Test--developed by the U.S. Navy National ResearchLaboratory.The Crack Tip Opening Displacement Test (CTOD). [20,21]

AWS Requirements. Annex III of the AWS Code sets forth the following impacttesting requirements'

Three Specimens. Table II1-1 calls for a set of three test specimens for eachtest location. The Engineer has the responsibility to specify the followingitems on the contract drawings or specifications:

Test temperature.Minimum average energy value per set of three (location).Minimum energy value per specimen from any set.

Five Specimens. An optional test--probably used Jn 75 to 80 percent oftests--allows a set of five test specimens for each location with the highestand lowest values discarded. The result is the average value for the threemiddle specimens. Discarding the highest and lowest values minimizes thevariations (scatter) normally associated with Charpy V-Notch test results ofwelds and Heat Affected Zone (HAZ). See Table II1-1, Note No. 2.

Specimen Location. Figure II1-1 notes the locations of the test specimensfrom the weld centerline, the Heat Affected Zone, and the weld face.

Scatter. Charpy V-Notch test results have large variations (great scatter) becauseof many potential differences in testing procedures This scatter of test resultssets up a difficult situation to make a judgment when only a single specimen istested at each location. Unfortunately the evidence--or lack of evidence--indicatesthe single test is generally the procedure followed. Differences in testingprocedures contributing to scatter include:

Material strength and thickness.Heat input of the weld specimen.Roiling direction of grain orientation.Variations in testing procedures.Small specimens.

18

Locations of tests.Personnel making the tests.

ASTM Codes. ASTM A370-92 and ASTM E23-88 give impact requirement testingprocedures for the various impact testing methods. The Codes also alert theEngineer to be careful in comparing the results of impact tests, including steel testspecimens machined from the same heat number lot. See:

Annex AS, "Notes on Significance of Notch-Bar Impact Testing" in ASTMA370-96.Appendix Xl, "Notes on Significance of Notched-Bar Impact Testing" in ASTME23-88. (Applies to all steel products.)

Notes Relating to the ASTM Codes.

. The Charpy V-Notch (CVN) impact test is especially appropriate for minimumoperating temperatures and maximum in service rates of loading.

. The notch behavior of face-centered cubic metals does show a broadrelationship of tensile test results. In contrast, body-centered cubic ferritesteel test results show very little relationship between tensile test and CVNimpact test results.

The property that keeps a notched-bar from cleaving (holds together underload), is its "cohesive strength." The bar fractures when the normal stressexceeds the cohesive strength. Fracture without the bar deforming is thecondition for brittle fracture.

Usually plastic deformation precedesfadure. Besides the normal stress, theapplied load also sets up shear stresses that are about 45 degrees to thenormal stress. Elastic behavior ends when the shear stress exceeds the shearstrength of the material and when deformation or plasuc yielding sets in.Fracture with the bar deforming is the condition for ductile failure.

. Size effect of the test specimens is another source of differences that causevariations in test results. The larger the specimen, the higher value the testresults; however, an increase in width will also increase the restraint of thenotch action tending to reduce the absorbed energy

. The temperature effect has great influence on the notched specimenbehavior. Steel temperature at the time of the test must be known, and theabsorbed energy test results must be recorded and compared torequirements--_ft , lbs. at_°F. Temperature influence is especially true forbody centered cubic ferrite steels.

. The testing machine also contributes to variations in test result valuesthrough items like:

Machine rigidity.Support anvil detail.

19

.

.

.

Pendulum striking of the specimen (not squarely).Details of the machine anchor bolts.

While Charpy or Izod tests may not directly predict the ductile or brittlebehavior of the steel specimens or of large masses (large structures), the testresults can serve as acceptance criteria.

The Engineer must recognize that the project Specifications in the BidDocuments should specify:

The dimensional detail of the specimens.Base metal material.Weld deposit material.The testing procedure.

The Engineer must know a structure's operating conditions, and set the testresults the Engineer is trying to achieve The engineer should also bethoroughly familiar with typical absorbed energy transition curves (ft. lbs. at°F) for the types of steel to be use on the project.

20

REFERENCE

1. Structural

2. Structural

3. Structural

4. Structural

5. Structural

6. Structural

7.

.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

LIST

Welding Code--Steel D1.1-96, AWS, Miami, 1996.

Welding Code--Steel, p. 41.

Welding Code--Steel, p. 235.

Welding Code--Steel, p. 235.

Welding Code--Steel, p. 235.

Welding Code--Steel, p. 237.

ManualofSteelConstruction:AIIowableStress Design, 9th ed., AISC, Chicago,1989.

Manual of Steel Construction

Manual of Steel Construction

Manual of Steel Construction

Manual of Steel Construction

Manual of Steel Construction

Manual of Steel Construction p

Welding Handbook, 8th ed., Vol

p. 5-69

p. 5-165.

p. 5-26.

p. 5-27.

p. 5-29.

5-30.

1, AWS, Miami, 1987.

AWS A5.1 Specification for Carbon Steel Electrodes for Shielded Metal ArcWelding, AWS, Miami, 1991.

AWS AS. 5 Specification for Low Alloy Steel Electrodes for Shielded Metal ArcWelding, AWS, Miami, 1996.

AWS A5.20 Specification for Carbon Steel Electrodes for Flux Cored ArcWelding, AWS, Miami, 1980.

AWS A5.29 Specification for Low Alloy Steel Electrodes for Flux Cored ArcWelding, AWS, Miami, 1995.

Fabricators' and Erectors' Guide to Welded Steel Construction, The James FLincoln Arc Welding Foundation, Cleveland, 1997.

ASTM A370-92' Standard Test Methods and Definitions for MechanicalTesting of Steel Products, ASTM, Philadelphia, 1992.

ASTM E23-88' Standard Test Methods for Notched Bar Impact Testing ofMetallic Materials, ASTM, Philadelphia, 1988.

21

ABOUT THE AUTHORS

Alvaro L. Collin is a Consulting Engineer with California registration in CivilEngineering and Metallurgical Engineering. He received a BS degree from theUniversity of California, Berkeley, in 1941 as a Civil Engineering major and aMechanical Engineering minor. He spent 24 years with Kaiser Steel Corporation asManager of Engineering of the Fabrication Division, Southern California, and SeniorDevelopment Engineer, Steel manufacturing Division, Oakland, CA. Al has beenconsulting the past 17years on welded construction, heavy equipment design andmaterial handling systems.

Mr. Collin is a life member of the Structural Engineers Association of NorthernCalifornia. He has been a member of the Board of Directors and the Steel andSeismic committees of SEAONC. He is a Iongtime member of the AmericanWelding Society, having served on the National Board of Directors, on the NationalQualification and Certification Committee, and as chairman to the Los Angeles andSan Francisco sections. Al has been awarded the National, District and SectionMeritorious Awards of AWS. Recently, he was awarded national honorarymembership in theAWS. He has served on AISC and AISI Code Committee TaskGroups, the SAC Joint Venture Task Group, and is a member of the EarthquakeEngineering Research Institute.

James J. Putkey is a consulting civil engineer in Moraga, California. He receiveda BCE degree from the University of Santa Clara in 1954. He served two years inthe U.S. Army, and then spent 19 years with the Erection Department of BethlehemSteel Corporation--Pacific Coast Division, and seven years with the University ofCalifornia--Office of the President. After leaving the University, Jim started his ownconsulting business. He has provided consulting services to owners, contractors,attorneys, and steel erectors for the past 1 7 years.

Jim is now "Semi-Retired." However, he still serves as a hearing officer for theUniversity of California, and occasionally writes construction related articles.

22

STRUCTURAL STEEL EDUCATIONAL COUNCIL

TECHNICAL INFORMATION & PRODUCT SERVICE

JANUARY 1997

QUICK REFERENCE GUIDE

for

STRUCTURAL STEEL WELDINGPRACTICES

This Guide provides a quick reference to codes and standards for selectedtopics regarding the welding of structural steel. The Guide mainlyaddresses welding practices, but contains some related design topics.Topics are grouped alphabetically, within the eight section titles in theStructural Welding Code--Steel, AWS D1.1-96.

The Guide consists of a Reference Table, Reference List, and Index ofTopics. To use the Guide, 1) locate the subject matter in the Index ofTopics, 2) find the topic in the Reference Table and note the correspondingreference list number and location, and 3) find the title of the referencelist number in the Reference List. Not all topics and topic locations arelisted. However, the topics listed and locations given should allow theuser to find most subject matter on welding practices.

The information presented in this Steel TIPs is for general informationonly, and should not be used without independent examination andverification of its suitability by the user.

REFERENCE TABLE

TOPIC REFERENCE CONTENT REFERENCE LIST

NO. LOCATION

1. GENERAL REQUIREMENTS

Safety Hazards present an welding and 1 Pg 2, Annex Jcutting 2a Pg 520

13 Pg 15 1-1

Structural Steel, Defined Elements of steel frame that 4 Pg 5support design load

Symbols Symbols to indicate joint type, 1 Pg 2, 3size, length, position, location, 2a Pg 195and process 3 Pg 4-152 to 155

6 Pg 6-14, 15, 1913 Pg 16 1-39

Terms and Definlbons Definlbons for terms used En the 1 Annex BAWS Code and for general 2a Pg 554welding terms 4 Pg 3

9 Pg 7113 Pg 16 1-1

2. DESIGN OF WELDED CONNECTIONS

Engineer's Role m Minimizing Steels to use, weldments to 10 Pg 18Weld Defects avoid, and submittals to require

Field Instruchons Informabon to erector on making 1 Pg 3welds 6 Pg 6-40

Lamellar Tearing Defimtlon, causes, and 1 Pg 3prevention of this phenomenon 2a Pg 137

5 Pg 2-198 Pg 449 Pg 6110 Pg 1613 Pg 6 1-9

Notches and Bnttle Fracture Locabons where notches can 1 Pg 162occur, and details to avoid 5 Pg 2-21notches

Sequence of Joint Welding When and how to convey welding 1 Pg 3, 230sequence 2a Pg 136

9 Pg 37, 67, 7110 Pg 20

Welded Joints (Connections) Requirements for design of I Pg 3welded joints and connections 3 Pg 4-152

5 Pg 2-3

REFERENCE TABLE

TOPIC REFERENCE CONTENT REFERENCE LIST

NO. LOCATION

3. PREQUALIFICATION OF WPSs

Fillet Welds Requirements and techniques. 1 Pg 49, 160, 1692a Pg 1545 Pg 2-48 Pg 31

Groove Welds Requirements and technaques 1 Pg 49, 1692a Pg 1495 Pg 2-5

Plug and Slot Welds Requirements and techn,ques 1 Pg 49, 169

Prequallfied Joint Details Joint preparation details for joints 1 Pg 52 to 99m prequahfied welding procedure 3 Pg 4-153 to 173specifications 6 Pg 6-12

11 Pg 29

Root Layer L•mltatlons on peenIng 1 Pg 1729 Pg 67

Root Opening Tolerances for root openings 1 Pg 49, 16413 Pg 2 2-6

Stnnger Beads What they are and why they are 1 Annex B(Pg 282)used. 9 Pg 67

Welding Procedure Spec•ficahons Scope and requirements I Pg 41,239(WPSs) 2a Pg 439

5 Pg 2-310 Pg 2612 Pg D-5714 All15 All

Welding Processes Common welding processes for 1 Pg 41, 51both prequahfied and quahfied by 2b Alltest 5 Pg 2-12

6 Pg 6-310 Pg 1213 Pg 5 0-1, 5 6-1

Electrogas Welding (EGW) Automahc, sohd or flux core wire 1 Pg 41, 105, 157electrode with gas shielding 2a Pg 12

2b Pg 2345 Pg 2-1713 Pg 5.5-3

REFERENCE TABLE

TOPIC REFERENCE CONTENT REFERENCE LIST

NO. LOCATION

Electroslag Weldtng ( E S W ) AutomatEc, solid were electrode(s) 1 Pg 41, 105, 157that melt m slag pool (not an arc 2a Pg 13welding process) 2b Pg 272

5 Pg 2-1710 Pg 1313 Pg 5 5-1

Flux Cored Arc W e l d i n g Sema-automatlc, with flux in core 1 Pg 41, 102(FCAW) of tubular electrode 2a Pg 7

2b Pg 1585 Pg 2-1610 Pg 1213 Pg 5 3-1

Gas Metal Arc W e l d i n g Seml-automahc, sohd wire 1 Pg 41, 102(GMAW) electrode with gas shleldmg 2a Pg 7

2b Pg 1105 Pg 2-1610 Pg 1313 Pg 5 4-2

Shielded Metal Arc Welding Manual, flux coated "sttck" 1 Pg 41, 102(SMAW) electrode 2a Pg 4

2b Pg 445 Pg 2-1210 Pg 1213 Pg 5 1-1

Submerged Arc Welding Automabc or semi-automatic, 1 Pg 41, 102(SAW) sohd wire electrode with arc 2a Pg 5

submerged m a granular flux 2b Pg 1925 Pg 2-1510 Pg 1313 Pg 5 2-1

4. QUALIFICATION

Potations of Welds and Welding Onentatlon and hmltahons of the 1 Pg 51, 102, 103,Potations four positions for groove and fillet 104

Flat welds, and corresponding 5 Pg 2-13Honzontal welding positions 6 Pg 6-13OverheadVertical

Quahficahon of Performance Requirements for quahficatlon 1 Pg 137(Welder) tests of welders 2a Pg 438, 455

8 Pg 3013 Pg 11 4-1

REFERENCE TABLE

TOPIC REFERENCE CONTENT REFERENCE LIST

NO. LOCATION

Qualification of W e l d i n g Requirements for quahficat•on 1 Pg 102Procedure Specifications tests for welding procedure 2a Pg 438, 451

specifications 8 Pg 3013 Pg 11.3-1, 6 2-1

5. FABRICATION

Arc Strikes Dispos•t•on of arc strikes. 1 Pg 1728 Pg 31

Backing (Backing Bars) Requirements and sketch of 1 Pg 159backing 5 Pg 2-49

13 Pg 2 2-6

Backing Bar Tack Welds Sketch of backing bar tack welds 5 Pg 2-23

Base Metal Approved base metal for 1 Pg 155,structure, weld tabs, backing, and Annex Mspacers 2a Pg 111

Cleaning Welds Slag removal between passes 1 Pg 172and on completed weld 13 Pg 2 1-5

Clearance for Welding Requirements for positioning I Pg 171electrode 5 Pg 2-55

6 Pg 6-38

Cracking, Cracks Causes and repair 1 Pg 171,Annex XI(Pg263)

10 Pg 1413 Pg 6 3-17

Discontinuities and Defects Avoiding and correcting defects I Pg 171, 174,175, 183

2a Pg 36410 Pg 14, 2113 Several Iocabons

D•stortlon and Shrinkage How to control and correct 1 Pg 1642a Pg 136, 2188 Pg 4310 Pg 2213 Pg 3 1-1

Electrodes (Consumables) Combinations with base metal, 1 Pg 41, 155,requirements, and classifications Annex M

6 6-710 Pg 1513 Pg 4 1-1

Environment Allowable temperature, wetness, 1 Pg 160wind

4

REFERENCE TABLE

TOPIC REFERENCE CONTENT REFERENCE LIST

NO. LOCATION

Equipment for Welding and Requirements 1 Pg 159Cutting 2a Pg 9

2b Pg 213 Pg 4 2-1, 4 3-1

Fitting Objecbves and inspection of 7 Pg 17placing parts together 13 Pg 4 4-1

Highly Restrained Welded Joints Causes and mltlgabon of 1 Annex XIrestrained joints (Pg 266)

CX12 (Pg 426)3 Pg 4-1529 All11 Pg 33

Peemng Reasons for and hmltat•ons 1 Pg 1729 Pg 6710 Pg 2013 Pg 3 1-7

Preheat and Interpass Reasons for, minimum 1 Pg 45, 158Temperatures temperatures, and other 10 Pg 16

requirements 13 Pg 3 3-1

Preparation of Base Metal Edge dascontmu•hes and their 1 Pg 160hm•tatlons 8 Pg 7, 8

Stress Relief Requirements for stress rehef by 1 Pg 158heat treatment 10 Pg 16, 18

13 Pg 33-1

Temporary and Tack Welds General requtrements for 1 Pg 163placement and removal 8 Pg 14

Thermal Cutting Processes and capabdltles 2a Pg 272b Pg 450, 4825 Pg 2-1713 Pg 13 5-1

Tolerance of Joint Dimensions Joint assembly requirements 1 Pg 164including sketches 7 Pg 50

Weld Tabs (and Weld Dams) Use and sketches on how to 1 Pg 172place 5 Pg 2-48, 49

8 Pg 3212 Pg B-6

REFERENCE TABLE

TOPIC REFERENCE CONTENT REFERENCE LIST

NO. LOCATION

Weldablhty Required capacltaes for steel to 1 Annex B(Pg 284)be welded 2a Pg 119

5 Pg 2-710 Pg 1113 Pg 6 1-1

Workmanship, Techmques Proper welding practmes. 1 All8 Pg 299 Pg 6710 Pg 1411 Pg 3313 Several Iocabons

6. INSPECTION

Inspection Checks Vanous items to check dunng 7 Pg 20welding

Inspectors and Inspecbon General requirements, and 1 Pg 173, 193, 403Objectives objectives of welding mspecbon 2a Pg 466

5 Pg 2-247 Pg 209 Pg 6813 Pg 11 2-1

Nondestructive Testing Descnptlon, use, and 1 Pg 175, 195(Examination) requirements 2a Pg 468

7 Pg 388 Pg 4110 Pg 23

L•quld Penetrant Inspection Descnphon and requirements 1 Pg 175, 1932a Pg 4735 Pg 2-24, 2510 Pg 2513 Pg 11 2-13

Magnetic Particle Inspection Descnptlon and requirements 1 Pg 175, 1932a Pg 4785 Pg 2-24, 2613 Pg 11 2-11

Radlographm Inspection Descnphon and requirements 1 Pg 177, 1952a Pg 4865 Pg 2-24, 2610 Pg 2513 Pg 11 2-6

REFERENCE TABLE

TOPIC REFERENCE CONTENT REFERENCE LIST

NO. LOCATION

Ultrasonic Inspection Description and requirements 1 Pg 183, 2042a Pg 5025 Pg 2-24, 2710 Pg 2413 Pg 11 2-14

Visual Inspect]on Descrlpbon and requirements 1 Pg 1752a Pg 4695 Pg 2-24, 257 Pg 2710 Pg 2413 Pg 11 2-3

Quahty Control Quality and an example of a 2a Pg 360, 462quahty control program 12 Pg E-17

13 Pg 11 1-1

7. STUD WELDING

Repairing Stud Welds Completion of flash by fillet weld 1 Pg 225, 228

Stud Arc Welding General requfrements and 1 Pg 223description 2b Pg 300

13 Pg 5 5-3

Visual Inspection and 15° Bend Bend test for studs that do not 1 Pg 228Test show full 360° flash

8. STRENGTHENING AND REPAIRING EXISTING STRUCTURES

Strengthening and Repairing General requirements 1 Pg 229Provisions

Work Plan Engineer required to prepare a 1 Pg 229comprehensive plan for the work

REFERENCE LIST

1. Structural Welding Code--Steel, ANSI/AWS D1.1-96, Miami, 1996.

2.a. Welding Handbook, 8th ed., Vol. 1, AWS, Miami, 1987.

2.b. Welding Handbook, 8th ed., Vol. 2, AWS, Miami, 1991.

3. Manual of Steel Construction: Allowable Stress Design, 9th ed., AISC, Chicago, 1989.

4. Code of Standard Practice for Steel Buildings and Bridges, AISC, Chicago, June 10, 1992.

5. Manual of Steel Construction, Vol. II--Connections ASD 9th ed./LRFD 1st ed., AISC, Chicago,1992.

6. Detailing for Steel Construct/on, AISC, Chicago, 1983.

7. Shop Inspector Training Guide, AISC, Chicago, 1985.

8. Quality Criteria and Inspection Standards, 3d ed., AISC, Chicago, 1988.

9. "Commentary on H•ghly Restrained Welded Connections," AISC Engineering Journal, 3dQuarter 1973/Vol. 10, No. 3, Chicago, 61-73; and "Discussion," AISC Engineering Journal, 1stQuarter 1975/Vol. 12, No. 1, Chicago, 36-68.

10. F. Robert Preece and Alvaro L. Collin, "Structural Steel Construction in the '9Os,' Steel TIPS,Structural Steel Education Council, Walnut Creek, California, September 1991.

11. James J. Putkey, " C o m m o n Steel Erection Problems and Suggested Solutions," Steel TIPS,Structural Steel Educational Council, Moraga, Califorma, December 1993.

12. Steel Moment Frame Connection, Advisory No. 3, SAC Joint Venture, Sacramento, 1995.

13. The Procedure Handbook of Arc Welding, 13th ed., The Lincoln Electric Company, Cleveland,1994.

14. Guidelines for Welding Procedure Spec/f/cations, SEAONC, San Francisco, October 1996.

15. Alvaro L. Collin, Welding Procedure Spec•fmatlons, Orinda, California, November 1995 (availablefrom Structural Steel Educational Council).

Index of Steel TIPS Publications

The following is a hst of avatlable Steel TIPS Copies wdl be sent upon request Some are in very hm•ted quanbty

· Seasmac Design of Special Concentrically Braced Frames· Selsmm Design of Bolted Steel Moment-Resisting Frames· Structural Detads to Increase Ductdlty of Connections· Slotted Bolted Connecbon Energy D•ss•paters· Use of Steel in the Seismic Retrofit of H•stonc Oakland C•ty Hall· Heavy Structural Shapes Jn Steel Tension Apphcattons· Economical Use of Cambered Steel Beams· Value Engineering and Steel Economy· What Desagn Engineers Can Do to Reduce Fabrication Costs· Charts for Strong Column Weak G•rder Design of Steel Frames· Se•smm Strengthemng w•th Steel Slotted Bolt Connechons· Se•smm Dessgn Pracbce for Eccentrically Braced Frames

INDEX OF TOPICS

Arc Stnkes, 4Backing (Backing Bars), 4Backing Bar Tack Welds, 4Base Metal, 4Cleaning Welds, 4Clearance for Welding, 4Cracking, Cracks, 4DESIGN OF WELDED CONNECTIONS, 1DiscontinuJhes and Defects, 4Distortion and Shrmkage, 4Electrodes (Consumables), 4Electrogas Welding (EGW), 2Electroslag Welding (ESW), 3Engineer's Role in Minimizing Weld Defects, 1Environment, 4Equipment for Welding and Cutting, 5FABRICATION, 4Field Instructions, 1Fillet Welds, 2Fitting, 5Flat Welds, 3Flux Cored Arc Welding (FCAW), 3Gas Metal Arc Welding (GMAW), 3GENERAL REQUIREMENTS, 1Groove Welds, 2Highly Restrained Welded Joints, 5Horizontal Welds, 3INSPECTION, 6Inspechon Checks, 6Inspectors and Inspection ObJectives, 6Lamellar Tearing, 1LJquad Penetrant Inspection, 6Magnetic Particle Inspection, 6Nondestructive Tesbng (Exammahon), 6Notches and Brittle Fracture, 1Overhead Welds, 3Peening, 5Plug and Slot Welds, 2Poslhons of Welds and Welding Poslhons, 3Preheat and Interpass Temperatures, 5Preparation of Base Metal, 5PREQUALIFICATION OF WPSs, 2Prequahfied Joint Details, 2QUALIFICATION, 3Quahficahon of Performance (Welder), 3Qualification of Welding Procedure

Specifications, 4

Quality Control, 7Radlographac Inspecbon, 6Repamng Stud Welds, 7Root Layer, 2Root Opening, 2Safety, 1Sequence of Joint Welding, 1Shielded Metal Arc Welding (SMAW), 3STRENGTHENING AND REPAIRING

EXISTING STRUCTURES, 7Strengthening and Repalnng Prowslons, 7Stress Rehef, 5Stringer Beads, 2Structural Steel, Defined, 1Stud Arc Welding, 7STUD WELDING, 7Submerged Arc Welding (SAW), 3Symbols, 1Temporary and Tack Welds, 5Terms and Defin•ttons, 1Thermal Cutting, 5Tolerance of Joint Dimensions, 5Ultrasonic Inspection, 7Vertmal Welds, 3Visual Inspection, 7Visual Inspechon and 15° Bend Test, 7Weld Tabs (and Weld Dams), 5Weldablhty, 6Welded Joints (Connections), 1Welding Procedure Speclficabons (WPSs), 2Welding Processes, 2Work Plan, 7Workmanship, Techniques, 6

Note Topms m bold caps (e g, INSPECTION)refer to the eight AWS Code sechon htles