SRS-429
-
Upload
valdemir-colares -
Category
Documents
-
view
15 -
download
0
description
Transcript of SRS-429
un.U-ENG-76-2022
CIVIL ENGINEERING STUDIES STRUCTURAL RESEARCH SERIES NO. 429
ORNL/SUB/4161-1 DIST. CATEGORY UC-77
SHEAR STRENGTH OF END SLABS OF PRESTRESSED CONCRETE NUCLEAR REACTOR VESSELS
By J. D. REINS, J. L. QUIROS, JR.
W. C. SCHNOBRICH and M. A. SOZEN
A Report on an Investigation Carried Out as Part of the Prestressed Concrete Reactor Vessel Program of the Oak Ridge National laboratory Operated by the Union Carbide Corporation for the Energy Research and Development Administration
UNIVERSITY OF ILLINOIS at URBANA-CHAMPAIGN URBANA, ILLINOIS JULY 1976
Printed in the United States of America. Available from
National Technical Information Service
U.s. Department of Commerce
5285 Port Royal Road, Springfield, Virginia 22161
Price: Printed Copy $7.00 Microfiche $4000
This report was prepared as an account of work sponsored by the United States Government. Neither the United States nor the Energy Research and Development Administration, nor any of their employees, nor any of their contractors, subcontractors, or their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness or usefulness of any information, apparatus, product or process disclosed, or represents that its use would.not infringe privately owned rights.
CONTEfJTS
SHEAR STRENGTH OF END SLABS OF PRESTRESSED CONCRETE NUCLEAR REACTOR VESSELS
1. INTRODUCTION .•.................. '.' ........................... .
Page
1 . 1 Obj ect and Scope......................................... 1 1 .2 Acknowl edgments. . . . . . . . . • . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . . . 2
2. OUTLINE OF THE INVESTIGATION... ...... ... ......... ... .....• .... 4
3. COMPUTER ANALYSIS OF END SLABS............... ...... ...... ..... 6
3. 1 I n t ro d u c t ion. . . . . . • . . . • • . . . . . . . . . • 0 • • • 0 • • 0 0 • • • • • • • • • • • 0 • • 6 3. 2 Linear Three Dimensional Analysis .•..•...... 0 .... 0....... 6 3.3 Two Dimensional Axi symmetric r~1odel........... ...•..... ... 11 3.4 Results ... e ••••••• 0 •••••••••••• 0 ••••••••••••••••• 0 0 •••• 0. 12 3.5 Stress State in Cryptodome Head •..... O. • • • • • • • • • • • • • • • • • • 14
4. OVERALL RESPONSE AND FAI LURE f\1ECHANI S~1S 0 • 0 •••••••••• 0 • 0 • • • • • • • 17
5. STRA INS t~EASURED ON HALLS OF PENETRATIONS •.............•. 0 •• 0.. •••••• ••• ••• ••• ••• •••••• ••• 22
5. 1 5.2
5.3
5.4
I nt roductory Remarks ... 0 •••••••••••••••••••••••••••••••••
Strains Measured in End Slab of Vessel PV30 •....... O •• 0 •••••••••••••••••••••••••••••••••
Strains Measured in the End Slab of Vessel PV32 .................... 0 ••••••••• 0 ••••••••••••••
22
23
26 Comparison of Measured and Calculated Strains ..•.....•............ o ••••••••••••• GO............ 27
6. SUMMARY 0 • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • •• 29
6.1 Analytical Studies........................................ 29 6.2 Experimenta 1 Studi es. . . . . . . . . . . . . . . . . . . . . . . . . . . . . • . . • . . . .. 30
7. REFERENCES..................................................... 32
TABLES
FIGURES
APPENDIX A
APPENDIX B
LIST OF FIGURES
Figure No. Page
l.la Vessels Tested by Karlsson ......................... ~ ... ~ .. 35
l.lb Vessels Tested by Karlsson ................................. 36
2.1 Penetration Patterns and Cross-section of Vessel .................................................. 37
3.1 Finite Element Grid Used in Three Dimensional Analysis ............... D· ................................. 38
3.2 Boundary Conditions Used in Three Dimensional Analysis ................................................ 39
3.3a Grid for Vessel with 6 - 511 Penetrations .................. 40
3.3b Grid for Vessel with 37 - 211 Penetrations ................. 40
3.4a Vertical Deflections of lO-in. Slab ....................... 41
3.4b Vertical Deflections of 12.5-in. Slab ..................... 42
3.5a Radial Stress in la-in. Slab with 5-in. Diameter Penetrations ................................... 43
3.5b Hoop Stress in 10-in. Slab with 5-in. Di ameter Penetrati ons. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 44
3.5c Shear Stress in la-in. Slab with 5-in. Diameter Penetrations ................................... 45
3.6a Radial Stress in 12.5-in. Slab with 5-in. Diameter Penetrations .................................... 46
3.6b Hoop Stress in 12.5-in. Slab with 5-in. Diameter Penetrations ................................... 47
3.6c Shear Stress in l2.5-in. Slab with 5-in. Diameter Penetrations ................................... 48
3.7 Comparison of Stresses Calculated by Axi-symmetric and by Three Dimensional Analyses for l2.5-in. Thick Head ........................ 49
3.8 Boundary Conditions Used in Cryptodome Analysis ................................................ 50
3.9a Radial Stresses in Cryptodome Shaped from 10- in. Slab............................................. 51
LIST OF FIGURES (cont1d)
Figure No. Page
3.9b Hoop Stresses in Cryptodome Shaped from lO-in. Slab................................................... 52
3.9c Shear Stresses in Cryptodome Shaped from lO-in. Slab............................................ 53
3.10 Principal Stresses on a Section Adjacent to Penetration Through 10-in. Head ........................ 54
4.1 Measured Relationships Between Internal Pressure and Total Deflection at Midspan of End Slab for Test Vessels with 10-in. End Slabs ................. 55
4.2 Measured Relationships Between Internal Pressure and Total Deflection at Midspan of End Slab for Test Vessels with l2.5-in. End Slabs ................... 56
4.3 Deflected Shape of End Slab for Vessels with Solid Heads............................................ 57
4.4 Deflected Shape of End Slab for Vessels with 5-in. Penetrations..................................... 58
4.5 Deflected Shape of End Slab for Vessels with 2-in. Penetrations..................................... 59
4.6 End Slab after Test, PV26 ................................ 60
4.7 End SI ab after Test, PV27................................ 61
4.8 End Slab after Test PV29 ............................... '0 62
4.9 End Sl ab after Test PV28 ...................... ~ . . . . . . . . . . 63
4.10 End Slab after Test PV30 ................................. 64
4.11 End Slab after Test PV31 ..... 0 •• 0 •••••••••••••••••••••••• 65
4.12 End Slab after Test PV32 ... 0 .. 0 .............. 0........... 66
4.13 End Slab after Test PV33 ......................... 0 ••••••• 67
4.14 Failure Prestress and Maximum Nominal Shear Stresses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
5.la Location of Strain Gages in End Slab with 2-in. Penetrations ................................. 0... 69
5.1b Location of Strain Rosettes in the Vertical Plane ............................................ 0.. ... 70
LIST OF FIGURES (cont'd)
Figure No. Page
5.2 Measured Vertical Strains, Level 1 Penetrations 1 and 3, PV30 ....................................... '.. 71
5.3 Measured Horizontal Strains, Level 1 Penetrations 1 and 3, PV30 ........... ~ ............... 71
5.4 Measured Compression Diagonal Strains, Leve 1 1 Penetrati ons 1 and 3 PV30.................... 72
5.5 Measured Tension Diagonal Strains, Leve 1 1 Penetra ti ons 1 and 3 PV30.................... 72
5.6 Measured Vertical Strains, Level 2 Penetrations 1 and 3, PV30 ........................... 73
5.7 Measured Horizontal Strains, Level 2 Penetrations 1 and 3, PV30 ........................ 00' 73
5.8 Measured Compression Diagonal Strains, Level 2 Penetrations 1 and 3, PV30 ... 0............... 74
5.9 Measured Tension Diagonal Strains, Level 2 Penetrations 1 and 3, PV30 ......................... 74
5.10 Measured Vertical Strains, Level 3 Penetrations 1 and 3, PV30 ........................... 75
5. 11 Measured Horizontal Strains, Level 3 Penetrations 1 and 3, PV30 ........................... 75
5.12 Measured Compression Diagonal Strains, Level 3 Penetrations 1 and 3, PV30 ..... ~ ............. 76
5.13 Measured Tension Diagonal Strain, Level 3 Penetrations 1 and 3, PV30 ........................... 76
5. 14 Measured Vertical Strains, Levels 1 and 2 Penetration 2, PV30.................................. 77
5.15 Measured Horizontal Strains, Levels 1 and 2 Penetration 2, PV30................................ 77
5.16 Measured Diagonal Strains, Levels 1 and 2 Penetration 2, PV30 ................................ 78
5.17 Measured Vertical Strain, Level 3 Penetration 2, PV30 .................................. 78
5.18 Measured Horizontal and Diagonal Strains Level 3, Penetration 2, PV30 ......................... 79
LIST OF FIGURES (cont1d)
Fi gure No. Page
5.19 Measured Vertical Strains, Level 1 Penetrations 4 and 5, PV30 ......................... 79
5.20 Measured Horizontal Strains, Level 1 Penetrations 4 and 5, PV30 ......................... 80
5.21 Measured Compression Diagonal Strains, Level 1 Penetrati ons 4 and 5, PV30................. 80
5.22 Measured Tension Diagonal Strains, Levell Penetrations 4 and 5, PV30................. 81
5.23 Measured Vertical Strains, Level 2 Penetrations 4 and 5, PV30 ......................... 81
5.24 Measured Horizontal Strains, Level 2 Penetrations 4 and 5, PV30 ......................... 82
5.25 Measured Compression Diagonal Strains, Level 2 Penetrati ons 4 and 5, PV30................. 82
5.26 Measured Tension Diagonal Strains, Level 2 Penetrati ons 4 and 5, PV30................. 83
5.27 Measured Vertical Strains, Level 3 Penetrations 4 and 5, PV30 ......................... 83
5.28 Measured Horizontal Strains, Level 3 Penetrations 4 and 5, PV30 ......................... 84
5.29 Measured Compression Diagonal Strains, Level 3 Penetrations 4 and 5, PV30....... ........... 84
5.30 Measured Tension Diagonal Strains, Level 3 Penetrations 4 and 5, PV30................. 85
5.3la Location of Strain Gages in End Slab with 5-in. Penetrations................................. 86
5.3lb Location of Strain Gages on Inside Face of End Slab with 5-in. Penetrations................... 87
5.3lc Location of Strain Rosettes in the Vertical Plane.... ... ... ... ... ...... ... ... .. . ... .. . ......... 88
5.32 Measured Vertical Strains, Levell 12:00 and 6: 00 Pos i tions, PV32o ............ 0 • • • • • • • • • • • • • 89
5.33 Measured Vertical Strains, Levell 3:00 and 9:00 Position~, PV32 ...... '0 ••••••••••••••••••• 89
LIST OF FIGURES (cont'd)
Figure No. Page
5.34 Measured Horizontal Strains, Levell 12:00 and 6:00 Positions, PV32 ....................... 90
5.35 Measured Horizontal Strains, Levell 3:00 and 9:00 Positions, PV32 ................... '. . . . . . . . . . . . . . . . .. . . . . . . . . . . .. 90
5.36 Measured Diagonal Strains, Levell. 12:00 and 6:00 Positions, PV32 ....................... 91
5.37 Measured Tension Diagonal Strains, Levell 3:00 Position, PV32 ...... ~ .................. 91
5.38 Measured Compression Diagonal Strains, Levell 9:00 Position, PV32 ......................... 92
5.39 Measured Vertical Strains, Level 2 12:00 and 6:00 Positions, PV32 ....................... 92
5.40 Measured Vertical Strains, Level 2 3:00 and 9:00 Positions, PV32 ........................ 93
5.41 Measured Horizontal Strains, Level 2 12:00 and 6:00 Positions, PV32 ....................... 93
5.42 Measured Horizontal Strains, Level 2 3:00 and 9:00 Positions, PV32 ........................ 94
5.43 Measured Diagonal Strains, Level 2 12:00 and 6:00 Positions, PV32 ....................... 94
5.44 Measured Tension Diagonal Strains, Level 2 3:00 Position, PV32 .......................... 95
5.45 Measured Compression Diagonal Strains, Level 2 9:00 Position, PV32 .......................... 95
5.46 Measured Vertical Strains, Level 3 12:00 and 6:00 Positions, PV32 ....................... 96
5.47 Measured Vertical Strains, Level 3 3:00 and 9:00 Positions, PV32 ........................ 96
5.48 Measured Horizontal Strains, Level 3 12:00 and 6:00 Positions, PV32 ....................... 97
5.49 Measured Horizontal Strains, Level 3 3:00 and 9:00 Positions, PV32 ........... D •••••••••••• 97
LIST OF FIGURES (cont'd)
Figure No. Page
5.50 Measured Diagonal Strains, Level 3 12:00 and 6:00 Positions, PV32 ...................... '98
5.51 Measured Tension Diagonal Strains, Level 3 3:00 Position, PV32 ........................ 98
5.52 Measured Compression Diagonal Strains, Level 3 9:00 Position, PV32 ......................... 99
5.53 Comparison of Measured and Calculated Strains at an Internal Pressure of 1000-psi for a lO-in. End Slab. Strains at 12:00 Position ......... 100
5.54 Comparison of Measured and Calculated Strains at an Internal Pressure of 1000-psi for a 10.0-in. End Slab. Strains at 6:00 Position ........ 101
5.55 Comparison of Measured and Calculated Strains at an Internal Pressure of 1000-psi for a 10.0-in. End Slab. Strains at 3:00 and 9:00 Positions ....... 102
5.56 Comparison of Measured and Calculated Strains at an Internal Pressure of 1000-psi for a l2.5-in. End Slab. Strains at 12:00 Position ....... 103
5.57 Comparison of Measured and Calculated Strains at an Internal Pressure of 1000-psi for a 12.5-in. End Slab. Strains at 6:00 Position ................. 104
5.58 Comparison of Measured and Calculated Strains at an Internal Pressure of lOOO-psi for a l2.5-in. End Slab. Strains at 3:00 and 9iOO Positions .............................. , ........ , ... 105
5.59 Comparison of Measured and Calculated Strains at an Internal Pressure of 1000-psi for a l2.5-in. End Slab. Strains at Penetrations 1 and 3 ............................................. 106
5.60 Comparison of Measured and Calculated Strains at an Internal Pressure of 1000-psi for a l2.5-in. End Slab. Strains at Penetrations 4 and 5 .............. 107
5.61 Comparison of Measured and Calculated Strains at an Internal Pressure of lOOO-psi for a l2.5-in. End Slab. Strains at Penetration 2 ..................... 108
Figure No.
A.1
A.2
A.3
A.4
A.S
A.6
A.7
A.8
A.9
A.10
A.11
A.12
A.13
A.14
A.15
A.16
A.17
A.18
A.19
LIST OF FIGURES (cont'd)
Stress-Strain Curve for Stressteel Rods a III II It •••• II ." II II) II) II II III • 0 • It • 0 III III lit 11 0 81 0 • II • II •• 0 0: II l1li • II II 0 GIl • III ...
Locations of Bands of 0.08-in. Wire Used .... e •••••••••
Schematic View of Circumferential Prestressing Rig ................................... .
I sometri c Vi ew of Pres tress i ng Ri g ................... .
Apparatus Used for Longitudinal Prestressing ......... .
Typical Liner Details (not to scale) ................. .
Steel Closure Plates ................................. .
Steel Base Plate ..... 0.0 ••• 0 ••• 0 ••• 00 •••••••••••••••••
Protective Steel Channels Across Top of Vessel ...... o •••••• 0 •••• 00 ••••••••••••••••••••••••••
Dial Gages Measuring End-Slab Deflections ........... 00
Closed-Circuit TV Cameras to Read Dial Gages 0 ••••••••••••••••• 0 0 •••••••••••• 00 •••••••••••••
Strain Rosettes ......... 0 •••••••••••••••••••••••••••••
Measured Pressure-Deflection Curves, Test PV26. 1 .................... 0 •• 0 ••••• 0 ••••• 0 •••••
Measured Pressure-Deflection Curves End Slab, Test PV26.2. 0 •••••••••••••••••• 0 ••••••••••
Measured Pressure-Deflection Curves Si de Wall, Test PV26. 2 ... 0 •••••••• 0 ••••••• 00 ••••••••
Measured Pressure-Deflection Curves End Sl ab, Test PV27. 0 •••••••••••••••••••••••••••••••
Measured Pressure-Deflection Curves Side Wall, Test PV27 ............................... .
Measured Pressure-Deflection Curves End Slab, Test PV29 ..... o •• 0 ••••• 0 ••••••••••••••••••
Measured Pressure-Deflection Curves Si de trJa 11, Test PV29 ..... 0 •••••••••••• 0 0 ••• 0 •• 0 •••••
Page
124
125
126
127
128
129
130
131
132
132
133
133
134
135
136
137
138
139
140
LIST OF FIGURES (cont'd)
Figure No. Page
A.20 Measured Pressure-Deflection Curves, Test PV28. 1 .................... -. . . . . . . . . . . . . . . . . . . . . . .. 141
A.21 Measured Pressure-Deflection Curves End Slab, Test PV28.2 .. 0 ••• 0 ••••••••••••••••••••••••••• 142
A.22 Measured Pressure-Deflection Curves Side Wall, Test PV28.2 ... · .............................. 143
A.23 Measured Pressure-Deflection Curves End Sl ab, Test PV30.................................... 144
A.24 Measured Pressure-Deflection Curves Side Wall, Test PV30................................... 145
A.25 Measured Pressure-Deflection Curves, T est P V 31 . 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 46
A.26 Measured Pressure-Deflection Curves End Slab, Test PV31.2 ......................... 0 •••••••• 147
A.27 Measured Pressure-Deflection Curves Side Wall, Test PV31.2 ................................. 148
A.28 Measured Pressure-Deflection Curves End Sl ab, Test PV32.................................... 149
A.29 Measured Pressure-Deflection Curves Side Wall, Test PV32 ........................... 0....... 150
A.30 Measured Pressure-Deflection Curves, Test PV33. 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
A.31 Measured Pressure-Deflection Curves End Slab, Test PV33.2 ...... o ••••••••••••••••••••••••••• 152
A.32 Measured Pressure-Deflection Curves Side Wall, Test PV33.2 ................................. 153
Table
4. 1
4.2
A.l
A.2
B.l
LIST OF TABLES
Page
Maximum Internal Pressure at Failure .................... 33
Nominal Shear Stresses in End Slab ...................... 34
Concrete Properties ..................................... 122
Longitudinal and Circumferential Prestress .............. 123
Chrono 1 ogy .............................................. 158
1. INTRODUCTION
1.1 Object and Scope
An investigation of the strength of prestressed concrete nuclear
reactor vessels is currently in progress at the Structural Research
Laboratory of the Civil Engineering Department, University of Illinois.
The project is part of the Prestressed Concrete Reactor Vessel Program
of the Oak Ridge National Laboratory sponsored by the Energy Research
and Development Administration. This report records the progress in
various phases of the project.
The objective of the current studies is to determine the effect of
penetrations in the end slab on the strength and behavior of flat end
slabs of prestressed concrete nuclear reactor vessels. A previous study
by Karlsson (1971) indicated that the shear strength of the end slab was
rather insensitive to size and arrangement of penetrations. Karlsson
rationalized this observation by demonstrating that, for the particular
failure mechanisms observed in the vessels tested (Fig. 1.1), the stress
conditions in the concrete were not critical in the immediate vicinity
of the penetrations. The same hypothesis would suggest that the end-slab
shear strength would become sensitive to the presence of penetrations if
the stress combinations acting on the concrete near the penetrations
exceeded the strength limit. The current study was initiated in order to
investigate this limit experimentally and analytically so that a 1
reasonably general design (or safety analysis) procedure could be
developed.
2
Analytical and experimental activities on the project are complementary
and are directed toward the development of an analytical model to simulate
the response of the end slab. The model is intended to reproduce the'
entire range of response, automatically differentiating between different
failure mechanisms, under increasing internal pressure. To develop and
substantiate such a model, the experimental work is designed specifically
to investigate (a) the strains leading to internal inclined cracking,
(b) stresses in the vicinity of the end-slab penetrations, and (c) stress
redistributions in the fully cracked end slab. The series of end-slab
tests have been planned with emphasis on span·-depth ratios and penetration
arrangements which reflect the geometry of slabs covering the main cavity
for the current generation of vessel designs.
The experimental program includes three series. The results of
the first two series, comprising eight test vessels, are reported
here. The outline of the experimental program is given in Chapter 2
along with a brief description of the test vessels and procedures.
Detailed information on fabrication and testing is provided in Appendices
A and B. Chapter 3 is devoted primarily to comparisons of the results
from analyses using axisymmetric models with those from three-dimensional
analyses in the linear range. Chapters 4 and 5 describe salient features
of the test results.
1.2 Acknowledgments
The reported work was carried out at the Structural Research Laboratory
of the Civil Engineering Department, University of Illinois, Urbana, as
part of the Prestressed Concrete Reactor Vessel Program of the Oak Ridge
3
National Laboratory sponsored by the Energy Research and Development
Administration. The program is coordinated by Dr. J. P. Callahan of the
Oak Ridge National Laboratory.
Several graduate students in Civil Engineering who were part-time
research assistants contributed to the project. The experimental work was
initiated by B. Oland, K. Clapp, and A. C. Stepneski and continued by
J. Reins. Data reduction and analysis were performed by J. L. Quiros.
H. O. Abdulrahman developed ISA, the axisymmetric-analysis program used
in studies of the test results. E. Chen aided in the analysis of the
end slabs.
Special acknowledgment is due Professor L. Lopez for making FINITE,
a general-purpose computer program for structural mechanics problems~
available to the project and giving unstintingly of his own time.
R. D. Metz, Laboratory Technician, was instrumental in developing
the prestressing equipment and the liner for the vessel. Without his
dedication and resourcefulness, the work would have been much delayed.
The success of the instrumentation is due to the careful work of
G. H. Lafenhagen, J. H. Sterner, and D. C. Hines. The experimental
work was supported by the staff of the Civil Engineering Machine and
Electrical Shops under the direction of Professor V. J. McDonald and
Mr. O. H. Ray.
All computational work was carried out on the DEC-10 and IBM 360/75
systems of the Digital Computer Laboratory of the University of Illinois,
Urbana.
The project is directed by W. C. Schnobrich and M. A. Sozen.
4
2. OUTLINE Of THE INVESTIGATION
The test specimens were small-scale cylindrical pressure vessels
designed to investigate the influence of penetrations in the end slab
on strength and behavior of the vessel. The overall dimensions of the
eight test ve~sels described in this report were 40 by 40 in. round
(1.02 by 1.02m) as shown in Fig. 2.1. The cavity, closed at one end by
the test slab and at the other by a 4 in. (O.lOm) steel plate, had a
diameter of 25 in. (0.64m).
Circumferential prestressing was provided by five bands, each
containing approximately 290 wraps of 0.08-in. (2mm) high strength
wire at an effective prestress at time of test of approximately 120 ksi
(830 MPa). Sixty Stressteel rods, uniformly distributed around the
perimeter in two rows, were used to develop a total effective longi
tudinal prestress force of approximately 2700 kips (12 x 106 N).
The target concrete compressive strength was 5500 psi (37 MPa).
No reinforcing bars were used in the end slab. The bars ·in the skirt
were placed to maintain integrity of the specimen during circumferential
prestressing.
A composite liner made up of thin sheets of steel, copper, and
neoprene was used for all specimens.
The two main variables in the experimental program were (1) the
thickness of the end slab which was either 10 or 12.5 in (0.25 or 0.32m)
and (2) the size and arrangement of the penetrations (Fig. 2.1). The
eight test vessels were distributed as indicated below with respect to
the two variables.
5
Solid slab Six penetrations, 5-in. Thirty-seven penetrations, 2 in.
Clear Span/Nominal Depth of Slab
2.5 2.0
PV26 PV27 PV29
PV28 PV32,PV33 PV30,PV31
The five-in. (0.13m) penetrations (PV27, PV32, PV33) were arranged
uniformly on an 8-in. (0.20m) radius'. Locations of the two-in. (0.05m)
penetrations (PV29, PV30, PV31), spaced at 3 in. (0.075m) center-to-center
to each other, are shown in Fig. 2.1. No reinforcing sleeves were used
in the penetrations which were closed by steel plates on the pressurized
surface of the slab.
The test vessels were first prestressed circumferentially. After
installation of the liner, a 4-in. (O.lOm) steel plate was placed at the
open end and the longitudinal prestressing was applied. The vessel was
then placed in the testing room and pressurized internally, using oil
after the cavity was filled with water, to failure over a period of
approximately three hours. In addition to the pressure, measurements
included deflections of the end slab and the side wall, and strains in
the concrete and the longitudinal rods, with the majority of the strain
gages concentrated on the walls of the penetrations if any.
The analyses in the first phase of the project were aimed primarily
at the interpretation of the experimental data which required a large
number of solutions with the critical factors varied parametrically.
This required the development of an interactive computer program
(Abdulrahman, 1976) which is suitable for linear axisymmetric and
plane-stress or plane-strain problems. Three-dimensional effects in
the solid and cracked end slab were studied using FINITE, a three
dimensional analysis program developed by Lopez (1975).
6
3. COMPUTER ANALYSIS OF END SLABS
3. 1 Introducti on
In order to analytically determine the stress and strain distribution
in the heads of the test vessels~ linear three dimensional finite element
analyses were performed on a segment of the full head of the vessel.
Three dimensional analyses are extremely expensive to obtain. The number
of equations and the bandwidth of those equations are of necessity very
large. The result of these factors is to destroy the solution efficiency
of most equation solvers. Therefore, because of the expense normally
involved in obtaining each 3D solution, one of the major objectives of
the 3D solutions was to provide a basis for confirmation of an approxi
mate axisymmetric solution which could be quickly and inexpensively run.
This approximate axisymmetric procedure can then be run to investigate
the influence of various parameter changes without the costs and time
delays associated with 3D procedures.
3.2 Linear Three Dimensional Analysis
Because of the high relative computational efficiency of the iso
parametric element over other three dimensional elements, tetrahedrons
etc., the twenty node isoparametric element programmed in the FINITE
system (Lopez, 1975) was selected for use in this investigation. This
element has several advantages over lower order elements. The twenty
node element, having three node points along each edge of each element,
has the capacity to properly model with a curved surface the region
7
around the circular holes that penetrate the slab. The element more
readily accommodates sharp strain gradients than would lower order
elements. The price paid for these is an increased bandwidth.
The element layout used in the analysis is shown in Fig. 3.1.
Advantage was taken of the symmetry of the problem. Even taking
full advantage of symmetry, eight hundred plus node points were
needed. Each node point has three translational displacement degrees
of freedom. This involves then twenty four hundred plus equations.
When considered with the wide bandwidth common to three dimensional
analysis (in this case 160 nodes or 480 degrees of freedom) this repre
sents a sizeable computational effort.
The FINITE program was selected over other general purposes (i.e.,
SAP and NASTRAN) and even special purpose 3D programs because of FINITE's
data generator capability and the options available for output. A
major disadvantage of most general purpose programs when applied to three
dimensional problems is the amount of pre and post processing necessary
to get the results into a reference frame convenient to the user rather
than that convenient to the programmer. Output from most programs results in
a simple global system. Although three dimensional analysis was necessary,
the flow of stress in the end slab is still basically polar rather than
cartesian.
One further reason behind using FINITE is its ability to reuse the
stiffness matrix of an element. The twenty node element does require a
significant computational effort to form the stiffness matrix. It is
more economical to store and retrieve the matrix each time it is
needed than to recompute it each time. FINITE has a coordinate trans
formation so it is possible to generate one stiffness matrix then for
8
example use it for all twelve elements near the center of the slab.
To include elements to model the side wall (skirt) as well as over
the head was beyond the current capacity of FINITE (833 nodes). It was
therefore necessary to replace the side wall by a system of springs and a
set of applied nodal loads to represent the interactio~ of the wall with
the head as load is applied.
Two spring systems were employed. A set of longitudinal springs
provide the axial stiffness and moment stiffness equivalent of the skirt.
To model the shear transfer or hoop effect, radial springs were also
connected to each node located on the outside bottom face of the slab.
The value of the spring stiffnesses were selected to match the stiffness
coefficients for a circular cylindrical shell with a shear and bending
moment applied to one end.
To model the loading, the solution for a cylindrical shell fixed to
the base slab was used. For the dimensions of the test specimen the
val ues are:
~>1 = 35.185 .6P o
in.-lbs/in.
8.388 .6P lbs/in.
This effect must be superimposed on top of the prestress. One thousand
psi was selected as the pressure for which the comparisons would be made
between 3D, 20 and experimental results. At that pressure there is a
near balance between internal pressure and external prestress effects
on the cylinder. Therefore, the net interactive forces are negligible
9
for this reference pressure. Displacement and stress plots for the 3D
solutions are shown with zero unbalanced interactive forces.
Grid Layout. Because of the mentioned limit of slightly more than eight
hundred nodes presently existing in the program, it was possible to use
only three rays of elements over that section of the head (300 segment)
taken out for analysis (Fig. 3.1 ).The stresses near the center of
the segment were expected to be reasonably constant or uniform with
respect to changes in the radial coordinate. Therefore the grid was
not continued all the way in to the exact center of the head. Allowing
this small artificial hole to exist at the center eliminated the need
to coalesce several nodes into one in the process of reducing the 20 node
element to one with a triangular plan form.
In the vicinity of the real holes or penetrations in the slab, the
grid layout selected was a compromise between the grid giving the best
stress resolution around the hole and that gridded in a regular pattern
thereby minimizing data input, within the constraints of the size
limitations set down by the program. The resulting pattern (Fig. 3.1 )
contains only one regular ray of elements for the cross section at the
radius of the holes. Any other grid arrangement would have meant either
a totally irregular grid or highly distorted element geometries in some
locations. Neither of these alternatives was acceptable. Also it was
felt essential that the elements framing the hole should not degrade to
triangular elements at the side of the hole. Thus the layout shown was
selected.
Four layers of elements are used through the thickness of the slab.
This provides ample points of stress definition to accommodate all but
very sharp stress gradients through the thickness. The finite program
10
allows stress output at either the points of integration or at the node
points. In order to have the values printed for nodes in the hole area
so that a comparison could be made with experimentally obtained strains,
nodal values were optioned. This was done recognizing that corner node
points represent the locations of poorest stress definition for an
element procedure. Simple arithmetric averaging was used (i.e., no
weighting on basis of tributary areas, etc.) in determining the nodal
stresses from the element values.
Boundary Conditions. Within the context of a displacement or stiffness
form of finite element analysis conditions can only be specified on the
displacements. The segment of the slab that was analyzed is shown in
Fig. 3.2 on which the boundary conditions are also indicated. Symmetry
lines of 8=0 and 8=300 require that the displacements normal to the
symmet~ surface be specified as zero. However, within the element method
no restriction can be placed upon the shear stress on those faces.
The inner surface, radius equal one inch has zero radial and circum
ferential displacements but allows normal (or in other words vertical)
displacements. The previous section discussed the interaction problem
at the junction of the slab with the cylindrical wall .. As noted
vertical and radial springs replace the interactive stiffness with the
cylinder.
The magnitude of the interactive forces was approximated based on
thin shell solutions of a fixed ended cylinder. The interactive bending
moment was split into statically equivalent vertical nodal forces. The
shear force was applied as radial nodal loads for those nodes along
the contact surface between the cylinder and the slab.
11
The pressure to the steel closure covering the five inch openings
was applied to the finite element model as a line load around the
circumference of the opening. Such a loading is a realistic model of
the actual force transfer. However, no account was taken of the
stiffening effect of the steel insert used as a seat for the closure
plug. The pressure loads were converted to consistent loads in all
cases.
3.3 Two Dimensional Axisymmetric Model
Two dimensional solutions for the test vessel were obtained using
the ISA program. The test specimen is modeled by an axisymmetric
element system as shown in Fig. 3.3 A four node quadrilateral
shaped torus element is the basic ingredient. Because of the inherent
excess stiffness of this element to strain gradients over the depth or
in other words flexural type actions, nonconforming modes are used.
These soften the element thereby improving the quality of the results.
Because only rectangular element configurations are used, questions
raised concerning the convergence of elements utilizing nonconforming
modes are not of consequence.
Modeling the vessel with penetrations as an axisymmetric structure
is accomplished by assuming that reducing the Youn~sModulus in the
appropriate directions has the same effect as the softening introduced
by the presence of the holes. An orthotropic material property matrix
is used therefore in the computation of the element stiffness matrix.
With a symmetric array of holes as with the five inch diameter penetra
tions, the hoop direction was assumed ineffective so that modulus was
taken as zero. The values of the radial and vertical moduli used in
12
the matrix were selected based on the ratio of the perimeter of
material remaining in the real vessel at the radius of the penetrations
versus the perimeter of a solid vessel. For the two inch holes this
means the moduli are reduced to
(2rr8 - 30)/2rr8 = 0.4
of the original value. The reduced orthotropic values apply to elements
61 to 75 of Fig. 3.3a while the rest of the elements (1 to 60) retain
the original isotropic values. For the vessels with the two inch pene
trations, the grid of holes is such that the properties are rectangular
rather than polar. Therefore an isotropic material matrix with modulii
reduced based on actual plan area to gross plan area was used. The
Poisson ratio was set to zero however to account for the lack of inter
action between the radial and hoop directions. From this reasoning
elements 1 to 20 used in the ISA solution had the modulus reduced to 3/8
of the modulus for the normal concrete. (Fig. 3.3b)
For the ISA program it was only necessary to specify symmetry
boundary conditions on the vessel center line and fixity at the base of
the skirt wall. With the use of nonconforming modes, however, the
boundary conditions are satisfied only at the node points, not between.
3.4 Results
The ten inch and the twelve and one half inch thick specimens
(PV27 and PV32) were both investigated by a linear three dimensional
solution with the FINITE program. The internal pressure used for each
analysis was the failure pressure for the appropriate vessel (2400 psi
for PV27 and 3075 psi for PV32). Solutions for internal pressure,
13
longitudinal prestress and hoop prestress were obtained by multiple
back passes of the various load vectors over the triangularized
version of the stiffness matrix. Because the analysis is linear~ .
these pressure values along with the elastic moduli can be scaled to
any value desired for purposes of making comparisons. The strains at
the levels of the gages placed in the penetrations and those at the
lower level of the vessel provide a test comparison with values from
the experimental and the orthotropic axisymmetric solutions from ISA.
Such comparisons are made in a subsequent chapter of this report.
Figure 3.4 compares the vertical deflections as measured during
the experiment and those calculated by the two dimensional ISA program
and by the three dimensional FINITE programs. The comparisons are quite
good. The 3D system appears to be overly stiff near the center line.
This is in part the result of shifting out one inch from the center of
the slab in order to avoid the use of elements with a triangular
shaped plan form.
Figures 3.5 to 3.6 show the stresses as computed from the
axisymmetric program and those from the 3D program. The stresses were
calculated for an internal pressure of 1000 psi and no prestress. The
load condition was selected to represent the test condition of the
already prestressed vessel. That is the vessel being subjected to an
internal pressure of 1000 psi, the prestressed vessel being the zero
or base of reference.
The radial stresses for the two solutions follow the same patterns.
Near the center line both sets of stresses show strong flexure with some
axial tension. The flexure diminishes as the hole area is approached
14
while the axial tension force builds. Then as the wall is approached
the reverse sign moment develops. The sharp increase near the reentrant
corner shows in both solutions.
Shear stresses in both solutions are low near the center region.
Then as the hole area is traversed there is a rapid increase. This is
the result of the vertical load being funneled through the rib area. On
the back side of the hole the shear stresses computed by both the 3D and
the axisymmetric solutions show a distribution departing from the para
bolic shape normally associated with flexure to a shape biased to the
bottom or insid~ surface. The latter effect must also be a manifestation
of the reentrant corner. Elements are trying to adapt to the stress
gradient partly by shear.
The hoop stresses also model well between the 3D and the 20
solutions. However, once in the hole area, the 20 solution reduces to
a zero hoop force because of the assumed zero hoop modulus. The 3D
solution on the other hand does show a flow of stress around the hole
with the consequent low but not insigificant values.
For further conformation the stresses at select locations are
compared directly in Fig. 3.7.
The comparisons between the 3D and the axisymmetric solutions
support the assumption that the actual vessel can be adequately analyzed
by a properly materialed axisymmetric structure.
3.5 Stress State in Cryptodome Head
Using the reconstructed shapes of the cryptodome for specimen PV27,
a three dimensional stress analysis was run using the FINITE program.
For each vessel the loading applied was the failure pressure, however,
15
elastic material properties were assumed. One modification was made for
PV27 where a reduced modulus was used for some of the highly stressed
elements in the vicinity of the hole.
The shape of the cryptodome and the boundary conditions used are
shown in Fig. 3.8. It is assumed that horizontal ~racking has progressed
through the skirt wall-end slab connection. It is therefore admissible
to model only the structure shown in the figure. The vertical movement
(rigid body motion) was constrained out of the system by a roller support
placed around the outer edge. Vertical forces simulate the longitudinal
prestress. The presence of the circular steel seating plate at the base
of each penetration was neglected in the solution as was the rigidity of
the anchor ring for the longitudinal prestress. Although the FINITE
program contains the ability to provide for relative constraints, that is
interrelate the displacements at a number of node points, that option was
not used because of its severe effect on the bandwidth and consequent
sizeable increase in equation solving time.
The loading used in the 3-D analysis is from a consistent formula
tion that is a work equivalent rather than a static equivalent. The
pressure applied to the plate covering the penetrations is applied as
a uniform line load around the perimeter of the holes.
The general pattern of the stress picture calculated by the three
dimensional analysis showed no great surprises (Fig. 3.9). Near the
center of the head the stresses convert from basically flexural for the
solid head to a strong membrane field after the dome has formed. The
calculated stress resultants for a section 1.5 in. from the center of
PV27 for example translate into radial axial compression force of
16
23 kips/in. while the corresponding moment is 3 in.-kips/in. In the
hoop direction the axial force is 14 kips/in. compression and a
moment of 2 in.-kips/in. At a section through the hole area the raqial
normal force has increased to 35.5 kips/in. which sums to a total force
of 60 kips. The corresponding forces for PV32 are 27 kips/in. axial,
3.8 in.-kips/in. moment in the radial direction. The vertical shear
stress on the hole section is shown in Fig. 3.9c. The shear is seen to
remain relatively constant over the depth. The summation of the shear
over the cross section sums to 40 kips for PV27. This total shear is in
agreement with the total vertical load. The non-vanishing of the shear
on the top surface, which might at first thought appear to be in error,
is the result of the free surface being inclined to the plane of reference
about which the stresses are being defined. The boundary condition of
stress normal to the free surface is adequately satisfied, adequate meaning
within the tolerances that are usually found with finite element solu
tions. For example, along the top surface at a section through the 8-in.
perimeter line, the stress normal to the surface is fn the sixty to
eighty psi range versus principal compressive forces parallel to the
faces of magnitudes seven to ten thousand psi.
Principal stresses calculated for PV27 on a section through the
vessel at the eight inches radius are listed on Fig. 3.10. A sizeable
percentage of this cross section is at a high (10,000 psi) stress level.
17
4. OVERALL RESPONSE AND FAILURE t,1ECHANISMS
The measured relationships between deflection- at center of slab
and internal pressure for all eight vessels are shown in Fig. 4.1 and
4.2. The entire set of recorded ptessure-deflection relationships are
included in Appendix A~ The deflections reported are the total deflec
tions with respect to the initial position of the slab and include the
small but finite extension of the side wall. In comparing the pressure
deflection curves, it must be considered that these refer to second
loadings for PV26, 28, 31, and 33. As recorded in Appendix B, the liner
in some of the vessels leaked during initial pressurization at various
levels (PV26, 800 psi; PV28, 3200 psi; PV31, 1200 psi; PV33, 1700 psi).
These vessels were relined and reloaded. The curves in Fig. 4.1 and 4.2
refer to the tests in which structural failure was achieved.
All curves show two typical ranges of response: an initial linear
and a following nonlinear range. As would be expected, the initial
linear range extends to higher pressures for the vessels with the
thicker slabs. Although a certain amount of IIductilityll may be read
into the curves, this is a moot question considering that the maximum
deflection at failure was less than 0.4 percent of the clear span or
one percent of the slab thickness for all test vessels. It should
also be pointed out that, had the only available sensor been the deflec
tion at midspan of the slab, structural distress would have been antici
pated simply from comparing the change in measured deflection over a
short time with that based on a linear analysis of the structure and
that warning would have been registered well in advance of failure in
18
terms of pressure although possibly not in terms of time.
Slab deflection distributions calculated using the axisymmetric
model, with the penetrations simulated by softened portions as described
in Chapter 3, are compared with the measured values in Fig. 4.3 through
4.5. Slab deflections are plotted relative to the center of the side
wall and as rates (in./psi) which refer to the initial slope of the
pressure-deflection record. Two data points are shown for the two tests
of PV28. Considering the possible scatter because of the rather small
magnitude of the data, the comparisons indicate that the axisymmetric
model is satisfactory for determining slab deflections in the linear
range for slabs with penetrations. As described in Appendix A deflection
dials had direct contact with the end slab only for-vessel PV26. Because
all dials measuring slab deflections were destroyed at the end of the
test of PV26, dials in following tests were located at some distance
from the surface of the slab with the connection provided by pretensioned
piano wire.
Shear failure of the end slab limited the structural strength of
all eight vessels tested. However, the mechanism of the .failure differed
depending on the thickness of the end slab and the presence of penetrations.
Despite the use of a hydraulic system for internal pressurization,
failure occurred explosively in every case. Chunks of slab concrete
were thrown about the testing room. This is attributable primarily
to trapped gas in the pressurizing fluid.
Cross sections and photographs of the failed end slabs are shown
in Fig. 4.6 through 4.13. The pressures registered at failure are
listed in Table 4.1 along with data on the geometry of the vessel, index
19
values for the circumferential and longitudinal prestresses, and
the compressive strength of the concrete in the end slab. (The index
value for the circumferential prestress is determined as the internal
pressure required to balance the force, at time of test, in prestressing
bands 1 and 2 located on and next to the €nd slab.)
Conditions of the end slab after testing indicated that final
collapse occurred always through failure of the concrete in a complex
state of stress but that the failures could be classified in two types
depending on the location of final distress in the concrete.
One type of failure is illustrated in Fig. 4.6 and 4.9 which refer
to solid slabs. As in the tests reported by Karlsson (1971), a three
dimensional inclined crack carves out from the end slab the kernel shown
in the lower photograph leaving behind a "cryptodome" to resist the
internal pressure. Collapse occurs as a result of failure of the con
crete in the cryptodome near the center of the slab. Viewed in two
dimensions (in the vertical plane), this mode of failure has character
istics similar to the shear-compression failure observed in reinforced
concrete beams.
The "shear-proper" or "punching" failure is illustrated best by
the state of the failed end slab of vessel PV30 (Fig. 4.10) having
a l2.5-in. end slab with two-in. penetrations. After initiation of
inclined cracking within the slab, concrete in the reduced area between
the openings fails allowing the central portion of the slab to extrude
along a nearly cylindrical failure surface.
The failure modes for each vessel are identified as SO (cryptodome)
and SP (shear-proper) in Table 4.1. Not every failure could be classified
20
as belonging clearly to one category or the other. The failed end slab
of vessel PV29 (Fig. 4.8), in addition to a well developed dome had a
vertical failure plane through the outermost line of penetration~.
Table 4.2 lists the nominal shear stresses, based on measured dimen
sions of the end slabs, for all test vessels. Failure pressures and maxi
mum shear stresses normalized with respect to the square root of the
compressive strength of concrete (in psi) are plotted against the slab
thickness in Fig. 4.14.
Both modes of failure involve complex internal stress conditions.
As hypothesi2ed by Karlsson (1971), maximum pressure for failure in
the cryptodome mode is a function of the overall geometry and concrete
strength with the influence of the prestressing forces affecting the
pressure through their influence on the initiation and trajectory of
the three-dimensional inclined crack. The shear-proper failure occurs
on a plane through the penetrations and close to the boundary, both of
which conditions complicate the stress distribution. Strain measurements
discussed in the next chapter imply that failure occurs after appreciable
redistribution of internal stresses. To generalize the results of the
tests must await the conclusion of the numerical studies some of which
are briefly discussed in Chapter 3. However, certain trends discernible
in Fig. 4.14 are of interest.
Consider Fig. 4.14a which shows the raw test results. As in the
series of tests by Karlsson (1971), the difference between the measured
strengths of the 10-in. end slabs with and without openings was quite
small. The perforated slabs had 92% of the strength of the solid slab.
The difference, if considered significant, is plausible. As long as
the final collapse is by material failure in the cryptodome near the
21
center of the slab, the penetrations, which are removed from this zone,
should have little negative influence on the strength of the slab.
For the l2.5-in. slabs, there was a meaningful difference in
maximum pressures measured for the solid and the perforated slabs,
consistently with the observation that the perforated slabs failed in
the section through the openings.
Figure 4.14b shows the maximum nominal shear stress for the
perforated slabs as a function of the square root of the compressive
strength of the concrete. The first cautious conclusion which may be
drawn from this plot is that the data for the slabs with 5- and 2-in.
penetrations must be considered separately. Assuming then that the
parameter used may be applied simultaneously to slabs with 10- and
l2.5-in. thicknesses, it may be concluded that, for the test specimens,
the nominal unit shear strength is approximately 391fT and 331fT c c
for the slabs with 2= and 5-in. penetrations, respectively. Taking
the results of the l2.5-in. slab as hard evidence, the insensitivity of
the end-slab strengths to size of penetrations in the results of the
10-in. slabs reported here and by Karlsson (1971) become plausible.
Evidently, the applicability of a statement of the shear strength
of the end slab in terms of a single nominal stress must be limited to
end slabs having geometries and loading conditions similar to those for
the test specimens. Studies toward the generalization of these observa-
tions are in progress.
22
5. STRAINS ~1EASURED ON WALLS OF PENETRATIONS
5.1 Introductory Remarks
Strains were measured at three levels on the walls of several
penetrations of five test vessels with perforated slabs. Because these
measurements provide information on changes in the load-resisting
mechanisms of the end slabs, all pressure-strain plots for two vessels
(PV30 with 2-in. and PV32 with 5-in. penetrations) are presented in
this chapter along with mean values of strains measured in the linear
range of response in all tests of vessels with perforated slabs. The
mean values are compared with those calculated using linear axisymmetric
models of the test vessels.
The locations of the 450 strain rosettes (3/4-in. gage length,
9/64-in. gage width) on penetration walls of vessels PV30 and PV32
are shown in Fig. 5.1 and 5.31. Walls of five penetrations in PV30
and two in PV32 were instrumented at levels of 2.25, 6.25, and 10.25-in.
from the pressurized surface of the slab. The locations of gages around
the circumference of the penetration walls of vessels PV32 (Fig. 5.31)
are designated by reference to the face of a clock viewed from the center
of the slab. In vessel PV30, rosettes were used at two diametrically
opposite locations (Fig. 5.1) around the circumference of five
penetrations. In vessel PV32, rosettes were attached at four circum-
ferential locations (Fig. 5.32) on the walls of two penetrations.
Designations of all gages are identified in Fig. 5.1 and 5.31.
Figures 5.2 through 5.30 contain the pressure-strain plots for PV30 and
23
Fig. 5.32 through 5.52 contain the plots for PV32. In reading and
comparing these plots it must be recognized that (1) horizontal
(strain) scales are not aJl the same, (2) minus sign in the horizontal
axis indicates tensile strain, (3) where the stress in the direction of
the gage is low, observed strain may be jnfluenced- appreciably by
stresses in orthogonal directions, and (4) gages at 45-deg. to the
horizontal read compressive or tensile strains depending on their
orientations.
Strain data were reduced and plotted using the DEC-10 and IBM 360/75
computer systems. Each measured concrete strain increment was corrected
by the corresponding mean increment indicated by a set of check gages 9
gages mounted on blocks of concrete located near the test specimen. The
maximum total strain correction was less than 1 x 10- 5, or less than the
limit of reliability of the strain measuring system.
Data from gages designated 4, 10, 17, 18, 34, and 70 in vessel
PV30 are not reported because of malfunctioning of those gages.
5.2 Strains Measured in End Slab of Vessel PV30
All strains recorded by strain rosettes installed at three levels
within five penetrations in the end slab of vessel PV30 are shown plotted
against the internal pressure in Fig. 5.2 through 5.30. Readings of all
strain gages indicate that the end slab responded reasonably linearly to
an internal pressure of 1000 psi. The first definite indication of
nonlinear response was given by the horizontal gages at level 1 in the
central penetration (Fig. 5.15). The calculated radial compressive
prestrain at this level is approximately 1 x 10-4. Thus, a tensile
strain increment of 4 x 10- 4 is an indication of the onset of local
cracking. Therefore, the readings of gages 20 and 29 are assumed to
24
indicate initiation of flexural cracks at the outside face of the end
slab at a pressure of 1000 psi. Larger strains indicated by the gages
suggest that the gages continued to function even after the develop
ment of the cracks, measuring not strain but average total deformation
over their gage lengths of 0.75 in. Oriented radially, the cracks
indicated by these gages must be the flexural cracks initiating on
the outer surface of the slab. The penetration of these cracks into
the slab appears to be limited. Horizontal gages 23 and 32 (Fig. 5.15)
continue to indicate a reasonably steady increase in tension up to over
a pressure of 2000 psi.
The next critical event is signalled by the diagonal gages at
level 2 of penetrations 1 and 3 (Fig. 5.8 and 5.9). Both the compressive
(Fig. 5.8) and tensile-strain (Fig. 5.9) readings indicate definite
nonlinear response at an internal pressure of approximately 1500 psi.
The tensile-strain readings are more significant in that, compared with
a calculated compressive prestrain of 2 x 10-4 in the direction and at
the location of the gages, a tensile strain incre~ent of 5 x 10-4 is
a strong indication of cracking. The readings of gages 42 and 15 are
complementa~, with the inference that gage 42 was straddling a crack
while gage 15 was not. The increase in strain rate of the gages reading
compressive strain (Fig. 5.8) also suggests a rearrangement of the paths
through which pressure is transmitted from the slab to the vessel wall.
The compressive-strain increment at which nonlinear response was
initiated (approximately 5 x 10-4) is too small to justify ascribing the
observed phenomenon to inherent inelastic action of the material (6300-psi
concrete). An increase in the ratio of compressive stress to internal
25
pressure must also have contributed to the increase in the observed
strain rate.
It would appear from the data in Fig. 5.8 and 5.9 that internal
inclined cracking was initiated in the end slab thin 3.5 in. of the
inside face of the side wall at a pressu~e of appr6ximately 1500 psi,
less than half the maximum pressure of 3210 psi.
The magnitude of the measured compressive strains at level 2
in penetrations 1 and 3 is also of interest. The final readings
indicated a compressive-strain increment of virtually 0.005 (Fig. 5.8).
The diagonal compressive-strains recorded at level 3 of the same
penetrations (Fig. 5.12) were of comparable magnitude though not quite
as large. Furthermore, strains measured in similar heights and
directions in penetrations 4 and 5, which are located closer to the
center of the slab span, were perceptibly smaller (Fig. 5.25 and 5.29).
(As would be expected, corresponding measurements in the walls of the
central penetration 2 are very small. See Fig. 5.16 and 5.17). The
measured diagonal compressive strains, then, did indicate the location
of the failure. They may also be used as evidence to provide an insight
into the mechanism of failure which is suggested to be, in one radial
plane, first a development of compressive struts followed by a primary
compression failure of the struts.
The sudden change in the response of the horizontal gages at
level 3 of penetrations 1 and 3 ( g. 5.11) is likely to have been
related to the development of cracking at the reentrant corner between
the slab and the side wall. Accordingly that event must have taken
place at an internal pressure of over 2000 psi.
26
Measurements of vertical strain, which are affected most critically
by "background noise il because of their typically low magnitude, do
confirm the changes in the load-carrying mechanism inferred from other
measurements. Consider the measurements of vertical strain in the
central penetration 2. At level 3, closest to the pressurized surface,
the increase in compressive strain is, relative to vertical strains at
other levels, large and steady with a slight increase in its rate at
high pressures (Fig. 5.17). At levels 1 and 2 (Fig. 5.14) the increase
in compressive strain stopped at approximately 1800 psi, indicating that
the pressure was being transmitted to the supports through compressive
dome action. Generally similar conclusions may be drawn from readings
of vertical gages on the walls of penetrations 4 and 5.
5.3 Strains Measured in the End Slab of Vessel PV32
Strains measured on the walls of two diametrically opposite
penetrations in the end slab of vessel PV32 are shown in Fig. 5.32
through 5.52. Because the penetrations in the end slab of vessel PV32
were limited to a single radial location, the information from the
strain gages is not of as wide a scope as that from the gages in PV30.
Strain measurements indicate changes in response at internal stresses
of approximately 1250 and 2000 psi. Horizontal gages 20 and 56 at level
1 (Fig. 5.34) signal the reaching of radial cracks to the "6 oiclock il
sides of the penetrations at approximately 1250 psi. Indications of
inclined cracking are provided by the measurements of gages 15 and 51
(Fig. 5.44), located at level 2 and H3 oBclock,1i at an internal pressure
of 2000 psi. The increased compressive strain rate, at 2000 psi, shown
in Fig. 5.45, attributable to a change in the internal load-carrying
mechanism, confirms the opening of inclined cracks at an internal pressure
27
of approximately 2000 psi. It would appear from the measurement of
the rosettes at mid-height (level 2) that permanent internal damage
occurred at a pressure of 65 percent of the ul mate.
The compressive strains at levels 2 and 3~ projected to the
failure pressure of 3075 psi, are all le$s than 0.04. Given the
observation that the mat~rial failed on a circular plane passing through
the axes of the penetrations, the indicated strains appear low for
confined concrete.
It is interesting to note that most gages on the 6 and 12 olclock
locations indicated negligible shearing strains, which suggests that the
gages were functioning properly.
5.4 Comparison of Measured and Calculated Strains
Figures 5.53 through 5.61 contain a series of comparisons of
measured and calculated strains in the range of linear response. The
plots show the variation of strain per 1000 psi internal pressure over
the thickness of the slab.
Measured values represent the mean initial slope of the pressure
strain plots obtained from gages attached to the walls of the
penetrations. Each datum point represents the average of all available
data for that location. The scatter range was as large as 60% for
mean values less than 1 x 10- 4 and less than 20% for mean values equal
to or greater than 1 x 10- 4. Readings from initial and second loadings
of test vessels which had to be loaded twice because of leakage in the
first test were treated as independent values.
Strains were calculated using the linear axisymmetric models
described in Chapter 3 with modifications to represent the effect of
the penetrations. It should be noted that the calculated strains refer
28
to an ideal radial plane without strain variations in the circumferen
tial direction while the strains were measured on the surfaces of the
penetrations. A better experimental check for the results of the
modified axisymmetric models would be at the radial plane bisecting the
angle between two penetrations.
The comparisons are, in general, favorable indicating that the
axisymmetric model may be used to investigate strain conditions around
the penetrations in the linear range of response.
29
6. SUMMARY
This report summarizes the experimental and part of the analytical
work carried out in connection with an investigation of the structural
strength of prestressed concrete reactor vessels. The project is part
of the Prestressed Concrete Reactor Vessel Program of the Oak Ridge
National Laboratory sponsored by ERDA. The objective of the current
phase of the work is to develop procedures to determine the shear
strength of flat end slabs of reactor vessels with penetrations.
6. I Analytical Studies
The assimilation and generalization of the experimental observa
tions demand two different levels of analytical capability. The
necessity for a large number of parametric studies to determine
sensitivity to variations in physical properties for design as well as
data interpretation requires a computer program which is easily
accessible and which produces solutions at moderate cost. On the other
hands the necessity for simulating the three-dimensional nonlinear
discontinuous phenomena in order to study local forces and deforma
tions in the end slab requires a rather large computer program.
To satisfy the first requirement, an interactive computer program,
ISA, based on a linear axisymmetric model was developed. ISA is also
suitable for problems of plane stress or strain and has been designed
to be developed further to include options for nonlinear analysis as well
as analysis of crack development within the end slab. The current version
30
of ISA is described by Abdu1rahman (1976).
Linear three-dimensional analyses of the end slab, in intact and
cracked conditions, have been made with the use of FINITE (Lopez, 1975).
These studies have confirmed the feasibility of using modified axi
symmetric models to determine strains around the penetrations in the
linear range of response.
6.2 Experimental Studies
The dimensions and properties of the eight cylindrical flat-ended
test vessels are described in Fig. 2.1 and Table 4.1. The main experi
mental variables in the tests reported were the thickness of the end
slab and the size and arrangement of end-slab penetrations. No
reinforcing steel was placed within the end slab. The penetrations, which
had no sleeves or liners, were closed by steel plates at the pressurized
surface of the slab. Prestressing was provided by 60 Stressteel rods
longitudinally and 5 bands of 0.08-in. (2-mm) high-strength wire
circumferentially, each band containing approximate1y 290 wraps of wire.
Prestress forces are tabulated in Tables 4.1 and 4.2.
The liner for the internal cavity comprised sheets 'of steel,
copper, and Neoprene. Each vessel was pressurized hydraulically to
failure over a period of approximately three hours. Maximum pressures
ranged from 2400 psi (16.6 MPa) to 3765 psi (26.0 MPa). Despite the
use of fluid as the loading medium, the structural failures were
violent.
All end slabs failed in shear, but in different modes (Fig. 4.6
through 4.13). As listed in Table 4.1, all three vessels with 10-in.
(0.25m) slabs and the vessel with a solid 12.5-in. (O.32m) slab failed
31
after the formation of a "cryptodome", as observed by Karlsson (1971),
within the end slab. The location of concrete failure was near the center
of the slab and away from the section of maximum shear stress. For this
mode of failure, the shear strength of the end slab was insensitive to
the presence of the penetrations.
The shear failures of 12.5-in. (O.32m) end slabs having penetrations
were characterized by failure of the concrete on an almost vertical plane
through the section of maximum shear stress and were designated as "shear
properfl or IIpunchingll failures. In this case, the strength of the end
slab was affected by the penetrations. Measured nominal shear stresses
are listed in Table 4.2.
Measured strains on the walls of the penetrations indicated the
initiation of the failure process (development of inclined cracks within
the end slab) and permanent internal damage at pressures less than half
the ultimate pressure for the test vessels with 2-in. (O.05m) openings
and 12. 5- in. (0. 32m) slabs.
32
7. REFERENCES
1. Abdu1rahman, H. 0., IIISA, Interactive Stress Analysis, A Program for Plane Stress, Plane Strain and Axisymmetric Analysis in Structural Continua,1I Civil Engineering Studies, Structural Research Series No. 428, University of Illinois, Urbana, July 1976.
2. Karl sson, B. I. and M. A. Sozen, II Shear Strength of End Sl abs wi th and Without Penetrations in Prestressed Concrete Reactor Vessels," Civil Engineering Studies, Structural Research Series No. 380, July 1971.
3. Lopez, L., II FIN ITE, A POLO 11 Subsystem for Structura 1 Meehan; cs, II Civil Engineering Systems Laboratory, University of Illinois, Urbana, June 1975.
33
TABLE 4.1 "
Maximum Internal Press~re at Failure
Penetra ti ons Prestress Index Fail ure r~ark Nominal Concrete No. oi a. Long. a Circum. b Pressure Mode c
Slab Comp in. psi psi psi Thickness Strength (mm) (r1Pa) (MPa) (MPa)
in. psi (mm) (r~Pa )
PV26 10 6710 5010 1560 2610 SO (254 ) (46.3) (34.5) (lO.8) (18.0)
PV27 10 6845 6' 5 5340 1560 2400 SO (254) (47.2) (127) (36.8) (10.8) (16.6)
PV29 10 5480 37 2 5330 1710 2400 . SO (254) (37.8) (50.8) (36.8) (11.8) (16.6)
PV28 12.5 6420 5760 1700 3765 SO (317.5) (44.3) (39.7) (11.7) (26.0)
II'
PV30 12.5 6300 37 2 5340 1690 3210 SP (317.5) (43.4) (50.8) (36.8) (11.7) (22.1)
PV31 12.5 4970 37 2 5500 1680 2800 SP (317.5) (34.3) (50.8) (37.9) (11.6) (19.3)
PV32 12.5 5720 6 5 5540 1780 3075 SP (317.5) (39.4) (127) (38.2) (12.3) (21.2)
PV33 12.5 4875 6 5 5400 1780 3100 SP (317.5) (33.6) (127) (37.2) (12.3) (21.4)
aTotal effective prestress force divided by horizontal cross-sectional area of cavity.
br'1ean effect; ve prestress force in prestressing bands 1 and 2 divided by vertical tributary area of cavity.
cSD = Shear failure after complete formation of cryptodome. SP = Shear failure by punching.
TABLE 4.2
Nominal Shear Stresses in End Slab
Mark Neas. Penetrations Maiimum - ----Snear- 51-ress Max-rm'Um Slab Pressure at Slab Edge a Shear Stressb
Thickness No. x Size v v/IF vm v /Ifl in. in. psi psi c psi m c
PV26 9.81 2610 1660 20.3
PV27 10.06 6 x 5 2400 1490 18.0 2370 28.8
PV29 9.94 37 x 2 2400 1510 20.4 2820 38.1 w ...j:::::.
PV28 12.38 3765 1900 23.7
PV30 12.22 37 x 2 3210 1640 20.7 3070 38.7
PV31 12.02 37 x 2 2800 1460 20.7 2720 38.6
PV32 12.30 6 x 5 3075 1560 20.7 2490 33.0
PV33 12.45 6 x 5 3100 1560 22.3 2480 35.6
aNomina1 shear stress at the slab-wall interface.
bNominal shear stress at the net section through the penetrations.
35
~O'I
17" 8" 8'1
7.5"
Fig. l.la Vessels Tested by Karlsson
17" ~,.tf'at ien!!
(1ltV20, flPVZI)
10"
40"
2-in. penetration
I I
\ \
/;"""
/
\
" ' .........
36
-~ ~
60 "-
Fig. l.lb Vessels Tested by Karlsson
\ \ \
Sea 1 i"9 plate
4 II 3/4t 1/2" . ~211
~ 1}41O
f
30-1" Oia. Holes at 12" Aport
S"R.
25"
I
37
12.5"R.
2.5"R.
1---------- 40" --------+4
0.75" Diameter Stress'"1 Rods
1.25" Thick Ring
O.OS" Diameter Prestren Wire
7.5" 12.5"
30 -I" Oia. Holes at 12" Aport
o 0
0- 0 0 o __ 0
-0=----'/"----:::- - ........ " 0 o 0 / "-
00/ 0000 / 00000
0, 000000 25"
I
12.5"
010000000 I 0\ f'\f'\f'\f'\(\(\ ,0 0', \JO\JO\JO\JO\J~O 0
o 0 \" 0 0 0 0 /1 0 0 o , / 0 o ' ______ - ./
o
1-0>--------- 40" --------....,
Bonded Reinforcement
I"
Fig. 2.1 Penetration Patterns and Cross-section of Vessel
12.5"R.
2" Dia.
46 5 5'2.~ 15
4 4S
4 ~15~5'2.
Fig. 3.1 Finite Element Grid Used in Three Dimensional Analysis
~\\OO '2.9
'2. \9S
4
w OJ
39
u = v = 0
u = 0
<t Slab
Cf. Penetration
Simulated Skirt
Fig. 3.2 Boundary Conditions Used in Three Dimensional Analysis
y
v
, 2 'II Ii S
1 6 61 66 i11 II 12 1iI 1lI IS
2 7 62 67 72 21 :12 21 24 :l5
3 8 63 68 73 :n 12 n 1'1 'lIS
4 9 64 69 7l! _I
IiJ2 I,f! iN IjS
5 10 65 70 7S '2 51 91 56
Fi a Gri d for i th 6 - 5" ve
penetratlons
5
11 111
12 2!1
13 J!i
14 111$
15 Ell
1 S 9 o
16 31 46 11 UI 13 20
17 32 4-7 71 2a :l9 ~
18 33 148 11 'Ja 'is 'K)
19 34 4-9 on lIS lj9 so
20 35 50 D1 sa Ell eo
~~1 36 51 III GO? lOt liM
~~2 37 52 lifo es en QI
23 38 53 WI1I 10 711 7Z
24 39 54 b "N "IS 15
~~s 40 55 71 "la "l!iI 00 I
~~6 41 56 II. «I.' Btl sa I
c.~7 42 57
IllO 8Ii In /!II
(~B 43 58 IB 00 SI 5!! i:
~~9 44 59 I
'ITJ SI.I Ii6 QIi I'
30 4S 60 'Iif1 m tall 100
F v p
l.. 2 ~ 1# S
S 10 15 20 25 ttr 11 12 1'3 tat
II '9 14 19 24 l§ 20 21. &! 2'J
3 B 13 18 23 25 t?JIIJ so ?i. ~
2 7 12 ' 17 22 71 :III 9S 'fa '6!.
1 6 11 16 21 'It! an i&!I 'i5 Iii!:)
;g. 3.3b Grid for esse 1 'wi th 37 - 211 enetrations
Ii ., II I'
30 35 LiO ItS III 1,
29 34 39 2'4 126 2B I ~ "V
28 33 38 '19 '" 1'.I1!:'.i 'iii
27 32 37 1@2 10 ifIl
" "" 26 31 36
61. 5lt' 91 ~ I
4:1 51 61 9b ~ as 167 S!!I I
Li2 52 62 \is II!O 111 G2
Li3 53 63 ID 6\! 6£ C!'!J
4:4 54 64 BJ IiIB iii! 'llO
4:5 55 65 it 1\2 71 "lIS
Li6 56 66 'J5 16 '77 'l6
Li7 57 67 7!11 Sll I!\I. 1m?
-liS 58 68
M M ~ 149 59_ 69 (f) 8! •
50 60 70
es
eo
!/II. l!I2 !f1 56
..j:::::. o
41.
42
~----------~--------~o o o
i Hole
Cf. Slab I
~ ~1 5000 psi
, , I II 'I U ,I
•
Fig. 3.5a Radial Stress in lO-in. Slab with 5-in. Diameter Penetrations
-
+ Tens.
,..::::::. w
~ o :r:
..0 o
(f)
+
c o
~ __ -b ____________ __
44
----
l~
Vl s::: 0
'r-+-> ro $...
+-> (]) C (])
c... $... (])
+-> CI.)
E ro
'r-0
C Or-
I L.()
..c +-> 'r-s:
..0 rtj
.--V)
= 'r-I
0
C or-
Vi Vi (]) S-
+-> V)
0.. 0 0 :c
..0 L.()
('Y')
O"l
l.J....
~ Slab
.1 'I I, " :, II
~ d II
" II
~ I,
'i I, II ~ II II II II
~
~ ~ 5000 psi
I , II U II II II II I' II
<t Hole I
5 II Penetrati on .... 1111_---
-Compo
Fig. 3.5e Shear Stress in lO-in. Slab with 5-in. Diameter Penetrations
..f:::>, <..n
(f. Slab 2-D 3-D
I- ~ 5000 psi
(f. Hole
I II •
jO!!lilf 5 Penetration >"1 --~'~'----Wh~M"~i~--~i'rT'----'----'
I '., 16, I I I B
I I I I I
Fig. 3.6a Radial Stress in l2.5-in Slab with 5-in. Diameter Penetrations
-Compo
+ Tens.
~ (j)
..c o
(f')
+
~--~------------
47
T~ l~
U)
c 0
+..J rd S-
+..J C!J C C!J 0-
S-C!J
+..J C!J E rd 'r-0
C 'r-
I L.O
..£:: +..J 'r-3
...0 rd r-(/)
C 'r-
I LD
N
C
U)
U)
C!J S-
+..J (/)
0... 0 0
:::c
...0 1..0
(Y')
en 'r-LL.
Cf. Slab
I~ " II II II II II II II II If II II
I ! I
'~ iSl
5000 psi·
Fig. 3.6c Shear Stress in 12.5 in. Slab with 5-in. Diameter Penetrations
+ Tens.
+::Q:)
• I
P = 1000 psi
E ::. ;"000 psi
--- 2- D R(,:~1lIt5
49
-- -- 3 - D Re~'Lllls With Reduced Intcrative Moments
Cl y T'y x
Radio I Shear
12.5 \ 12.5 "
\ :--, \
, '" , ~
I I , I \ \ \ \ \ \ \ \ \
/ /
/ /
/ /
°0 00
0' (psi)
3.7a Stresses at R = 8 inches
p = 1000 psi
E = 3000 psi
--- 2 - 0 Results
---- 3 - 0 Results With Reduced Interative Moment
O'y
Radio I
u(psi)
O'z
Hoop
3.7b Stresses at R = 3.75 inches
u (psi)
7:'0
0' (psi)
I 1500
T'yx
Shear
u (psi)
Fig. 3.7 Comparison of Stresses Calculated by Axisymmetric and by Three Dimensional Analyses for l2.5-in. Thick Head
z(w)
On This Front Face u::O J v=O
Use Springs With k :: 1.0 E 06 To Specify The Horizontal Disp. Perpendicular To This Face is Zero
Fig. 3.8 Boundary Conditions Used in Cryptodome Analysis
Face
On The Bottom J Face A long This Edge w = 0
U1 o
C£. Sia b 't. Hoi e
I 1 I I \... 5" Penetration I I I I I I
10,000 psi
Fig. 3.9a Radial Stresses in Cryptodome Shaped from lO-in. Slab
c..,..,
+ Tens.
~ Slab Cf. Hoi e
! 5 81 Penetrati on
t+-- "'.
I I I I I
1-' .,
10,000 psi
Fig. 3.9b Hoop Stresses in Cryptodome Shaped from lO-in. Slab
+ Tens.
- Compo
U1 N
i Slab ~ Hole
~ 5" Penetrati on .. i I I I I I
10,000 psi
Fig. 3.9c Shear Stresses in Cryptodome Shaped from lO-in. Slab
-Compo
(".1 W
54
<t 67 -9 -253 -770 1
58 96 84 80 -6588 -6598 -9606 -11,597
-93 -778 -1537 200 24 -157
-7142 -10,184 -11,662
-27 -361 -1021 -1970 123 130 136 -158 7578 -7461 -10,425 -11,518
-229 -973 -2232 -2 -98 -446 -7388 -10,106 -11 1156
c: 0 ..... c -198 -291 -851 -1688 -2555 ""' .....
- 515 QJ 478 63 -185 -794 c: -9589 -9851 -11,031 -11,213 -10,753 QJ
a..
""' -442 -1167 -291 I QJ *= -694 - i i77 -984 QJ T -11,816 E -11,906 -9785 c
0
LO 221 51 -1058 -2354 -3331
-1862 -1773 -2687 -1442 -1190 -12,416 -12,216 -11,653 - 9014 -8421
-2417 - 2251 -3764 -7780 - 4741 -2408 - 17,035 - 10,387 -7267
12328 3i87
B 3378
-9646 -8506 -4425 1278 94 -25,336 -14,700 -8174 -1315 -2220
-5074 -4916
Fi g. 3.10 Principal Stresses on a Section Adjacent to Penetration Through 10- in. Head
3.0r-
U')
~ ~
CD "-::J 2.0 U') U')
CD "-a.. 0 f!:: "-CD 1.0 +-f!::
0
----------- t¢P
t¢PfIIIIII'"
_fIIIIII'" ...,.-
10.01 in. a ~ 1
Deflection
Fig. 4.1 Measured Relationships Between Internal Pressure and Total' Deflection at Midspan of End Slab for Test Vessels with 10-in. End Slabs
---20 0 a.. :?! --- 15 CD "-::J U') U')
CD "- (J1
a.. (J1
10
o-c: "-CD +-f!::
5
0
m ~
... Q.) .... :::ll m m Q.) .... a..
0 c: ll.... Q.) +-c:
3.01- ~ ~ ~ -1120
I-
200t / PV2j! ]'5 10
I.°r / J5
~--------~--~----~----~--~----~----L---~----~--~----~----~--~----~--~~--~O
Def lection
Fig. 4.2 Measured Relationships Between Internal Pressure and Total Deflection at Midspan of End Slab for Test Vessels with 12.5 in. End Slabs
0 a.. ~
CD '-:::ll m m Q.) .... a..
<.iI 0"1
0 c: ll...." Q.) +-c:
57
10
U) I Q 8 )( • PV 26 -S.
....... c 6
-Q) ... 0 a::
.§ Calculated, Axisymmetric Model ... 0 Q)
...... 2 Q)
0
CQ I 0
8 x • PV 28, Test I - o PV 28, Test 2 "m Q.
....... C 6 -Q) ..... 0 0::
4 c • Calculated) Axisymmetric Model .Q ...
0 0 • Q) 0 ...... 2 Q)
0 •
0 18
ct Rod ius, in.
Fig. 4.3 Deflected Shape of End Slab for Vessels with Solid Heads
58
10
CD D
0 8 • PV 27 x -U) Q.
"': 6 It::
Q) Colcu loted, Axisymmetic Model ..... c a: 4 It:: .2 +-(.) Q)
lit- 2 Q)
0
0
• PV 32 IiIO I 0 PV 33 0 8
x -m Q.
"""-: 6 It::
ItU Calculated, Axisymmetic Model +-
c a: 4 •
2
o 8 10 12 14 16 18
<t Radius, in.
Fig. 4.4 Deflected Shape of End Slab for Vessels with 5-in. Penetrations
CD I 0 )( -en c.
........ C
Q) +-c a:: c: 0 +-(.) Q)
...... Q)
0
CD I 0 )(
fI) c.
........ c:
Q) +-c a::: c: 0 +-(.) Q)
'i-Q)
0
10
6
4
2
0
8
6 -
2
o ct
59
Calculated, Axisymmetric Model
..
Caicuioted, Axisymmetric Model.
12
.. PV 29
• PV 30 o PV 31
14 16
Fig. 4.5 Deflected Shape of End Slab for Vessels with 2-in. Penetrations
18
60
1D I
N I'-
---- -->
-W
a.. ~ I
<.0
:~: .. ~ :~ ....... - .. -
-:"-<' ..", ..... --.,. .. ~ .... ~ .:- ~: .... :: .... -~""'-""'-~-_-"!'I"'"""'f
1 ~'i~{~~~-\)~;P~;:~i~);!;S~~ ~
= ~
U) I
N 00
> N
a.. ::::::. I
W
®
\..0 N > 0...
4-J III OJ l-
!'.QJ
--4-J 4-1'0
...0 ro .....
(/)
-0 s::::
W
or-I..L.
61
> 0..
®
+..J l/l Cl) I-
---------------- ~ +..J 4-res
..Cl rd
r--(/)
'"'0 C
W
'r-u...
24~ PV 29 A 2S'O
\"s- ~'/~II ':f' r ','" . ' II • ;,.z -v/,,1' .~ i.,>.. :, ":':' CS-314 ~ .... .: ,",.". 'rA "J '.' •••• ,- "",
® I':) :' .... '.:.:: .. :.:!.. ':(;":~'.;, :~,'~~'I r:~ N .. :.:'-... /), ..... ,. ',.- ';'Jj' ':\""'1 \..:V
t\'~,{ffi:;~i/ 6-3/4" 6-3/4" :V{~;{:ir~l.
®
::. !.i: ;;.',': I....- Penetration ,:,~~:I..,:· ...... ,:
:~{~,~\1~ I ~i}:~i
, 35~ , PV 29 ~320
:s::'3i4 ii', r~, :~;:.:';':': ~ ;;;~~'~~'" ~. ". 1.6 '.' ,~ Jj " ,
tA{~~N&/(}\.. . ~':~ ~. ::'~:::::.,:.~ I S -1/2" I 6 5/8 11
•
. rf:~I:f~:; i1&t;5:
" '1"6'~':3;4 I~~' • I ' .. _.at:.
....... - .A: J .... ,
Fig. 4.8 End Slab after Test PV29
(~
0'1 1',)
63
o o
~
co C\J
> a..
/..
'r-I.J...
64
o
1 (
0 0 3: rr>
0 rr>
c 0
> :.;:: 0
a.. lI.. - > a..
(!,) c o (!,)
a..
" " '" "-/
/
-y
~l I
.....
-00
o (Y")
>-0.....
+J (/)
Q) i-
~ OJ
------It: ro
..0 ro
r--(./)
"0 £:
l.J.J
o
65
-0
0 ~
tf)
--§ > -0 0.. L.. -Q) 0 c
~
> 0..
/ /
=
r--('V)
>-0-
+J (/) (])
l-
SClJ
------~ ro
..Q ro
r--(/)
-0 s::
W
N t()
> / = a.. ~ / f()
/ I
0
66
> Cl.
®
t':': .~(~: ~;~::
t¥ 1i~J ~~~tI"~;;\'Ji~~g~f')1~~; 'r-lL..
32~
" PV 33 4 270
/"
® .. ' " .,/ '::~l ~6-~
, J .~:: .. ~ .:: "~ II.~· ., .: ~ .;""".\ :.: ••.•• : ' •• " :, ::
,', " '0' • \:" 'A' ./~. '.'. I. I@S ", . '., ~ p ,. "- "" '. /l ' •. ,.; IJ, '. " ',/J '( \ ~. ". ,.. '" ' ..' .) ~ . i.: . :"':Y.·~'·~~~ ':\: ';:.~'.: :-::: .:. ,::'1 :: ::·_<~,,·I. '::' ... :~', Penetration Wall ~ ~:,' ~'. ',:'1 :,6'
.. ~\~ "f'~:'; :~i·:I:J· ~ .:. ····r,~· .. :·:/~·; :",<':'I)~':, 10-1/2 11 10-1/4" '.~'.'I·!··:
-Mri; ·ti.\~t.i;; • l ~/j.·:·... 'if' W •• " "
",: .... " . "\, " .. .. --: .. -. , " .... '/1.... .:.:..:-..dJ~
45~ PV 33 A380
®
Fig. 4.13 End Slab after Test PV33
:~.:}.~ :~; ~:'.:': :J:::: ; :', ::. j~.::
..... ~ - ' . . '",~.".' . :' :'06 ••.• .': ,_ ·,:i .. ~;.::..:...::...
®
0"1 -.....J
68
40~--~I----~I--~I~----T--~
II
• II
.. II
A.. 3000 30 ~ -
«J) 20 .. a. I l:;U CU lb... ::» .......
~ «J) «J)
CU +- .... « -
15 en
Q) .... 10 .... ::» Q.. 0 «J) 2000 ~
Q) 20 ---U) -.c
CU .... en Q..
0 0 C
c 'E .... CU - 10 0 ,. Solid Slab c Z
.. 5- in. Penetrations
E I 2 - in. Penet rations :::» E )(
1000 0 10 ~ -~
5
I (0) ( b )
I I 10 0 i I I I 0
9 10 II 12 13 14 9 10 II 12 13 14
Slab Thickness - in.
Fig. 4.14 Failure Prestress and Maximum Nominal Shear Stresses
69 N
W E S J I
W E I I
I 3 10 12 Top View
*2 ,*" 37 39 46 48 *38 ,*47
4 6
*5
13 15
*14
40 42 ,*41
49 51
*50
7 9 16 18 Outside Surface ,*8 *17 Of Head 2-1/4"
43 45 52 54 ~44 *53
Penetra t ion I 19 21 28
30 Level I *20 *29 Penet rat ion :3
¢
22 24 31 33 N
Level 2 *23 *32 ~ i
N
-¢
2527 34 36
~ Y Level :3 *26 *35
64 66 ,*,57 2-1/4" 56 *65
Penetration 2
73 75 82 84
*74 *83
(Typical Dimensioning) 68 60 *59 67 69 ,*68 76 78 85 87
*,77 ,*86
61 63 ,*62
70 72
*71
79 81 88 90
*80 *89
Penetration 4 Penet ra t ion 5
Fig. 5.la Location of Strain Gages in End Slab with 2-in. Penetrations
-~ I
C\J
@ I. 7, 30 " .14
70
! .Penetration
® 10.40" -I- 7,30" -I
No.
Fig. 5.lb Location of Strain Rosettes in the Vertical Plane
(See Fig. 5.la for location of penetrations in plan).
#-i
en (L
o :::r
I W a: ~~ CD(\) (j) w a: (L
.-J ceo z'-: a: w }z: lIc-1
o
71 ~----------------------~
PV 30
It GAGE.a t 4 ~IGE.a '37 ~ GflGE.a 46
~--------4-------~+----L--__ +-__ ------+---____ ~ 0-10. 0 -6.0 -2.0 2.0 6.0 10.0
STRRJ N (Xl 0-4 J
Fig. 5.2 Measured Vertical Strains, Levell Penetrations 1 and 3, PV30
o :::r ~---------------------------------------.-------.
(T')
o ~D
C5cr) 11,.;7---__ _ I--t
U)
lL I
W rr ~DI .-J~ U)C'I U) W a::. lL
---1 a: o z'-: a: w I-Z lIc-1
PV 30
)it GRGEa2 oIL GffiE n 11 I!I m;E 11 38 121 GRGE 0 47
o 4---------~--------+_--------+_--------~------~
tl-25.0 -19.0 -LS.O -7.0 -1.0 STRR] N (X 10-4 )
Fig. 5.3 t·1easured Horizontal Strains, Levell Penetrations 1 and 3, PV30
5.0
72 ,
0 __ -----------------------------------------------. :::r
~
en D-
I W CCo => . (/Jru en w a: 0...
~ a::t:) z'-: cc w }-z: 1---1
PV 30
x GRGE:I1 3 A GAGE:I1 48
a~--------~------~--------_+--------~--~--__4 ~.O 4.0 8.0 12.0 16.0
STRRIN (XI0-4: 7
Fig. 5.4 Measured Compression Diagonal Strains, Levell Penetrations 1 and 3 PV30
20.0
a~------------------------~----~--~--------__. :::r
I--i
en D-
J w a:a => . enru ([) w a: 0..-
--l a:a z'-: a::. w tz f.-..-t
PV 30
1£ GAGE.t1 12 .III GAGE.t1 39
D~--------~-------4--------_+--------_r------__4 D -4:0.0 -32.0 -24.0 -16.0 -8.0
STRRIN (Xl 0--4 J
Fig. 5.5 Measured Tension Diagonal Strains, Levell Penetrations 1 and 3 PV30
0.0
73
0
::r
(T)
0 .......... 0 X· -(1'")
1-1 (j) Q..
I W 0: 0 ~ . (j)(U (J) W a: (L
-1 PV 30 er o
GRGE 12 l3 z~ It a: e. GiiGE ti 40 w I/) GMGE .11 49 J-Z 1-1
0
°-15.0 -9.0 -3.0 3.0 9.0 S T RR 1 N (X 1 0 -4 )
Fig. 5.6 Measured Vertical Strains, Level 2 Penetrations 1 and 3, PV30
15.0
-~--------------------------------------~------~ ::J'
I--iI if) Q..
1 W a: o :::::> • Cf)("\l Cf) w a: Q..
---1 cro z:~ a: w Jz: 1-1
PV 30
Jt GffiE Sl 5 A GflGE Sl l4 e GRG£: Sl 41 f) Gfl('E it 50
-19.0 -l3.0 -7.0 -1.0 STRR] N (X 10-4 )
Fig. 5.7 Measured Horizontal Strains, Level 2 Penetrations 1 and 3, PV30
5.0
74
o ~T----------------------------~----~----------~
"""""" (J) Q...
J W 0: 0 ::> . (f)(\1 (J) w cr:: Q...
....J CIa z:~ a:: w tz: ~
PV 30
o +---------~------~---------+--------~------~ e:'o.o 10.0 20.0 30.0 40.0
STRR 1 N (Xl 0-4 )
Fig. 5.8 Measured Compression Diagonal Strains, Level 2 Penetrations 1 and 3, PV30
o
50.0
~~--------------------~------------------------.
,....... (J)
D-J
W 0: 0 ,:::) . (j)0J en W a: (L
....J a:o z~ a: w ...... z ~
pv 30
a ~--------~------~---------+---------~------~
o-l!O.O -32.0 -24.0 -16.0 -8.0 STRAIN (X10-4 )
Fig. 5.9 Measured Tension Diagonal Strains, Level 2 Penetrations 1 and 3, PV30
75
D ~~----------------------------------------------~
f--1
(f) (L
1 W a: ~C: enC\I (J) w cr:. (L
.--J a:a Z"': a: W IZ $---1
PV 30
)( GllGE n 7 .I\. GAGE 1I t6 e:J GAGE 11 43 en GflGE 1I 52
o +---------+---------+---------~--------~------~
tb.o 6.0 12.0 18.0 24.0 STRRIN (X]O-Lt)
.. Fig. 5.10 f1easured Vertical Strains, Level 3 Penetrations
1 and 3, PV30
D
30.0
~~-----------------------.----------------------~
('I')
o ..-fO
~~ f-.-t (j) 0-
J W a:o :::> • (J)nJ (J) W a: lL
.--J a: o z"": a: w 1-z: ......,
PV 30
-6.0 -2.0 2.0 6.0 STRR I N ex 10-4 )
Fig. 5.11 Measured Horizontal Strains, Level 3 Penetrations 1 and 3, PV30
10.0
76 o ~?----------------------------
1-1
(J)
0-j
W CL o ::::) . (J)N (J) W a: 0-
-1 0:0 Z"': a: W tZ 1-4
~.o 10.0 20.030.0 STRRl N (Xl 0-4 )
pv 30
)t GRGE.Il 5 .. GAGE.Il 54
40.0
Fig. 5.12 f1easured Compression Diagonal Strains, Level 3 Penetrations 1 and 3, PV30
o
50.0
~T---------------------------------~----------~
t>---t
en 0-
J W cr. o :J. en(\! en w a::. a.. -.J er o z"': a: W JZ ......"
rv 30
1)( GBGE AI 451
o +---------r-------~~------_+--------~------~
°-25.0 -20.0 -15.0 -10.0 -5.0 STRAIN (Xl0-4)
Fig. 5.13 Measured Tension Diagonal Strain, Level 3 Penetrations 1 and 3, PV30
-0.0
77 o ~r----------------------------------------
I-i
en 0-
J W 0: 0 ::J . enr:v (f) w a: 0-
--1 0: 0
z'-: a: w IZ 1--1
10.0 20.0 30.0 51AR IN (Xl 0-5 )
PV 30
x G8GE .u .I!I OOGEJ:l to GriCE J:l I!J G8GE .t.1
40.0
,
IS 22 28 31
50.0
Fig. 5.14 Measured Vertical Strains, Levels 1 and 2 Penetration 2, PV30
o -;:f""f""'--T ----------g
1-1
en 0-
J W 0: 0 => . r;.-----__ Cf)(\) (f) w CC CL
-' ceo z'-: a::.. w IZ ~
o
pv 30
JIt GflGE k! 20 I! 00'3E 1:1 23 I!J C~GE ~ 29 c GrlGE: 11 32
+---------+---------+---------~--------~------~ °-25•0 -20.0 -lS.O -10.0 -s.O -0.0
STRR1N (XI0- 4 )
Fig. 5.15 Measured Horizontal Strains, Levels 1 and 2 Penetration 2, PV30
78 o~ ______________________________ .-____________ --. -::1
I-i
(f) CL
J W a:.oC--____ ~ ~ . (f)\V (j) W a:. CL
--.J 0: 0
Z'-: a: W 1-z: 1'-1
PV 30
x GRGE AS 21 .II. GrlGE JS 24 eJ GAGE A1 30 e!J Oi1GE AS 33
D~ ________ ~ ______ ~ ________ -+ __ ~ ____ ~ ______ ~ f::I-10.O -7.0 -4.0 -1.0 2.0 S.O
STRR J N (X 1 0 -4 )
Fig. 5 .. 16 Measured Diagonal Strains, Levels 1 and 2 Penetration 2, PV30
o ::I' -------------------------------------------------,
I-i
en 0...
I W CCo => . (f)1;\J
en w a: Q..
..J 0:0 z"': a: w Iz: 1-1
PV 30
o ~--------~--~---4--------_+--------_r------~
c o.o 3.0 G.O ~LO 12.0 STAR] N (Xl 0-4 )
Fig. 5.17 Measured Vertical Strain, Level 3 Penetration 2, PV30
15.0
79
(:)
'::J' ~------------------------r-----------------------'
1--1
Cf)
0-J
W IT:CJ ~ . Cf)\U (f) W a: 0-
---1 ere Z"-: a: W IZ ~
PV 30
x G'rlGE it 26 .!l GrlGE it 27 e> GiiGE.u 35 I!l GiiGE.t:! 36
o 4---------4---------+----L----+---------+--------4
°-10.0 -6.0 -2.0 2.0 6.0 STRRIN(X10-4 )
Fig. 5.18 Measured Horizontal and Diagonal Strains Level 3, Penetration 2, PV30
10.0
-?-------------------------------------------------, ::J'
PV 30
)C GAGE tt 55 4 GAGE 11 64 e:l GAGE 21 73 C!l OOGE Jl 82
9J.o 2.0 lJ.O 5.0 8.0 STRRIN(Xl0-4 )
Fig. 5.19 Measured Vertical Strains, Levell Penetrations 4 and 5, PV30
10.0
80
o ~T-~--~-----------------------------------------'
1---1 (J) (L
J W cc. o -;:::) . (J)\\I (J) W cc (L
-1 era z:~ cc. w ~ z: '-I
PV 30
)( GRGE ~ 56 .& <frlGE ~ 65 e QiGE ~ 74 t!J G):IGE J:j 83
o.+-________ +-________ +-________ ~--------+_------~ D-10.0 -8.0 -6.0 -4.0 -2.0
STRR J N (X 10 -4 )
Fig. 5.20 Measured Horizontal Strains, Levell Penetrations 4 and 5, PV30
o
-0.0
~----------------------~----------------------~
D--t if) (L
::r
j w a: o -;:::) . cnru (f) w a: Q..
-l crt:) z"": a: w tZ I--i
PV 30
)( nAGE ti 57 All GAGE.It 75
(:)
4---------+---------+---------~--------+_------~
~.O 2.0 4.0 6.0 8.0 STRRIN (XIO-4 J
Fig. 5.21 Measured Compression Diagonal Strains, Levell Penetrations 4 and 5, PV30
10.0
81 0
~~I ~O
X· ~rn
I--f
en (L
J W CL. o ~ . c.nru en w cr. 0...
-..J PV 30 ITO
GRGE .it 66 Z..-: )(
a: A GfiGE Jt 84: W }-Z 1-.,..\
0
0-10. 0 -8.0 -6.0 -4.0 -2.0 STRRIN (XIO-41
Fig. 5.22 Measured Tension Diagonal Strains, Levell Penetrations 4 and 5, PV30
o
-0.0
~~------------------------------------------------~
'""-1
en LL
J W a:. 0 :::; . Cf)C\l Cf) w rr LL
-1 a: o z":' lr.. w IZ ~
~.o 2.0 4.0 B.O S T RR ] N (X 1 0 -4 )
Fig. 5.23 Measured Vertical Strains, Level 2 Penetrations 4 and ~, PV30
PV 30
If GAGE.ti 58 4 OiiGE.It 67 (!) GflGE.ti 76 I!) G8GE lj 8S
8.0 10.0
82
0
::r
('r)
0 ~o
X· ......, (T)
1-1
(f)
£L J
W 0: 0 :::) . (f)\\J (f) lIJ CC ll...
-1 PV 30 0: 0
GBGE It SS z::~ ,. 0: .. OMGE It 68 W
€I GriCE It 77 t- c OOGE .u 86 Z I-f
D
°-10. 0 I I I
-8.0 -6.0 -4.0 -2.0 STRR 1 N (X] 0-4 )
Fig. 5.24 Measured Horizontal Strains, Level 2 Penetrations 4 and 5, PV30
o
-0.0
:::::J' ~------------------------------------------------.
~
(f)
£L I
W a: o :::) . (f)(\J (J) w a:. Q..
-' (['0 z~ a: w rz I-f
PV 30
l!r GAGE.IS 60 4 CAGE.I:I 78
o ~--------+---------+---------+---------~------~
°0.0 . 4.0 B.O 12.0 16.0 ST RR J N (X 1 a -4 1
Fig. 5.25 Measured CompressiDn Diagonal Strains, Level 2 Penetrations 4 and 5, PV30
20.0
83
0
::J'
r"'"'
(TJ
0 ,....-40 X· -en ,., Cf)
CL I
w a: o => . Cf)\U en W a: 0....
.-.J rv 30 0: 0
GRGE!!. 69 2"": iii! a::.. .& GBGE li 87 lLJ t-Z ~
a
(:)-20.0 -16.0 -L2.0 -8.0 -4.0 STRR J N (X 10-4 )
Fig. 5.26 Measured Tension Diagonal Strains, Level 2 Penetrations 4 and 5, PV30
o
-0.0
:::r ~-----------------.----------------------------,
dl--t (f)
0.... J
W a: o :::> . (J)(\I (f) w a: 0....
.-.J a: o z.-.: a:::. w tz: ~
PV 30
~ GflCE A:I 61 A GflGE A:I 79 El GAGE l:l 88
D +---------~------~---------+--------~------~
b_10•O -5.0 -0.0 5.0 10.0 STRR 1 N (X 1 0-4 )
Fig. 5.27 Measured Vertical Strains, Level 3 Penetrations 4 and 5, PV30
15.0
84 D
~~--------------------------------------------~
1-1 CJ)
CL 1
W a: D ::J . CJ)('\.I CJ) w a: CL
-.J ITO Z"':' cr. w .z ~
---
PV 30
)( GflGE l1 62 &. OOGE li 71 tl GflGE tI 80 fI GRGE I:l 89
a +---------~------_4--------_+--------~----
t:J- 10. O -8.0 -6.0 -4.0 -2.0 STRRl N (X1 0-4 )
Fig. 5.28 Measured Horizontal Strains, Level 3 Penetrations 4 and 5, PV30
o
-0.0
~~--------------------~----------------------~
1-1 if) a...
I W a: o ~ . (f)('\J (f) w a: 0....
..J 0:0 z""': a: w JZ ~
o
PV 30
~ GAGE n 63 A GRGE Ii 8t
~--------~-------4----------+---------+-------93.0 5.0 10.0 15.0 20.0 25.0
STRR J N (X 10-4 )
Fig. 5.29 Measured Compression Diagonal Strains, Level 3 Penetrations 4 and 5, PV30
85
b
:::f
r-"'
(1')
0 ,.-40
X· --(Y')
,...... en lL
J W a: ~C:: cn('\J en w 0: D-
--' PV 30
ero GAGE A4 72 Z~ !It
0: A GRGE !t 90 w J-Z I--t
0
°-20. 0 -16.0 -12.0 -8.0 -4.0 STRA] N (XI0-4 )
Fig. 5.30 Measured Tension Diagonal Strains, Level 3 Penetrations 4 and 5, PV30
-0.0
N
86
N ::::::
, .. 5 in. Die. ~ ,
.: .... r· • ........
::.:.~:J.::::>::.,:
W E ~ --~~~~~--~~--~--~----~~~~~
12:00
I 3
Level I -¥2
Level 2 4 6
5
7 9 Level 3 -1L s
3:00
10 12 II
13 15
-¥-14
16 18 17
s Top View
6:00 9:00
19 21
-¥20
28 30 -f29
22 24
-f23
31 33
-¥32
25 27
-f26
34 36 T35
th Penet
:::'\t ....... 8
N
i
fH~~,i~ra::~~; ~~~~ 7 in. Dia. ~
Sect ion A
12:00 3:00 6:00 9:00
37 39 -f38
46 48 -f47
55 57 -fL56
64 56
~55
40 42 -¥41 49 51
-!L SO
58 60 ~59
67 69 -fL68
43 45 -¥44 52 54
-¥53
61 63
-¥62
70 72
-¥71
South Penet ion
Fig. 5.31a Location of Strain Gages in End Slab with 5-in. Penetrations
87
N
+---+-+----+-- W
s
Fig. 5.31b Location of Strain Gages on Inside Face of End Slab with 5-in. Penetrations
-~
s I
C\.I
I C\.I
88
6:00 5"
N
Level
Level 2
Level 3
Fig. 5.3lc Location of Strain Rosettes in the Vertical Plane (*Numerals designating location in plan. See Fig. 5.31a)
89
0
-:::J'
,-...
('I")
0 ...-iD
X • ......... IT>
l'-I
(f) CL
I W a: o :::> . (f)ru (f1 w a: CL
.--J PV 32 ITD
GAGE 1t t Z...; x a:: A GAGE 1t 19 w
f!J GAGE 1t 37 }-
a GAGE .1t. 55 z: 6---i
0
t:I-10.O -7.0 -4.0 -1.0 2.0 STRRIN (XIO-41
Fig. 5.32 Measured Vertical Strains, Levell 12:00 and 6:00 Positions, PV32
o
5,0
~~---------------------------.---------------------~
PV 32
" GRGE At 10 £ GAGE At 28 e> GAGE At 4:5 rJ GBGE.ti 64:
(:)
~----------~--------~r---~----+---------~------~ t':J-5•O -3.0 -1.0 1.0 3.0
STRRIN (Xl0-4?
Fig. 5.33 Measured Vertical Strains, Levell 3:00 and 9:00 Positions, PV32
5.0
~
(f) 0-
J W
o ::J'
IT: 0 .:::> • (f)0I (f) W a: Q...
-.l 0:0 Z~ a: W
90
~----------------------------
PV 32
~ GAGE a 2 & GAGE j1 20 e GAGE a 38 a GAGE -'1 56
o 4---------~--~--~--------_+--------_r~~--~
b-3O. 0 -23.0 -16.0 -9.0 -2.0 S T RR I N (X 1 0 -4 )
Fig. 5.34 Measured Horizontal Strains, Levell 12:00 and 6:00 Positions, PV32
(:)
::r ~----------------------~-----------------------,
PV 32
o ~--------+-------~r---~---+---------r------~
e:;,-LO.O -s.o -2.0 2.0 6.0 S TRR I N (X 1 0-4 )
Fig. 5.35 Measured Horizontal Strains, Levell 3:00 and 9:00 Positions, PV32
10.0
91 0
"::J'
(T)
0 ....-to X· ........ ('1')
~
en a....
J w a:. 0 => . en(\) (f) W a: CL
.-J PV 32 a: o
GAGE n 3 z"': )(
a: &I GReE Jl 21 W e GRGE .i! '39 1- e GRGE J1 57 z l-1
0
°-20.0 -16.0 -12.0 -8.0 -4.0 STRRiN (X10-4 1
Fig. 5.36 Measured Diagonal Strains, Levell 12:00 and 6:00 Positions, PV32.
o
-0.0
~T---------------------------------------------__ ~
(i')
o ".-co (SM ~
if) 0....
I W a: o :::> . (J)N (J) W a: Q.
...J (IO z""': a:: w Iz: .......,
C;:;-10.0
PV 32
-8.0 -6.0 -4.0 -2.0 STARIN (XI0-4 )
Fig. 5.37 Measured Tension Diagonal Strains, Levell 3:00 Position, PV32
-0.0
1-1
en a.-
t w
CI
a:CJ => . enru c.n w a:: D...
.-J er o z....:·-a::. w J-z: B--I
o
92
PV 32
)( GRGE J! 30 & GAGE.lt 66
2.0 1.1.0 6.0 8.0 STRRIN (XI0-41
Fig. 5.38 Measured Compression Diagonal Strains, Levell 9:00 Position, PV32
o
10.0
§:-r---j ---r----_
S~ 1-1
en 0.-
j W a:: 0 ~ . cnru if) W cr: D...
.-J er o z....:1-cr. w J-z: I--j
tJ-B.O
PV 32
j( GAGE $ 4 A GflGE. 22 ~ GAGE. IlO m GAGE J! 58
-5.0 -2.0 1.0 4.0 STRRIN (XI0-41
Fig. 5.39 Measured Vertical Strains, Level 2 12:00 and 6:00 Positions, PV32
7.0
93, 0
::r
r-"'
CJ')
0 ..-. t::J
?5(,~
i'-i Cf)
CL , W CI: a :.:J . Cf)(\J (J)
W a: CL
-.J pV 32
(['0 GAGE .It 13 Z"':- It' OOGf.. a 31 cr a
W fi!J GAGE a 49 I- «!l GAGE AS 67 Z i--1
0 ~
c:JO•O 2.0 y.o 6.0 8.0 STRA1N (Xl 0-4 1
Fig. 5.40 Measured Vertical Strains, Level 2 3:00 and 9:00 Positions, PV32
10.0
~T-----------------------------------------------' :::f
~
(f) 0-
J W 0: 0 => . (f)f\I (j) W a: 0..
.-J 0:0 z"-: a:. w l:z 6--iI
pv 32
k GRGE At 5 4 GAGE a 23 III GAGE.I:! 41 e GAGE.ti 59
O~ ________ ~ ______ -4 _________ -+I __________ ~ ______ ~ (:)-10.0 -8.0 -6.0 -1.400 -2.0 -0.0
5 T RR J N (X 1 0 -4 )
Fig. 5.41 fleasured Horizontal Strains, Level 2 12:00 and 6:00 Positions, PV32
94
o ~?-----------------------------------~----------.
"""'"t (j) (L
I W a: o ~ . (j)fIJ r..n w a: a... --l 0:0 z...; a:. w ~ Z D-f
PV 32 )t GAGE.II! 14 &. GAGE Ji '32 e ~qGE Ji sa B GAGE..Ii 68
O~--------~-------4---------+------~-r------~
°-15•0 -11.0 -7.0 -3.0 1.0 STRAIN (X10-4)
Fig. 5.42 Measured Horizontal Strains, Level 2 3:00 and 9:00 Positions, PV32
o
5.0
~T-------------------------------'----------------'
B-j
en a..
I w 0: 0 :::J . (f)(\J U1 w a: 0....
-' 0:0 z....:; a: w ~ z ~
rv 32 II:: GRGEJi 6 III. GfIGE.it 24 (!) GRGE JlI. 42 !!!l GBGE.Rl 60
o ~--------~------~---------+--~----~------~
o-LO.O -7.0 -!iQO -1.0 2.0 STRAlN (XI0-11 )
Fig. 5.43 Measured Diagonal Strains, Level 2 12:00 and 6:00 Positions, PV32
s.o
95
0
:::r
('I')
0 ,..-to X . ....... (i')
1--11
(f) a....
J w CCo ~ . (f)\U en w cc 0...
.-J PV 32 cr: o
GRGE l!. t5 z"': j(
a: ... c..qc;c: oX 5 t W
-1
0
b_20. 0 -15.0 -10.0 -5.0 -0.0 STRRI N (Xl 0-4 1
Fig. 5.44 Measured Tension Diagonal Strains, Level 2 3:00 Position, PV32
o
S.Q
~~-----------------------------------------------,
('I')
o ......-40
C~ 11-1 en 0... , w 0:. 0 :J . cnru en w a: n
---l a: o z...: a: w tZ ~
PV 32
o ~------~~-------+---------+---------~-------1
~.O 5.0 10.0 15.0 20.0 STRR 1 N (X 1 CJ'4l
Fig. 5.45 Measured Compression Diagonal Strains, Level 2 9:00 Position, PV32
25.0
96
o ~----------------------------------------------.
'::J'
(T)
o or-4 a C~ ...... (j) 0....
1 W rr:o :::> • (j)N if) W a: {L
-1 a: o Z""': a::. w rz #--\
pv 32
~ GAGE jJ 7 dII CflGE tI 25 () GRGE II l! 3 !!l GRGE.IS 61
o ~--------~-------4---------+--------~------~
<=tJ.O 5.0 10.0 15.0 20.0 STRR1N (Xl 0-4 )
Fig. 5.46 r1easured Vertical Strains, Level 3 12:00 and 6:00 Positions, PV32
o
25.0
~~--------------------------------.------------~
PV 32
o +---------~------~--------_+--------~------__4
0 0•0 S.O 10.0 15.0 20.0 STRR I N (X 1 0-4 1
Fig. 5.47 Measured Vertical Strains, Level 3 3:00 and 9:00 Positions, PV32
25.0
97 ' o ~~------------------------------~--------------~
I'--l
(f) a...
J w rr:o ~ . (f)('\J
en w a::. a.. .-J a: o z"': cc w r z ~
PV 32
)( GflGE.It 8 01\. QiGE.It 26 IlII GflGE.It 44 I!!I GflGE.u 62
<::::> +---------~------_4--------_+--~----~--------~
b_10. 0 -7.0 -4.0 -1.0 2.0 5.0 STRAIN (XI0-4 ]
Fig. 5.48 f·1easured Horizontal Strains, Level 3 12:00 and 6:00 Positions, PV32
o ~~---------------------------------------------~
pv 32 jlt GAGE. 17 &. Gf}GE. 35 fa GAGE· 53 It!! GBGE. 71
, . , -s.O -6.0 -4.0 -2.0 -0.0
STRR I N (X 1 0-4 )
Fig. 5.49 Measured Horizontal Strains, Level 3 3:00 and 9:00 Pos it ions, PV32
98
o .~------------~------------.---------------------,
(T)
o
::J'
....--10
C~ '-(f)
CL J
W 0: 0 -:::::J. (j)fV (f) W a: Q...
-' 0:0 Z"': CC w tZ ~
PV 32
.)( G.'iGE.t! S
.iii ffiGE ... 27 /;I) G.C/GE ~ 45 ID GAGE,j, 63
o ~--------~-------4---------+---------r--------4
b ooO 2.0 14.0 6.Q B.O STRRIN (Xl 0-4 1
fig. 5.50 Measured Diagonal Strains, Level 3 12:00 and 6:00 Positions, PV32
I:)
10.0
~T---------------------------------------------~
PV 32
o ~--------r-------~----------+--------~------~
b_1S. 0 -12.0 -9.0 -6.0 -3.0 5 T RR 1 N {X 1 0 -4 J
fig. 5.51 r-1easured Tension Diagonal Strains, Level 3 3:00 Position, PV32
-0.0
99' b
::1 -----------------------------------------------,
pv 32
6.0 L2.0 18.0 24.0 STRA1N(Xl0-4 1
Fig. 5.52 Measured Compression Diagonal Strains, Level 3 9:00 Position, PV32
c
10.0--
(a) Vertica I ( Horizontal (c) Diagonal
S.O Calculated
6.0 B\
~ \
4.01--" ,~ / ... "'-~ .... -- \ i \ I
\ \
~
~ \ I. Tension Tension
.. I I I -5 -4 -3 -2 -I 0 -I 0
Strain x 4-
10 Per 1000 psi Internal Pressure
Fig. 5.53 Comparison of Measured and Calculated Strains at an Internal Pressure of 1000-p~i for a 10.0-in. End Slab. Strains at 12:00 Position (See Fig. 5.31a)
0 0
2 3
10.0
(a) Vertical ( Horizontal (c) Diagonal
c:
.. 8.0 Q) (.) 0 ..... !l... :J en "'0 Q) N !l... :J W w Q) !l... a..
E 0 !l...
LL
6.0
t..I .. II¥J 1.11
4.0
2.0
Tension Tension
00 2 :3 4 5 -5 -4 -3 -2 -I o -3 -2 -I o
Stra X 104
1000 psi Pressure
Fig. 5.54 Comparison of ~1easured and Calculated Strains at an Internal Pressure of 1000-psi for a 10.0-in. End Slab. Strains at 6:00 Position (See Fig. 5.3la)
0
2
c
.. (U (.)
0 ... Il... :::J en
"'0 (U NI .;: :::J Vb Vb (U Il...
a.. If"'"
(U (,) C It) .... Vb
0
(a) Vertical
6.0
4.0
2.0
0 0 2 3 4 5
(b) Horizontal
r I I I
1
Tension
, , , ~
(c) Diagonal
Tension
Strain x 104
Per 1000 psi Internal Pressure
Fig. 5.55 Comparison of Measured and Calculated Strains at an Internal Pressure of 1000-psi for a 10.0-in. End Slab. Strains at 3:00 and 9:00 Positions (See fig. 5.3la)
5
o N
12.5
c
cu 10.0 U o
II®!h. :l
(J)
"'Cl CU 7.5 IN! !h. :l M M CU !h.
Q..
CU U
5.0
c 2.5 o -M
o
(a) Vertiea I
Calculated
"'-, ,{MeaSUred
"'- , , ,
(b) Horizontal (e) Diagonal
" "'tJ Tension
4 Strain x 10 Per 1000 psi Internal Pressure
~ , , , , ~
Tension
Fig. 5.56 Comparison of Measured and Calculated Strains at an Internal Pressure of 1000-psi for a l2.5-in. End Slab. Strains at 12:00 Position (See Fig. 5.31a)
2 3
o w
12.5
C
10.0 Q) U 0 ~
!I... ::::Il
(I)
7.5 "'0 Q)
.~ II-::::Il lin lin Q) \\mo 5.0 0..
(a) Vertical
\
~ \ \ \ \ \
(b) Horizontall (c) Diagonal
Q) 2.5 ~ u c o .... lin
o
dill ~.' Ill! o 2
Strain x 104
Per 1000 ps,i Pressure
Fig. 5.57 Comparison of Measured and Calculated Strains at an Internal Pressure of lOOO-p~i for a 12.5-in. End Slab. Strains at 6:00 Position (See Fig. 5.31a)
3
o ..p:.
12.5 m
(0) Vertical (b) Horizontal I (e) Diagonal .5
Q)
I u
~ 10 .... lI- I ::D
(f) , "'C
, Q) ,
.!:::! 7.5 lI- I ::D , en en Measured p Q) II-
a.. I E I I \\ , \ 0
I (.J1
0 II-
LL I Q)
2.5~ " '\ ~alCUlaled I u
J c 10 -en 0
-Tension ~I ( -4 -3 -2 -I 0
4 Strain x 1O Per 1000 psi Internal Pressure
Fig. 5.58 Comparison of Measured and Calculated Strains at an Internal Pressure of 1000-psi for a 12.5-in. End Slab. Strains at 3:00 and 9:00 Positions (See Fig. 5.3la)
c!
Q) (.)
o '+h :::J
(J')
"'C Q) N l!o.. :::J en en Q) h..
Cl.
E o h
LL
Q) (.)
c o +U)
o
12.5 ( a) Vert ical
10.0
7.5
5.0
2.5
00 2 3 4
( b) Horizontal
Tension
5 6 -4 -3 -2 -I 0
(c) Diagonal
I I
'I
Tension
Strain x 104 Per 1000 psi Internal Pressure
o 2
Fig. 5.59 Comparison of Measured and Calculated Strains at an Internal Pressure of lOOO-p~i for a l2.5-in. End Slab. Strains at Penetrations 1 and 3 (See Fig. 5.2)
:3 4
o 0"1
5·
II:
Q) (,)
ell ..... ,..,. :J if)
Q) (,)
II: ell -U»
o
12.5
10.0 q
7.5
5.0
2.5
00
(a) Verticcli
\ \
\
2
\
3
\ b
4
(b) Horizontol (c) Diagonal
o I I I I J \ \ \ \ \
b
5 6 -4 -3 -2 -r 0 -4 -3 -2 -I 0
4-Strain x 10 Per 1000 psi Internol Pressure
2
Fig. 5.60 Comparison of Measured and Calculated Strains at an Internal Pressure of 1000-psi for a 12.5-in. End Slab. Strains at Penetrations 4 and 5 (See Fig. 5.2)
\
~
3 4 5
o -.....,J
12.5 m
(a) Vertical (b) Horizontal (c) Diagona I
at:
.,. 10.0 @) \ u 0 \
'3-\ f!.",
:l \ U')
\ Calculated "'0 7.S \ @) N \ \ f!.",
~ ~ :l en U) \ @)
\ \ '\ 0 f!.",
Q.. \ co 5.0 \
E \ \ 0 ~Measured \ f!.", \ u.. \ \ @) \ \ u b at: fJ 0 -. en £5
Tension I Tension
~I I I Iv -4 -3 -2 -I 0 I 0 2 :3
Strain x 104
Per 1000 psi Internal Pressure
g. 5.61 Comparison of Measured and Calculated Strains at an Internal Pressure of lOOO-psi for a l2.5-in. End Slab. Strains at Penetration 2 (See Fig. 5.2)
A1.1 Concrete
109 '
APPENDIX A
A1 MATERIALS
The vessels were cast from concrete mixed in the laboratory. Two
different mixes were used, one for the cylindrical skirt and one for the
end slab.
The first mix was used for the skirt up to a level of 2-in. below
the reentrant corner. Two batches of concrete were required to cast the
skirt. The concrete contained pea-gravel aggregate, sand and type III
cement. The proportions by weight of cement: sand: gravel were 1.00:
2.77: 3.07 and the water cement ratio was 0.67.
The second mix was used for the slab and the top 2-in. of the skirt
and was made in one batch. The concrete contained crushed limestone ag
gregate, sand and type III cement. The proportions by weight of cement:
sand: gravel were 1.00: 3.51: 3.40. Type III cement was used with a
water cement ratio of 0.80.
Five 6 x 12-in. cylinders were cast from each of'the first two batches.
Eight 6 x 12-in. cylinders and eight 6 x 6-in. cylinders were cast from
the third batch used in the end slab of the vessles. The properties of the
batches used in the end slabs of the vessels are shown in Table A.l.
Al.2 Longitudinal Reinforcement
PV 26 was prestressed with the rods from the previous series of tests,
namely 0.775-in. diameter stressteel rods with an ultimate stress of
110
140 ksi. However, the threads on several of these rods stripped off
during prestressing.
New rods and bolts were purchased and used in all of the subsequent
tests. The results of a tensile test of a 30-in. stressteel rod are
plotted in Fig. Al. The strain was measured using an eight-in. extensom
eter. The ultimate stress in the rod, which had a measured cross~sectional
area of 0.471 sq. in. was 140 ksi.
Al.3 Circumferential Reinforcement
The wire used to prestress the vessels circumferentially was obtained
from ARMCO, Kansas City, Missouri. The wire was 0.08-in. diameter extra
high strength high carbon rope. Two strain gages were attached on opposite
sides of wire samples cut from the unstressed coils. These samples were
subjected to tensile testing and the stress versus average strain was
plotted. The Young's Modulus for the wire was found to be 30 x 106 psi.
All of the samples failed at the grips of the testing machine. The failures
occurred at an average load of 1400 lbs or approximately 280 ksi. Thus
it can be assumed that the actual ultimate load is somewbat higher.
Al.4 Liner Materials
The neoprene used to seal the pressure vessels was purchased in 100-ft
rolls. The sheets were 36-in. wide and 1/16-in. thick. It was specified
as "60 Durameter Shore A Black Neoprene Sheeting, Type #260. 11
The O-ring material was obtained in 100-ft lengths. The diameters of
the 3/16~in. and 3/4-in. O-ring stock were 0.210 ± 0.010 in. and 0.750 ±
0.010 in. respectively. The material was specified as "70 Durameter Buna-N
0- ri ng Cord Stock. II
111
The caulking used to seal and hold the neoprene intact was specified
as "General Electric Construction Sealant, SE-1204 Neutral in 1/12 U.S.
Gallon Paper Tubes. II
Sheets of 0.104-in. thick steel and 16 oz. soft copper were used on
the sides and end slabs of the vessels.
112
A2 FABRICATION
A2.l Casting and Curing
All pressure vessels in the current series, PV 26 through PV 33, were
cast in the same outer steel form. This form was rolled from 5/l6-in.
steel plate and reinforced with rolled 2 x 2 x l/4-in. angles. The inner
form was basically a closed steel cylinder. Its height could be adjusted
by adding 2 l/2-in. steel bands around the bottom or open end of the cylinder.
This made it possible to vary the thickness of the concrete vessel head
Both the inner and outer forms were bolted to a l/2-in. thick base plate.
Sixty holes were drilled 3/8 in. into the base plate to receive the 7IB-in.
diameter aluminum rods which form the openings for the longitudinal pre
stress rods. The tops of the rods were secured by a template of l/2-in.
steel which was attached to the outer form by sections of 4-in. wide channel.
The center of the template was cut out to allow easier access when casting
and trowelling. For the vessels which had penetrations in the head, holes
were drilled and tapped in the top or closed end of the inner form. Steel
pipes having the desired length and diameter were then bolted to the form.
All vessels were cast in three batches. The first two batches con
taining pea gravel aggregate, were used for the skirt of the vessel up to
a level approximately 2-in. below the top of the inner form. The third
batch, which contained a limestone aggregate, was used to complete the re
maining skirt and head slab. The concrete was vibrated internally with an
electric vibrator during casting. The sidewalls of the vessels were rein
forced with 40 No.4 rebars, providing a reinforcement ratio of approximately
113
one percent. The bars were placed longitudinally around the outside of
the skirt with about a 1/2-in.cover provided.
After the concrete had been placed and vibrated, the surface was
trowelled until the concrete had begun to set up. The oiled aluminum rods
and penetration pipes were periodically twisted during this time until the
concrete was firm enough to allow the pulling of the rods and penetration
pipes. This was accomplished by first removing the 1/2-in. steel template
and then very carefully rotating and lifting the rods and penetrations out.
The vessel and cylinders were then covered with wet burlap and plastic.
On the second day after casting, the forms were struck and a grinder was
used to smooth out any rough spots on the surface of the vessel head. The
cylinders were removed from their forms and placed around the vessel and
again covered with wet burlap and plastic. The wet curing process continued
until the seventh day after casting.
A2.2 Circumferential Prestressing
The pressure vessels were circumferentially prestressed in the Civil
Engineering Machine Shop on a specially built prestressing rig. The 0.08-in.
diameter wire was applied in a series of five "belts'l as shown in Fig. A.2.
Each of the fi ve "be 1 ts II contai ned between 290 and 300 wraps of wi re app 1 i ed
in six layers of about 50 turns each. The wires in each band were kept from
slipping out of position by a series of steel plates bolted into coupling
nuts cast into the concrete. The coupling nuts were cast into the vessel
by drilling holes in the outer form and bolting them from the outside against
the inside of the form. A 0.5-in. washer bolted to the coupling nut provided
114
bearing for anchorage. When the forms were struck, the bolts were unscrewed
and the forms removed, leaving the openings of the nuts exposed. Screw-in
clamps on each of the bands served as tie-offs. for the beginning and end of
each band of prestressing wire.
The prestressing operation was initiated by securing the wire in one of
these clamps. The first wrap of prestress was applied at a reduced load to
facilitate the proper alignment of the wire on the vessel. Subsequent wraps
of the D.08-in.diameter wire were applied at a tension of between 700 and
730 lbs. Ten steel rods were used to bolt the vessel securely on the lathe.
A schematic diagram of the prestressing apparatus is shown in Fig. A.3.
The extra high strength wire was shipped in coils weighing around 500 lbs.
It was necessary to rewind the coils of wire on a large spool to provide a
more uniform rate of feed. The spool of wire was then mounted on a stand
at the rear of the prestressing rig. A rope was wrapped around the spool
axle and kept taut during the entire operation. This prevented the feed
spool from gaining momentum and letting out more wire than was needed. The
wire was first passed around a friction pulley a total of five times and
then pulled over the first of the mounted pulleys. The wire was then passed
under the pulley mounted on the 1500 lb. weight and back up and over the
second mounted pulley. Finally, the vJire was passed under a smaller pulley
attached to a dynamometer and secured on the vessel (Fig. A.4).
Two automoti ve brakes were mounted on the axl e of the fri cti on pull ey
and they ultimately controlled the rate at which the wire was fed from the
spool. It was the brakeman's function to control the rate of feed so that
it equalled the rate at which the rotating lathe wrapped the wire onto the
115
vessel. The brakeman was able to gage this proper rate by watching the
suspended 1500 1b weight and its relative motion. If the weight was rising,
that indicated that the lathe was taking on wire faster than the spool was
feeding it. Thus, the pressure on the brake had to be reduced. Conversely,
if the weight was descending, the brake pressure had to be increased. Once
the wrapping operation was begun, the weight was kept suspended at all times
to maintain full tension in the wire. Approximately two hours was required
to wrap one "be1tll of 295 turns.
After all five bands had been applied, the vessel was unbolted from the
turret lathe and inspected for cracks. In each case, a series of two or
three circumferential cracks was found on the inside walls of the vessel at
a spacing of approximately nine in. from the open end.
The dynamometer or load cell attached to the last pulley was connected
to a strip chart and continuous strain readings were taken during prestressing.
From these readings, a value for the average tension in the wire was obtained.
The effective prestress was determined with the help of information on
shrinkage and creep characteristics of the concrete used (Reference, Vol. II,
Fig. A. 3). The lIanchoringll stress for each wrap was mea·sured. The instan
taneous stress reduction on each wrap caused by subsequent wraps was calcu-
lated using a linear elastic model of the vessel. The following expressions,
based on experimental data, were used for time-dependent strain changes in
the concrete:
10-3 = ----::---=--:--
Sc 1250 + 30,000 tp
(A.l )
116
Esh = (A.2) 30,000 1700 + tc
where EC = creep strain per psi
~h = shrinkage strain
tp = time after prestress, days
tc = time after casting, days
Both expressions are intended to apply to the concrete used and
for values of t less than 120 days. The rate-of-creep method was used to
determine the effective prestress with an integration interval of one
day. The calculated reduction in stress for the prestressing bands
near the end slab ranged from approximately 15 to 20 percent (Table A.2).
A2.3 Longitudinal Prestressing
Sixty stressteel bolts were used to prestress all of the vessels. Strain
gages were placed on 15 of the rods and were calibrated in the laboratory.
Loading of the bolts was accomplished with a 30-ton Simplex jack with the
scheme shown in Fig. A.5. A continuous steel plate 1 1/4-i~. thick was used
as a bearing plate. The fifteen gaged rods were pulled first, with strain
readings taken before and after. The load in the bolts after the jack was
released varied from 40 to 45 kips. After the rest of the bolts were pre
stressed the gaged bolts registered loads of less than 40 kips. This loss was
attributed to the effects of creep and to the fact that the loading of a bolt
adjacent to an already loaded bolt tended to reduce the force in the loaded
bolt. To minimize this effect, the rods were pulled a second time in the
117
same manner. In this way, a force approaching 45 kips was attained in each
of the rods. A final set of readings from the gaged bolts was taken immedi
ately before the test so that the vertical prestressing force was known at
the time of the test.
A2.4 Liner
A detail of the typical liner used for all -the vessels is provided in
Fig. A6. For the vessels having penetrations, steel plugs or plates were
used to cover the holes on the inside of the vessel (Fig. A. n. A welded
steel can~ 0.104-in. thick, was then grouted into place with the use of an
electric vibrator. The 16-oz. soft copper can was soldered in next with all
copper to copper and copper to steel connections tinned and sweated. The
vessels were then lightly prestressed longitudinally and pressurized to 50
psi gas pressure to check for leaks. A layer of 1/16-in. thick neoprene was
placed over the copper and secured with rubber cement and General Electric
Silicone Caulking. A 3/4-in. neoprene O-ring was also installed at the
junction of the end slab and the sidewall. An aluminum expansion ring was
used to hold the neoprene securely in place around the bottom of the side
wall. The seal between the steel base plate and the sealing ring was made
by compressing a 0.210-in. O-ring into the groove in the base plate (Fig. A.8).
118
A3 TEST SETUP
All testi ng was conducted in the basement at the east end of the .Ci vi 1
Engineering Building. The vessels, having been lined and prestressed, were
transported by crane and fork lift to the test room. On -the day prior to
testing, the vessels were filled with water to within approximately 1/2-in.
of the end slab. An oil pump was used to pressurize the vessel to failure.
This procedure greatly reduces the violent release of energy that occurs when
using gas pressurization. However, it was still necessary to contain the
explosion by the use of steel channels bolted across the top of the prestress
rods in a criss-cross pattern (Fig. A.9).
During the tests, the door to the test room was barricaded and warning
signs were posted in all adjacent corridors. All operations were conducted
remotely in an area at the east end of the Crane Bay, directly above the test
room located in the basement. Here strain measurements were taken and re
corded on a teletype equipped with a paper tape, deflection readings were
taken from the two television monitors, and the internal pressure of the ves
sel was monitored and controlled by the test personnel.
119
A4 INSTRUMENTATION AND TEST PROCEDURE
In general, the instrumentation of the vessels consisted of deflection
dials across one diameter of the head and down the side on a line at one
end of this diameter, strain gages on the inside surface of the head and the
surfaces of the penetration walls and load cells on the fifteen prestressing
rods.
Deflections across the head of the specimen and on the side wall were
measured with O.0005-in. Brown and Sharpe Dial Indicators. For PV 26 the
dial gages were connected to push rod extensions in direct contact with the
surface but due to the explosion at failure, all of the dials on top were
destroyed. For PV 27, push rods were again used for the sidewall gages.
However, the ~ad dial gages were connected to piano wires which were strung
over ball bearing pulleys and attached to metal tabs glued to the specimen.
Tension springs connected to the back end of the dial gage plunger kept the
piano wires taut. This system proved unsatisfactory due to the great amount
of internal friction which markedly reduced the sensitivity of the gages.
Thus, for PV 28, the head gages were mounted directly above the specimen and
connected to piano wires running vertically down to the vessel head and at
tached to the metal tabs glued to the surface (Fig. A.10). The gages were
protected from damage by a series of steel channels bolted over the head.
This method proved satisfactory and was adopted for all subsequent tests.
The dial gages were read with a closed circuit television hookup with the
monitors situated on the first floor at the east end of the Crane Bay. Two
television cameras equipped with telephoto lenses were used to read the de
flection gages (Fig. A.ll).
120
Strain gages were used on the inside surface of the concrete and in the
penetrations were limited to two types: the BLH type A12 which has a one-in.
gage length and 3/32-in. gage width, and the BLH type AR-2-S6 which is a
rosette having a gage length of 3/4-in. and gage width of 9/64-in. Careful
steps were taken to ensure a smooth surface and good bonding for all gages.
The concrete surface was first sanded to a smooth finish. A hydrocal paste
was then applied over the surface to fill in any holes or indentations. The
surface was again sanded down and a layer of cement glue was placed on the
concrete and allowed to set to a smooth, hard finish. The gages were then
attached to this prepared surface with Eastman 910 cement (Fig. A.12). In ad
di t i on, a soft rubbery protecti ve coati ng was placed over a 11 stra in gages on
the inside surface of the vessel. The wires from the inside gages were run
down the inside wall of the vessel and out between the concrete skirt and
the one-in. steel ring. Channels 1 in. wide by liB-in. deep were cast into
the concrete to accommodate the gage wires.
Load cells were used to measure the changes in force in the prestress
bolts. Four strain gages were cemented to the outside of these bolts and
wired into a full bridge. Before they were used, the load ~ells were cal
ibrated in a testing machine.
Strains were read by a Pivan switching strain indicator located at the
east end of the Crane Bay in the Civil Engineering Building, directly over
the testing room one floor below. The load cells were calibrated with a
10k ohm resistor while all others required a 60k ohm resistor.
Pressure was applied to the inside of the specimen by a high-pressure
hydraulic pump with a maximum capacity of 10,000 psi. During a test the
121
gas pressure was increased in increments and once the pressure was set and
became stable all measurements were taken. Approximately 3 to 4 minutes
were usually required for each set of readi~gs. The size of the pressure
increments varied among the tests.
TABLE A. 1
Concrete Properties
Age Modulus of Splitting Compressive Strength
Mark @ Test Slump Elasticity Strength Batch 1 Batch 2 Batch j-[End-Slabl Days in. psi x 106 psi psi psi psi
PV26 98 3 1/2 3.9 445 7320 7910 6710
PV27 102 4 3.8 460 7620 7610 6845
PV29 52 3/4 .3.8 450 5760 6660 5480 N N
PV28 117 2 3.9 440 7620 812.0 6420
PV30 93 2 3/4 3.7 495 6350 5890 6300
PV31 85 2 1/2 3.7 380 4910 5890 4970
PV32 114 2 1/2 3.8 450 4940 5560 5720
PV33 95 1/2 3.7 375 5030 5670 4875
TABLE A.2
Longitudinal and Circumferential Prestress
Longitudinal Prestressing Circumferential Prestress Mark
Indexa Mean Force in Bands 1 & 2b Age Force Age Initial Final Final ·c Days per rod Index
kips psi days kips kips psi
PV26.1 56 40.0 4890 19 199 159 1590
PV26.2 93 41.0 5010 199 155 1560
PV27 126 43.7 5340 27 194 156 1560
PV29 45 43.6 5330 29 199 171 1710
PV28.1 83 44.1 5390 56 203 175 1760 N w
PV28.2 111 47.1 5760 203 170 1700
PV30 85 43. 7 5340 30 206 169 . 1690
PV31.1 64 43.8 5350 30 201 171 1710
PV31.2 80 45.0 5500 201 168 1680
PV32 112 45.3 5540 83 203 178 1780
PV33.1 76 44.4 5430 34 205 173 1730
PV33.2 84 44.2 5400 205 170 1700
aTota1 force divided by horizontal cross-sectional area of cavity
bMean force in bands 1 and 2 around end slab at time of prestressing
cMean effective prestress in bands 1 and 2 around end slab at time of test
124
140
120
100
Based On Actual Area 0.471 in~
60
40
.002 .004 .006 .008 ,010 .012
Strain
Fig. A.l Stress-Strain Curve for Stressteel Rods
4"
4"
I.' r
" ." :.: 4: " J ...
" . 'V. :
r" .. "
I' ..... ,
.0.'
,4 " ': ./0
•• •• 1
,',' .-. ,:.~ ~:
.. ' .. : :::1;. ·t:V J
:.',
, .... :: ... , .. ~
... " 0." /
: ". '.
:: ~.;:. '.
" , •• , t·
::A·~· ... ,11
125
Steel Retain ing Rods
Base Plate Of Turret Lathe
Fig. A.2 Locations of Bands of O.OS-in. Wire Used
1/2" Steel Bands
}
Band
Band 2
Approximately 295 Wraps In Six Layers
Band 4
Band 5
Spool
Fr iction
Fig. A.3 Schematic View of Circumferential Prestressing Rig
Turret Lathe
N (j)
Fig. A.4 Isometric View of Prestressing Rig
N -......J
Vessel ~--------------=-- ~
128
I I I I I I I I I I I I I I I I I
Pull rod
Stressteel nut
30-ton Simplex jack
Coupling
Jack chair
Stressteel rod Stressteel nut 3" 0.0. bearing plate
311 0.0. sleeve to protect instrumentation
Vessel
1/4" thick bearing ring
Fig. A.5 Apparatus Used for Longitudinal Prestressing
lID Steel
129
Pressure Vessel
I/t6 II Neoprene
G. E. Si licon Caul k
O-Ring
A II Copper to Copper and Copper to Steel Connections Tinned and Sweated.
1/1611
Neoprene
16 oz. Soft Copper
0.104 in. Thick Steel
114 II Grout Layer
0.210 in. 0- Ring
411
Steel Plate
Caulk
Ring
Fig. A.6 Typical Liner Details (not to scale)
130
I ..
Fig. A.7 Steel Closure Plates
stee\ Plate 4 in.
, Of Groove Oe1al\
131
Groove
Plate Base
Fi 9- 1\. Base Plate 8 steel
1\ Oiam. 30 - Id Ho\es Dr i\\e
132
Fig. A.9 Protective Steel Channels Across Top of Vessel
Fig. A.10 Dial Gages Measuring End-Slab Deflections
133
Fig. A.11 Closed-Circuit TV Cameras to Read Dial Gages
Fig. A.12 Strain Rosettes
... ;--.
134 2.5
2.0
1.5
1.0 fJ)
..:s: 12 34 5 6 7 8 Dial Ga(Je ...
cu 0.5 ~
:J fJ) fJ)
cu ~ a a..
01 c:
2.0 ~
cu -c:
1.5
1.0
13 12 II 10 9 8 Dial Gage -::::...~::::::::;;;;.---
0,5
°0 0.001 0.002 0.003 0004 0.005 0.Q()6 0.007 0008 0009 0.010
Deflection, in.
fJ)
~
~ i,5 ~
:J en en 4» ~ 1.0 a..
OOtO
Deflection 1) in.
Fig. A.13 f'1easured Pressure-Deflection Curves, Test PV26.1
135
3.0 Max. Pressure Reached
en 2.5 3 4 5 6 7 8 Dial
Gage ~
Q,) »-::::lI 2.0 en
211 . 10 at 2-1/2" ~ 2 1/2" r I I I \ I I I I '1: ~ [I -
en Q,) »-a..
1.5 c c »-Q,) .....
1.0 c
0.5
3.0
9 8 Dial en 2.5 Gage ~
... Q,) »-::::lI 2.0 en cn Q,) »-
a.. 1.5
c c »-Cl) .....
1.0 c
0.10
Oaf lect ion, in.
Fig. A.14 Measured Pressure-Deflection Curves End Slab, Test PV26.2
Gn ~
CIt) l!...
~ .Gn CIt) l!...
a..
OJ c: l!... CIt) -c:
3.5
3.0
2.5
2.0
1.5
1.0
0.5
136
0.002 0.004 0.006 0.008 0.010 0.012 0.014 0.016 0.018 0.020
Ultimate::: 2610
2500 psi
- - --- 2000 psi
--- 1000 psi
Deflection, in.
0.10
Fig" A .. 15 ~:1easured Pressure-Deflection Curves Side Hall, Test PV26 9 2
137
3.0
Max. Pressure Reached en 2.5 ~
Q) i>- Dial Gage I 2 3 4 ::J 2.0 en
5 6 7 8
2" . 10 at 2-1/2" ~ 2 1/2"
I", I I I I I I I ,: IG -en Q) i>-
a.. 1.5
c J:: i>-Q) +-
1.0 It:
0.5
3.0
U) 2.5 ~
.,. Q) i>- 13 12 119 10 8 Dial Gage ::J 2.0 en en Q) i>-
a.. 1.5
c It: i>-Q) +- 1.0
0.5
0.0! 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10
Deflection, in.
Fig. A.16 t1easured Pressure-Deflection Curves End Slab, Test PV27
138
3.0
2.5
15 141920 18 Dial Gage
o 1.5 c Dia I Gage /6 Was Not Functioning . b. W -C
1.0
0.5
o~--~~--~~--~----~----~----~----~----~----~------o 0.002 0.004 0.006 0.008 0.010 0.012 0.014 0.016 0.018 0.020
U I t imate :: 2400 psi
2000 psi
1000 psi
Deflection, in.
0.10
0.08 I
0.04
/t-\ .fP \ ,- '\ I I \
I \ I I '\
0.06
, 0.02 ' ........ _ ... ---'" -~ .................
Fig. Ae 17 Measured Pressure-Deflection Curves Side Wall, Test PV27
en ~
Q) a... :l en en Q) a... a..
0 £: a... Q) -£:
en ~
... C1> ).,.
:::lI en en C1> a...
a..
0 £: a... C1> -
3.0
2.5
2.0
1.5
1.0
0.5
3.0
2.5
2.0
1.5
1.0
0.5
Max. Pressure Reached
i 2 3 4
II 10
0.01 0.02 0.03 0.04
139
5 6 7 Dial Gage
16 ·~)'/::/h·::~i:~\:t:~rM1~j~;
9 8 7 Dial
0.05 0.06
Gage
0.07 o.os
'~:~.::..~.::: ~<~~:::
@"~~
0.09
Deflection, in.
Fig. A.1B Measured Pressure-Deflection Curves End Slab, Test PV29
0.10
3.0
2.5
2.0
1.5
0.5
140
16 15 14 17 r 20
18/19 Dial Gage
0.002 0.004 0.006 0.008 0.010 0.012 0.014 0.016 0.018 0.020
Deflection, in.
Oil 0.08
I
Ultimate = 2400 psi I 2300 psi
0.0\
A 1i:i:ZmIII ____
2000 psi 0.04
--- 1000 psi ,.;' f '" ,I' " 0.02 ,.;' .. ..... ..... -"" .............. _ .. ..--
Fig. A.19 Measured Pressure-Deflection Curves Side Wall, Test PV29
.." ~
... GJ ~ :;::) .." .." GJ ~
a..
0 c ~
GJ .... C
o c
2.5
2,0
1.5
1.0
0.5
0
G; 0.5 ....
141
0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09
Deflection, in.
.004
Deflection, in.
Fig. A.20 Measured Pressure-Deflection Curves, Test PV28.1
0.10
.020
142
It/) 2.5 211
I' 10 at 2-1/2" ~ 2-1/2"
I I I I I I I l'I:IAi
oX
(l) .... :::s 2.0 It/) It/) (l) .... a..
1.5 c t: .... (l) ....
1.0 t:
0.5
3.0
0.5
0.04 0.05 0.06 0.07 ooa Q09 0.10
Deflection» in.
Fig. A.21 ~·1easured Pressure-Deflection Curves End Slab, Test PV28.2
143
ge
4J) 2.5 ~
OJ
"'" ~ 2.0 4J)
OJ "'" a.. 0 1.5 c "'" OJ -C
O~----~----~----~----~----~----~----~----~----~-----d o 0.004 0.008 0.012 0.016 0.020 0.024 0.028 0.032 0.036 0.040
Deflection, in.
0.10
0.08
Ultimate = 3765 psi 0.06
3400 psi
.......... __ ........ - 3000 psi 0.04
--- 2000 psi 0.02 .......... --- 1000 psi -.......... ---
Fig. A.22 Measured Pressure-Deflection Curves Side Wall, Test PV28.2
144
Max. Pressure Reached 2 3 4 6 7 Dial Gage
3.0
en 2.5 ~
Q) Ibm. ::J (/)
2.0 en Q) Ibm.
a.. 1.5
c II:: Ibm. Q) - 1.0 II::
0.5
II 9 8 7 Dial Gage
3.0
en 2.5 ~
... <lJ) Ibm. :l 2.0 en en <lJ) Ibm.
a.. 1.5
0 II:: Ibm. <lJ)
Diai Gage 10 Did Not Move - 1.0
0.5
0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 Q09 0.10
Deflection, in.
Fig. A.23 Measured Pressure-Deflection Curves End Slab, Test PV30
en ~
OJ ~
::J en en OJ ~
a.. c c::: ~
OJ .... c:::
145
3.5
17 20 19 18 Dial Gage 3.0
2.5
2.0
1.5
1.0
0.5
o~----~--~~--~~--~~--~----~~--~----~----~----~ o 0.004 0.008 0.012 0.016 0.020 0.024 0.028 0.032 ·0.036 0.040
Ultimate::: 3210 psi
3100 psi
----- 3000 psi
---2000psi
---1000 psi
in.
Deflection, in.
0.10
0.08
0.06
0.04
0.02
Fig. A.24 Measured Pressure-Deflection Curves Side Wall, Test PV30
146 2.5
2.0
1.5
1.0 f/)
~
... QJ 0.5 b.. ::J f/) f/) QJ b.. a a..
c c:
2.0 b.. QJ -c:
1.5
10 8 II .9 7 Dial Gage
1.0
0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10
Deflection, in.
8 Dial Gage
aOlo
g. 25 ~1easured Pressure-Deflection Curves, Test PV31.1
VJ ~
(J) ~
::::J VJ VJ (J) ~
a..
0' c ~ (J) +-c
3.0
2.5
2.0
1.5
1.0
0.5
3.0
2.5
1.5
1.0
0.5
147
Max. Pressure Reached
0.01
Dial Gage 4 Was Not Functionjng Properly -o
c,o
Gage
8 7 Dial Gage
Dia I Gages 10, II Were Not Functioning
0.02 0.03 0.04 0.05
Def lect ion, in.
Fig. A.26 Measured Pressure-Deflection Curves End Slab, Test PV31.2
0.10
148 3.5
3.0
Dial Gage 19 ~ 2.5 ~
OJ ~
~ 2.0 en w ~
0-
0 1.5 c ~
OJ -C
0.5
0.002 0.004 0.006 0.008 0.010 0.012 0.014 0.016 0.018 0.020
Deflection, in.
0.10
0.08
0.06 2700 psi
----- 2000 psi 0.04
--- 1000 psi 0.02
Fig. A.27 Measured Pressure-Deflection Curves Side Wall, Test PV31.2
149
Max. Pressure Reached
3.0
Dial Gage
en 2.5 ~
CD 1!... ::J 2.0
211
• 10 at 2-1/2" I ~ 2-1/2"
I, I I I I I I I 'I:~ en en CD 1!...
a.. 1.5
c c 1!... <U +-
1.0 c
0.5
3.0 13 12 111098 Dial Gage
en 2.5 ~
"" <U 1!... ::J 2.0 en en <U 1!...
a.. 1.5
c c l....
1.0
0.10
Deflection, in.
Fig. A.28 Measured Pressure-Deflection Curves End Slab, Test PV32
en ~
OJ ll:a"
~ en OJ ll:a"
Cl.
0 c ~
3.5
3.0
2.5
2.0
1.5
0.5
150
15 14 16 17 19 20 18
0.002 0.004 0.006 0.008 0.010 0.012 0.014 0.016 0.018 0.020
Ultimate::: 3075 psi
2800 psi
0.10
0.08
0.06
--~-- 2000 psi 0.04
--- 1000 psi 0.02
in.
on, in.
Fig. Aa29 r1easured Pressure-Deflection Curves Side Wall ~ Test PV32
V)
..lIl:: ...
w "'" ::::J V) V)
w "'" a..
e c "'" w -c
iIV
"'"
e c "'" iIV -C
2.5
2.0
1.5
1.0
0.5
0
2.0
1.5
0.5
2.5
2.0
1519 1.5
1.0
0.5
°0
1 51
Dial Gage
Dial Gage
0.01 0.02
14- 20
I I 0.002 0.004
0.03 0.04 0.05 0.06 0.07 0.08
Deflection, in.
18 Dial Gage
Gages 16» 17 Did Not Change Throughout The Test.
g. A.30 Measured Pressure-Deflection Curves, Test PV33.1
0.09 0.10
0.020
152
Max. Pressure Reached
12:3 4 5 67,8 Dial Gage . --...-:~P-3.0
2.5
Q) ~
:J 2.0 w w Q) ~
n. 1.5
0.5
13
1.0
0.5
0.04 0.05 006 0.07 0.06 009 OoiO
, in.
Fig$ A~31 Measured Pressure-Deflection Curves End Slab, Test PV33.2
153
18 19 20 Dial Gage
2.5
2.0
1.5
1.0
0.5
OL-----~----~----~----~----~----~----~----~----~ ____ __ o 0.004 0.008 0.0 12 0.0 16 0.020 0.024 0.028 0.032 0.036 0.040
Ultimate::: 3100 psi
3000 psi
2000 psi
1000 psi
Deflection, in.
0.10
0.08
0.06
0.04
0.02
Fig. A.32 Measured Pressure-Deflection Curves Side Wall, Test PV33.2
154
APPENDIX B
TEST DATA
INTRODUCTION
Eight vessels were tested in the current phase of the investigation.
This appendix contains specific information on the different character
istics of each vessel. The dates of casting, prestressing, and test
ing of each specimen are recorded in Table B.l.
A brief description of each vessel is provided in this section.
Since the basic materials and procedures used for each specimen were
detailed in Appendix A, they will not be repeated in each vessel
description. The graphs of pressure-deflection readings for each
test are included at the end of Appendix A.
B.l. Test Vessel PV26 (Solid, 10-in. head)
Since PV26 was the first vessel tested in the current series it
had many unique features which were later modified or eliminated from
the other vessels. It was circumferentially ~restressed going from
the bottom to the top, an operation which resulted in radial cracks in
the end slab after the bottom four bands were completely prestressed.
These cracks closed upon prestressing the last' band located around the
slab.
PV26 was also the only vessel that was longitudinally prestressed
using the rods from the previous series. The vessel was originally
lined with only a welded steel can grouted into place. However, the
first test of the vessel had to be aborted after reaching 800 psi due
155
to leaks through the welds of the steel can. The specimen was lined
with copper and neoprene over the steel as described in Appendix A.
This served as the liner for all subsequent tests with only minor
changes. The second test proved successful with an ultimate pressure
of 2610 psi obtained. The deflection readings of the head surface
were obtained using push rod extension gages in direct contact with
the concrete. The readings taken in the two tests were quite good using
this setup. However the explosive manner of failure destroyed all
of the head deflection gages and thus a new method had to be used for
later vessels.
B.2 Test Vessel PV27 (six 5-in. penetrations, 10-in. head)
PV27 and all subsequent vessels were circumferentially pre
stressed going from top to bottom. In addition, new stressteel rods
were purchased for the longitudinal prestressing. The vessel was
tested successfully on the first attempt, reaching an ultimate pressure
of 2400 psi. A new deflection gage system was designed for this vessel.
The head dial gages were mounted to the side of the vessel out of
the direct line of the explosive failure path. Metal tabs were glued
to the specimen and piano wire was threaded into these tabs and tied
off. The wire was then strung over ball bearing pulleys and across the
head to the spring loaded gages on the side. This system was effective
in protecting the gages from damage but the readings taken from these
gages indicated a great amount of friction was present in the system.
Thus, a new system was adopted for the next vessel.
156
B.3 Test Vessel PV28 (Solid, l2~-in. head)
The deflection gage system used for all remaining vessels was
first used for PV28. The gages were mounted over the vessel, directly
above the metal tabs glued to the surface. Piano wire was tied to
the tab, stretched taut and attached to the spring loaded gages. Steel
channels were run across the top of the vessel between the piano wires
and bolted to the prestress rods. This system gave good readings and
at the same time contained the explosion and protected the gages. An
internal pressure of 3170 psi was reached during the first test of
PV28 before the leaks in the vessel could not be outrun by the pump.
The vessel was relined and tested successfully three weeks later to
an ultimate pressure of 3765 psi.
B.4 Test Vessel PV29 (Thirty-seven 2-in. penetrations, 10-in. head)
Fabrication and testing procedures had become well established
with the successful testing of three vessels. No major changes were
made for PV29. It was cast and tested to failure in a relatively
short period of time. No concrete strain gages were applied to the
vessel.
B.5 Test Vessels PV30, PV3l (Thirty-seven 2-in. penetrations, l2~-in. head)
PV30 and PV3l were nominally the same in size and penetration
pattern. However, there were several important differences in their
physical properties and behaviors. The actual head thickness of PV30
was found to be l2.22-in. while PV3l had a thickness of 12.02-in. The
compressive strength of the concrete head was 6350 psi for PV30 and
4950 psi for PV31. PV30 failed on the first attempt at 3210 psi in a
157
symmetric manner. PV3l developed a leak in the liner on the first test
and reached an internal pressure of only 1200 psi. Upon retest PV31
failed at 2800 psi in a somewhat unsymmetric shear failure.
B.6 Test Vessels PV32, PV33 (Six 5-in. penetrations, l2~-in. head)
PV32 and PV33 were also designed and tested as a check against
each other. Their properties were in closer agreement than were PV30
and PV31 and the results reflect this similarity. PV32 and PV33 had
concrete compressive strengths of 5720 psi and 4875 psi respectively.
The average head thickness of PV33 was l2.45-in. as compared to 12.30-
in. for PV32. These two factors of concrete strength and slab thick
ness appear to have offset one another. PV32 failed at 3075 psi
while PV33, after an aborted first test due to leakage, failed at
3100 psi. After PV33 developed a leak in its first test at 1700 psi,
a new type of copper liner was used for the second test. A copper
can was fabricated to fit into the steel liner grouted into the vessel.
The use of a copper can provided a better fit in the reentrant corners
and thus reduced the amount of expansion that high, pressures would pro
duce. PV33 was retested with this new liner and was pressurized to
failure without a single leak.
158
TABLE It 1
Chronology
Mark Casting Circumferential Longitudinal Testing Prestressing Prestressing
PV26.1 1-29-75 2-17-75 3-26-75 4-16-75
PV26.2 5-2-75 5-8-75
PV27 3-13-75 4-9-75 7-17-75 7-25-75
PV28.1 6-5-75 7-31-75 8-27-75 9-4-75
PV28.2 9-24-75 9-30-75
PV29 9-15-75 10-14-75 10-30-75 11-6-75
PV30 10-20-75 11-19-7 5 1-13-76 1-21-76
PV31.1 12-9-75 1-8-76 2-11-76 2-17-76
PV31.2 2-27-76 3-2-76
PV32 12-17-75 3-9-76 4-7-76 4-9-76
PV33.1 2-20-76 3-25-76 5-6-76 5-11-76
PV33.2 5-20~76 5-25-76