Effect of Manganese and Type of Current on the Properties and

Effect of Manganese and Type of Current on the Properties and Microstructure of All-Weld-Metal Deposited with E7016-1

Electrodes

With DCEP, weld metal toughness deteriorates above 1.45% Mn; with AC, up to 1.6% Mn is beneficial to weld metal toughness.

BY E. SURIAN AND T. BONISZEWSKI

ABSTRACT. Basic low-hydrogen all-position electrodes were used to study the effects of manganese content variation and the type of current (DCEP or AC) on chemical composition, microstructure and mechanical properties of all-weld-metal in the standard AWS test. Six electrodes were made to cover the Mn range of about 1.4-1.8%. With a given electrode on AC, the Mn content was about 0.08% lower, and the nitrogen content was about 40 ppm higher than with electrode DCEP. In the as-deposited weld metal, the acicular ferrite content increased with Mn content in the range of 1.45-1.7%. Vickers hardness, the yield stress and the tensile strength increased with the Mn content, but scatter in this relationship was caused by nitrogen, which raised the hardness and strength. At the same time nitrogen appeared to be detrimental to elongation. Along the centerline of all-weld-metal where the Charpy notch was located, nearly all the microstructure was of reheated type. In the DCEP weld metal, progressive deterioration in toughness occurred above 1.45% Mn, but in the AC weld metal up to 1.6% Mn was beneficial to toughness. At any Mn-level and by a number of different criteria, the AC weld metal was less tough than the DCEP weld metal, with the average AC transition curves lying usually below the DCEP curves plotted for the minimum results of the scatter band. The superior toughness of

f. SURIAN is Chief of the Development Dept., CONARCO Alambres y Soldaduras S.A., Buenos Aires, Argentina. T. BONISZEWSKI is a Welding Consultant based in the United Kingdom.

the DCEP weld metal was associated with the N content of just above 100 ppm, while the AC weld metal had N > 150 ppm. With both types of current, the extra-stringent classification of E7016-1 could be readily achieved with the electrode formulation at all the Mn levels tested.

Introduction

Historical Background

Since the late 1970s, a number of studies have been undertaken to find out the optimum Mn content of maximum toughness in C-Mn steel weld metal deposited from all-position basic low-hydrogen electrodes. In Europe, the impetus for those studies came from the requirements for the construction of offshore platforms for the North Sea. This quest for maximum toughness has been coupled with micro-structural studies which showed that in the as-deposited regions, acicular ferrite is the desirable constituent, whereas poly-

KEY W O R D S

Manganese Content AC vs. DCEP Current Microstructure E7016-1 Electrodes Chemical Composition Mechanical Properties Weld Metal Yield Stress Tensile Strength Charpy-V Toughness

gonal ferrite, grain boundary ferrite and various aligned ferrite plates (resembling bainite) have to be avoided.

In 1977, Evans (Ref. 1) showed that when using AWS E7018 electrodes on DCEP, at four nominal Mn-levels of 0.65, 1.0, 1.4 and 1.8%, in the ISO all-weld-metal test with three passes per layer welded with 1 k j /mm heat input, the optimum Mn-content for maximum toughness was about 1.5% in the as-welded and postweld heat treated (PWHT 580°C/2 h) conditions. Although the volume fraction of the acicular ferrite continued to increase above 1.5% Mn, up to 1.8% Mn, the beneficial effect of this microstructure on toughness appeared to be counteracted by the increasing strength.

Further work by Evans reported over 1978-80 for the same series of four electrodes used on DCEP throughout showed the following findings for the all-weld-metal deposits.

1) Whereas in the ISO test with three passes per layer and 1 k j /mm heat input 1.5% Mn was the optimum value for maximum toughness, in the AWS A5.1-69 test with two passes per layer and about 2 k j /mm heat input 1.3% Mn was the optimum value for the maximum toughness (Ref. 2).

2) With heat inputs of 0.6, 1.0, 2.2 and 4.3 k j /mm, the respective optimum Mn-contents for maximum toughness appeared to be: 1.15, 1.5, 1.3 and 1.25%, with Evans claiming that the overall optimum Mn content remained at approximately 1.4% (Ref. 3).

3) A change from the flat position (with the two heat inputs of 2.2 and 4.3 kj/mm) to the uphill position (with the two heat

348-s I SEPTEMBER 1992

inputs of 3.4 and 3.9 kj/mm) did not appear to make much difference to the optimum Mn-content, which remained at about 1.4% (Ref. 4).

At first sight, some of those findings appear incongruous because, according to some studies (Refs. 5, 6), one would expect that with increasing heat input and hence decreasing cooling rate, more and more Mn would be required for micro-structural refinement leading to maximum toughness. That this could be so is evident in 2 above where for 0.6 and 1.0 kj /mm, the optimum Mn contents are 1.15 and 1.5%, respectively. The apparent confusion arises from the fact that the changes in the heat input and welding position have led to the Charpy notch sampling completely different macro- and micro-structural regions:

1) With 3 or more passes per layer, both the as-deposited and reheated regions are sampled. Therefore, the effect of the heat input (cooling rate) is congruous: at 0.6 k j /mm less Mn (1.15%) is required to develop optimum conditions for maximum toughness than at 1.0 kj /mm (1.5% Mn).

2) At heat inputs from 2 k)/mm (AWS test) upward, mainly, if not only, the reheated regions are sampled. Here, the acicular ferrite plays no role, and for such spuriously 'normalized' weld metal there appears to be one optimum Mn value of about 1.4%, as for the properly normalized weld metal (Ref. 7).

Therefore, there is currently a consensus in industry that the maximum toughness is achieved with the E7018(—1) electrode at the optimum composition of about 1.5% Mn-0.3% Si in the weld metal.

However, the pressures of technological requirements from the North Sea offshore industry increased the interest in the AWS E7016 Class electrodes (Refs. 8,9). In contrast to E7018, the electrodes of E7016 Classification have the following characteristics and usability advantages:

1) Unlike E7018, the E7016 electrodes have little or no Fe-powder in their coatings, and hence there is more space in them for CaCC>3 which gives improved gaseous protection and decreased nitrogen pickup.

2) Consequently, E7016 electrodes can be eminently suitable for the deposition of sound (porosity-free) root passes, especially in very deep V-groove welds in plate thicknesses of 50-100 mm (2-4 in). Also, sound root passes can be deposited in single-sided butt joint welds where there is no access for back gouging or grinding.

3) With no Fe-powder in the covering of E7016 electrodes, good operability is obtained with thinner coatings than those required for good operability of E7018 electrodes. Therefore, with E7016 electrodes, the arc and molten pool visability is improved, especially in deep V-joints. In

single-sided welding, especially where large gaps have to be bridged and in all-position welding with difficult access, E7016 electrodes give weld pools much easier to control than those produced with E7018 electrodes.

4) Other factors being equal, as the Fe-powder is decreased and eliminated from the coatings, the weld metal oxygen content decreases from about 350-400 ppm (E7018) to about 300-250 ppm (E7016), and this is associated with progressively improving toughness (Ref. 10). Consequently, where good CTOD toughness is required in plain C-Mn weld metal (without 1% Ni), fabricators would normally use the E7016 and not the E7018 electrodes.

Because of this interest in E7016 electrodes, Taylor (Ref. 11) examined the effect of the four Mn levels at 1.2, 1.5, 1.75 and 2.0% on the toughness and micro-structure of their weld metal. However, instead of using DCEP as Evans (Refs. 1-4) did, Taylor (Ref. 11) used his E7016 electrodes on AC. AC is often preferred when welding large steel structures, such as offshore platforms, because the magnetic arc blow is avoided with AC. For this reason, industry prefers E7016 electrodes over E7015 ones which could also have all the advantageous features of E7016 electrodes, but which are usable on DCEP only.

Taylor (Ref. 11) tested the standard AWS all-weld-metal deposited at 2 k j /mm and single-V butt joint welds made uphill (ASME 3G) with the four heat inputs of 2.5, 3, 3.6 and 5 kj /mm. Both test types showed that the optimum Mn-content for

maximum toughness was about 1.7%, especially for uphill welding with 3 kj /mm, which is a very common practical condition. As expected, the acicular ferrite increased and the grain boundary ferrite decreased with the increasing Mn-content.

Currently, high-quality E7016 electrodes supplied to the North Sea offshore industry and claimed to be suitable for the CTOD testing have their weld metal composition controlled at 1.6% Mn and 0.4% Si. Especially, the Mn-content in E7016 electrodes appears to be marginally higher than that established by Evans (Refs. 1-4) for the maximum toughness in E7018 weld metal. This might be needed to compensate for the decreased Mn-recovery on AC which was shown by Morris (Ref. 12) to be about 0.07% lower than that on DCEP.

The Purpose of This Work

Because of the apparent discrepancy between the findings of Evans (Refs. 1-4) for E7018 electrodes and those of Taylor (Ref. 11) for E7016 electrodes, it was decided to carry out further work on the E7016 electrodes for which the data are rather scant at present.

The objectives of this work were to study:

1) The effect of the Mn content in the narrow range of 1.4-1.8% using six electrodes, as opposed to the four electrodes used by Evans over the wide range of 0.6-1.8% Mn.

2) The effect of two different current types, DCEP and AC, applied to each of those six E7016 electrodes.

Table 1 — Formulation of E7016 Electrodes Used to Study the Effects of the Mn-Content and Type of Current on All-Weld-Metal Toughness

Ingredients in Dry Powder

Hoganas Fe-powder ASC 100.29 Electrolytic Mn Stabilized FeSi45 Mg-powder Total metals

White pigment-grade Ti02

Potassium titanate K2TiC>3 Zirconium silicate ZrSiO,} Limestone CaCC>3 Calcium fluoride CaF2 Extrusion aids

Core Wire: standard rimming steel

Covering Parameters Extrusion die for 4-mm rod: 6.30 ± 0.04 mm Linear coating factor FL = 1.58

D 2 - d 2

Volumetric coating factor Fv = "

Contents for 4-mm rod,

varied 2.7-3.0* varied 4.3-5.8* varied 9.0-7.2*

2.5 18.5

7.0 3.0 4.0

50.5 16.0 1.0

100.0 31.5

d2 1.48

NOTE: A small amount of Fe-powder was used as ballast to accommodate variations in the Mn and FeSi45 powder contents. To maintain an approximately constant Si recovery in the weld metal, the FeSi45 powder was decreased with the increasing Mn powder

W E L D I N G RESEARCH SUPPLEMENT I 349-s

45°

Fig. 1 —All-weld-metal test weld preparation and the sequence and number of weld passes deposited with 4-mm-diameter electrodes. Approximate heat input —2.0-2.1 kj/mm; plate width on each side —127 mm; total plate/weld length —500 mm; interpass temperature — 710- W0cC.

Experimental Design

The Electrodes

Six electrodes were produced on 4-mm-diameter core wire according to the formulation given in Table 1. Manganese was varied in the covering, using electrolytic Mn-powder in the range 4.3-5.8%. To maintain approximately constant Si recovery in the weld metal, the ferrosilicon had to be decreased with increasing Mn. The linear coating factor was 1.58, as opposed to 1.70 for Evans's E7018 electrodes (Ref. 1).

Welding Conditions

The weld preparation used is shown in Fig. 1, which corresponds with the current AWS A5.1-81 standard, and the welding conditions for the two current types are given in Table 2. All welding was done in the flat position. It should be noted that the 'natural' arc voltage was slightly higher on AC than on DCEP. As with the AWS test used by Evans (Ref. 2), the nominal heat input was about 2 k j /mm.

Testing Conditions

The tensile test specimens of Minitrac type were dehydrogenated prior to test

ing at 250°C for 16 h. The Charpy test specimens had their notches normal to the plate surface and positioned on the weld centerline, where the two passes meet in each weld layer—Fig. 1.

Results and Their Analysis

Electrode Operability Characteristics

The electrodes operated satisfactorily, with the arc being smoother on AC than on DCEP. The slag covered the weld pool uniformly and it cracked on cooling, thus being easy to remove. The deslagging was better with AC than with DCEP. This could be associated with smoother bead appearance for AC than for DCEP where some coarse ripples could offer keying points for the slag. However, the total number of ripples per unit length was similar at about 20 per cm, for both current types.

Chemical Composition of the Deposits

Table 3 gives the representative compositions of the weld metal deposits. There were some variations between the analyses obtained from pads and the all-weld-metal test samples. These variations were not systematic, and the values in Table 3 are the advantages of all the analysis values available.

As expected, the Mn content in the weld metal increased with the progressive additions of Mn powder to the electrode coverings —Fig. 2. With Electrode 3041 (5.5% Mn powder), there was hesitation in this trend with DCEP, but the change was monotonic with AC. On average, the Mn-recovery curve for DCEP in Fig. 2 lies about 0.08% above the curve for AC, and this is in good agreement with the difference of 0.07% Mn reported by Morris (Ref. 12).

In general, the Si-content was maintained at a steady level of about 0.3%, but for the last electrode, No.3042, it was slightly lower. There was not much effect from the type of current used on the Si recovery.

The oxygen contents (Table 3) did not appear to vary systematically with either the Mn content or the type of current. The crossover plots in Fig. 3 suggest that the oxygen contents remained at a constant level between 250 and 280 ppm. According to the data reviewed by Ahlblom (Ref. 13), such an oxygen level is at the optimum for the achievement of best toughness in C-Mn weld metal.

Despite some scatter, the nitrogen content was systematically and consistently higher with AC than with DCEP—see Table 3 and Fig. 3. Ignoring any general drift in the N contents with the Mn additions (Fig. 3), and averaging the N contents over the Mn range, the N content was about 40 ppm higher in the AC weld metal than in the DCEP weld metal. This trend is in agreement with the results of Abson (Ref. 14) who showed that, at a given arc voltage, the N content of the weld metal is much higher with AC than with DCEP. For E7016 electrodes at 24 V, Abson found the difference to be in excess of 150 ppm. Abson's analyses were done in the root regions of double-V welds where the arc would be exposed to air from both sides, unlike in a weld with the backing bar —Fig. I. The higher N content of the AC weld metal is probably a result of the diminished protection of the arc environment from air when the alternating current passes through zero. At that instant, the decom-

Table 2 -

Current Type

DCEP AC

- Welding Conditions for 4-mm

Arc Voltage, V

24 25

Current, A

170 170

-Diameter Electrodes

Travel Speed, mm/s

2 2

Heat Input, k)/mm

2.04 2.13

Interpass Temperature

110-180°C 110-180°C


Table 3 —

No.

3037 3038 3039 3040 3041 3042

3037 3038 3039 3040 3041 3042

Chemical

coating

4.3 % 4.6 % 4.9 % 5.2 % 5.5 % 5.8 %

4.3 % 4.6 % 4.9 % 5.2 % 5.5 % 5.8 %

Analyses of the Weld Metal Deposits

Type

DCEP DCEP DCEP DCEP DCEP DCEP

AC AC AC AC AC AC

Deposit No.

960 961 962 963 964 965

966 967 968 969 970 971

C

0.07 0.07 0.08 0.07 0.07 0.08

0.08 0.08 0.08 0.08 0.08 0.08

Mn

1.45 1.55 1.65 1.72 1.70 1.80

1.36 1.45 1.58 1.62 1.64 1.73

Si

0.30 0.31 0.31 0.30 0.33 0.24

0.31 0.31 0.33 0.32 0.33 0.27

Weld Metal Analysis

S

0.007 0.007 0.007 0.007 0.007 0.008

0.008 0.008 0.009 0.009 0.009 0.008

P

0.014 0.014 0.014 0.014 0.014 0.014

0.013 0.013 0.013 0.013 0.013 0.013

Cr

0.04 0.04 0.04 0.04 0.04 0.04

0.04 0.04 0.04 0.04 0.04 0.04

wt-%

Ni

0.03 0.03 0.03 0.03 0.03 0.03

0.03 0.03 0.03 0.03 0.03 0.03

Mo

0.015 <0.01 <0.01 <0.01 <0.01 <0.01

0.01 <0.01 <0.01 <0.01 <0.01 <0.01

Ti

0.01 0.01 0.01 0.01 0.01 0.01

0.02 0.015 0.01 0.01 0.01 0.01

Sn

0.01 0.01 0.01 0.01 0.01 0.01

0.01 0.01 0.015 0.01 0.01 0.01

ppm

O

272 281 250 270 255 253

266 240 259 249 250 264

N

128 126 118 138 106 109

191 161 153 148 152 172

NOTE: Co, Nb, V, W , As and Al were all below 0.01"... These results are averages from weld metal pads and all-weld-metal specimens (metallographic and mechanical). Oxygen and nitrogen determined in tensile pieces.

position of CaC03 might be retarded and there would be less CO2 available to displace air from the arc environment.

Metallographic Examination

Columnar/As-Deposited Microstructure. This was examined in the final passes— i.e., No. 14, as indicated in Fig. 1 —after the usual etching in Nital. It was noted that there was a marked difference be

tween the widths of the columnar grains (i.e., the prior austenitic grains) in the deposits made with the two current types. The measured values are given in Table 4 and are also plotted in Fig. 4 against the Mn-content of the weld metal. At 1.3-1.5% Mn, the columnar grain widths were approaching 200 pm or the DCEP deposits, but they were below 100 pm for the AC deposits. With the increasing Mn content to 1.8%, these widths decreased to

150 pm and below for the DCEP deposits, but remained generally unchanged for the AC deposits.

It is believed that this is the first time that a difference in the columnar grain widths has been observed to be associated with the type of current used for the deposition of C-Mn weld metal from basic low-hydrogen electrodes.

The narrower grain size with the AC and the narrowing of the grain size with

1.9

1.8 _

1.7

1.6

1.5 -

i.k _

1.3

1.2

300

250 _

200

150

100

50

DC+ OXYGEN

- - . . — - S ^

® DC+1

• AC 5 0 DC+ 1

O AC J

0XYC-EN

NITROGEN

' . . 0

I I I I 1 I ' ' 5.0

1 j 1 1 r

5.5 6.0

Mn-POWDER IN ELECTRODE COVERINGS,

k.o

Mn-POWDER IN ELECTRODE COVERINGS, %

Fig. 2 — Variation in the Mn content recovered in the weld metal with the Mn powder additions to electrode coatings plotted for DCEP and AC deposits. Note: On average, the Mn recovery in the DCEP weld metal was higher by about 0.08% than in the AC weld metal.

Fig. 3 —Plots of oxygen and nitrogen contents in the weld metal against the Mn powder additions to the electrode coatings, for both the DCEP and AC deposits.


Table 4 — Average Widths of Columnar Grains (Prior Austenitic Grains) in As-Deposited Weld Metal of Final Capping Passes

Electrode No.

3037 3038 3039 3040 3041 3042

Mn Powder

in Coating

4.3 % 4.6 % 4.9 % 5.2 % 5.5 % 5.8 %

Electrode on

Deposit No.

960 961 962 963 964 965

Weld Metal

Mn, %

1.45 1.55 1.65 1.72 1.70 1.80

DCEP

Columnar grain

width, Mm

185 200 187 168 146 123

Deposit No.

966 967 968 969 970 971

Electrode on

Weld Metal Mn, %

1.36 1.45 1.58 1.62 1.64 1.73

AC

Columnar grain

width. j j m

81 66.5

100 81

104.5 76

the Mn content on DCEP (Fig. 4) could be the result of both the increased N content (see Table 3) and the Mn affecting the phase formation during and immediately after the solidification. Mild steel weld metal solidifies normally with delta ferrite being the primary phase and, on further cooling, austenite nucleates at the delta grain boundaries and transforms the original delta ferrite, to form the columnar grains resembling the original solidification structure. As both Mn and N are austenite stabilizing elements, it is possible that with

their increasing contents, accompanied by the inherent inhomogeneity of the weld metal, the number of nucleation sites increases, thus leading to some grain refinement. Also, it is possible that, with the increasing N and Mn contents, in some places solidification with austenite as a primary phase may occur.

The microstructural constituents in the columnar structure have been determined quantitatively (Ref. 15) and their volume fractions (in %) are given in Table 5. As it would be expected for the Mn-levels

200 —

150

100

50

1.70

Mn-CONTENT IN WELD METAL, % Fig. 4 —Widths of columnar grains in the as-deposited regions plotted against the Mn content of the weld metal.


studied, the predominant microstructure was acicular ferrite. There was a minor difference between the DCEP and AC deposits concerning the presence of polygonal ferrite (PF) which was noticeable in the AC weld metal. An example of the change in the appearance of the micro-structure with increasing Mn content is given in Fig. 5 (A-D): polygonal ferrite and ferrite side-plates shrunk progressively (Fig. 5A-5C) giving way to acicular ferrite. At the highest Mn-content (1.73%), the side plates were replaced by ferrite aligned with other phases (Fig. 5D), which is a bainitic type of microstructure associated with increased hardenability (Refs. 5,6).

The data from Table 5 are plotted in Fig. 6 to highlight the trends:

1) For each current type, in the central portion of theMn content of 1.45-1.65%, the content of the acicular ferrite rose monotonically with the Mn content. With DCEP, the increase in the acicular ferrite content was at the expense of grain boundary ferrite.

2) At the upper end (1.7-1.8%) of the Mn range, the content of acicular ferrite decreased, having passed through a maximum. This is a new observation because in the previous work (Ref. 1), the acicular ferrite content had been shown to increase monotonically up to 1.8% Mn. The decrease in the acicular ferrite content at 1.7-1.8% Mn in the present work is associated with the increased formation of the aligned ferrite (AC) and grain boundary ferrite (GF)—see Table 5 and Fig. 5D.

3) The anomalies in the trends of Fig. 5 are further highlighted in Fig. 7. The AC weld metal at the lowest Mn content of 1.36% showed an unusually high acicular ferrite content of 62% (Table 5). This would appear to be associated with a very high N content (191 ppm) in this weld metal as shown in Table 3 (Deposit No. 996). Nitrogen is an austenite stabilizer and it would promote hardenability, as Mn and Ni do. It is also possible that the apparent peak of 72% acicular ferrite in the DCEP deposit, No. 963, is also associated with its highest N content within the DCEP weld series —see Table 3.

It has been noticed that in the AC weld metal, the acicular ferrite was generally finer and more acicular than in the DCEP weld metal. Figure 8 shows the examples for two electrodes:

1) For DCEP, the acicular ferrite became markedly finer when the Mn content increased from 1.55 to 1.72%.

2) For AC, even at 1.45% Mn, the acicular ferrite was finer than for DCEP at 1.55% Mn, and therefore further refining was not conspicuous when the Mn content increased to 1.62%.

The finer morphology of acicular ferrite in the AC weld metal is likely to be associated with its increased N content (Table

3), which would promote the hardenability of austenite.

Reheated Microstructure. Within the all-weld-metal areas reheated above the AC3 temperature by subsequent passes, there are two regions: 1) a fine-grained region adjacent to and just above the AC3 isotherm; and 2) a coarse-grained region, well above the AC3 temperature and adjacent to the solidus isotherm.

In the present work, the average grain diameters were measured for the finegrained regions only, and the results are given in Table 6. These show that over the Mn range of 1.4-1.8%, the grain intercept remained constant at about 7 pm regardless of the current type used. This is at some variance with the results of Evans (Ref. 1) who reported that the average grain diameter (the mean linear intercept) decreased as 6.7-5.8-4.9-4.1 pm series for the increasing Mn contents of 0.66-1.0-1.42-1.82%.

The central region of all-weld-metal: the Charpy notch locum. Along the vertical weld centerline position where the two passes meet in the layers, as shown in Fig. 1, the distributions of the different micro-structural regions have been determined and are shown in Figs. 9 and 10 for the DCEP and AC weld metals, respectively. It can be seen that the columnar structure was primarily in the final pass, which stood above the plate surface. Within the plate, the metal was mainly of the reheated type, as would be expected from Fig. 1.

In Figs. 9 and 10, the central 10-mm depth of the plate covered by the Charpy-V notch is indicated. The quantities of the different microstructural regions covered by the notch are given in Table 7. It can be seen that the notch sampled predominantly the reheated weld metal, which was in the range of 85-100%. Therefore, the acicular ferrite and other microconstituents of the as-deposited columnar regions would not have played any significant part in determining the Charpy toughness in this work.

Mechanical Properties

Vickers Hardness. The measurements at 10-kg load were carried out along the weld centerline as shown at the top of Table 8, which gives the results obtained. The maxima and minima from Table 8 are plotted against the weld metal Mn contents in Fig. 11. In each case, at a given Mn-level, the hardness of the AC weld metal is systematically higher than that of the DCEP weld metal. The higher hardness of the AC weld metal is associated with its higher N content (Table 3), which is an important hardener of the ferritic weld metal (Refs. 16-18).

There is a general tendency for the weld metal hardness to increase with the Mn content—Fig. 11. There is some fluctuation in the results plotted, and this flue-

i- f i t /

M J S ^ ? £ ' 5 » ; ? J

^ . . • : • •:•>•• . * ,

Fig. 5 — Typical microstructures within the columnar as-deposited final passes of the A C weld metal. A-Mn = 1.45%; B-Mn = 1.58%; C-Mn = 7.64%; D-Mn = 1.73% (100X).

Table 5 — Volume Fractions of Microstructural Constituents in Columnar Grain Structure of As-Deposited Weld Metal in Final Passes

Electrode No.

3037 3038 3039 3040

4 I 3042

Mn in Coating

4.3% 4.6% 4.9% 5.2% 5.5% 5.8%

Deposit No.

960 961 962 963 964 965

Electrode on

Mn, Ol

1.45 1.55 1.65 1.72 1.70 1.80

DCEP

Microstructural components

Al-

43 53 59 72' 65 67

AC

4 3 5 1 3 2

O/

GF

53 44 36 27 32 SI

Deposit No.

966 967 968 969 970 971

Electrode

Mn, %

1.36 1.45 1.58 1.62 1.64 1.73

on AC

Microstructural components, %

AF

621

46 52 62 67 50

AC

6 3 4 8 2

11

PF GF

5 27 3 48 3 41 3 27 3 28 4 35

Legend: AF —acicular ferrite; GF —grain boundary ferrite (veins); PF—polygonal (blocky) ferrite; AC —aligned ferrite wi th other phases; carbides, austenite and martensite. 1. Extra high N contents in their respective series; see Fig. 3.

Table 6 — Mean Grain Diameters in Fine-Grained Regions of Reheated All-Weld-Metal Areas

Electrode No.

3037 3038 3039 3040 3041 3042

Mn Powder

in Coating

4.3% 4.6% 4.9% 5.2% 5.5% 5.8%

Depos No.

960 961 962 963 964 965

Electrode on

Weld t Metal,

Mn, %

1.45 1.55 1.65 1.72 1.70 1.80

DCEP

Mean Grain

Intercept, pm

7.2 7.6 7.5 7.3 7.1 7.1

Deposit No.

966 967 968 969 970 971

Electrode on

Weld Metal Mn, %

1.36 1.45 1.58 1.62 1.64 1.73

AC

Mean Grain

Intercept, pm

7.0 7.3 7.1 6.9 7.0 7.3

WELDING RESEARCH SUPPLEMENT I 353-s

Table 7 —

Electrode No.

3037 3038 3039 3040 3041 3042

Contents of Different Microstructural

Deposit No.

960 961 962 963 964 965

Electrode on

Regions and Constituents along Charpy-V Notch Front

DCEP

Microstructural Zones along the Charpy Notch Front

Weld Metal

0 9

10 5

13 7

Total

100 91 90 95 87 93

Reheated Weld Metal

Coarse Grain

27 24 27 23 26 14

Fine Grain

73 67 63 72 61 79

Deposit No.

966 967 968 969 970 971

in the All-Weld-Metal Test

Electrode on AC

Microstructural Zones along the Charpy Notch Front

Reheated Weld Meta Columnar

Weld Metal

8 15 0 0

15 0

Total

92 85

100 100 85

100

Coarse Grain

47 24 41 38 31 56

Fine Grain

45 61 59 62 54 44

tuation can be accounted for by some variation in the Si and N contents in each weld metal series.

Tensile Properties. The results for the individual deposits are given in Table 9. It is quite clear that at a given Mn level, e.g., 1.45 or about 1.7%, the AC weld metal is much stronger than the DCEP weld metal. This is in accord with their difference in hardness (Table 8 and Fig. 11), with the AC weld metal being the harder one. As with the hardness, the raised tensile properties of the AC weld metal must be associated with its higher N content—see Table 3.

The yield stress and the tensile strength

of the two weld metal types are plotted against their Mn contents in Fig. 12, showing a gradual increase in the strength parameters with Mn. The trends for the DCEP weld metal are reasonably consistent: both the yield stress and the tensile strength rise with the increasing Mn-content, although there may seem to be a dip in strength at the highest Mn level of 1.80%. For the AC weld metal, there is much scatter at above 1.6% Mn. This is clearly associated with the lower N content of the two deposits in comparison with the remainder in the series. The strength parameters of the AC weld metal

Table 8 — Vickers Hardness (HV 10) Measured along the Centerline of All-Weld-Metal Deposits on Transverse Macrosections

Electrode No.

3037

3038

3039

3040

3041

3042

Deposit No.

960

961

962

963

964

965

Electrode on

Weld Metal Mn, %

1.45

1.55

1.65

1.72

1.70

1.80

DCEP

HV10 avg.

max-min

179 218-160

180 223-166

185 230-164

191 240-177

192 242-176

196 248-175

Deposit No.

966

967

968

969

970

971

Electrode on

Weld Metal Mn, %

1.36

1.45

1.58

1.62

1.64

1.73

AC

HV10 avg.

max-min

187 224-170

190 236-172

197 243-180

197 243-178

205 248-182

210 250-186

with N at about 150 ppm appear to be close to those of the DCEP weld metal with N at about 120-140 ppm. The AC weld metal with N at 160-190 ppm is much stronger than the remainder, and the DCEP weld metal at 1.80% Mn has probably lost some strength because it has about 110 ppm N only and its Si content is on the low side (0.24%).

The elongations of the two weld metal types are plotted against their Mn content in Fig. 13, showing a trend of decreasing values with Mn. In the DCEP weld metal series, two groups are apparent: one where N = 120-140 ppm, and a second where N = 100-110 ppm, with the latter having higher elongation values at a given Mn content. The elongation of the AC weld metal shows much scatter when plotted against the Mn content, but the results cluster around those of the DCEP weld metal where N = 120-140 ppm. Within the AC weld metal series, again the effect of N can be perceived: at the lower N-contents (close to 150 ppm) the elongations tend to be higher within the overall plot against the Mn content.

Charpy Impact Toughness. The Charpy transition scatter bands for all the results, with their mean transition curves are plotted in Figs. 14(A-F) and 15(A-F) for the DCEP and AC all-weld-metals, respectively. It is interesting that, in general, the scatter bands for the DCEP weld metal are much wider than those for the AC weld metal. This could be because of the DCEP

Table 9 — Tensile Properties of All-Weld-Metals Deposited with Two Types of Current from E7016 4-mm-Diameter Electrodes with Different Mn Contents

Electrode No.

3037 3038 3039 3040 3041 3042

Deposit No.

960 961 962 963 964 965

Weld Metal Mn, %

1.45 1.55 1.65 1.72 1.70 1.80

Electrode on

Yield Stress

N/mm 2

443 481 498 520 500 508

DCEP

Tensile Strength, N/mm2

525 548 596 618 618 578

Elongation, o,

26.8 26.0 24.0 22.8 28.8 24.8

Deposit No.

966 967 968 969 970 971

Weld Metal Mn, %

1.36 1.45 1.58 1.62 1.64 1.73

Electrode on

Yield Stress

N/mm 2

498 520 535 478 460 548

AC

Tensile Strength N/mm 2

598 613 603 563 578 628

Elongation,

% 26.4 27.6 22.8 26.4 25.6 21.6


100

50

DC'ELECTRODE

—r 1.3

I ' l GF I

GRAIN |BOUNDARYI FERRITE I

AF ACICULAR FERRIJTE

1 FIELD I

I

~1 1.5

5 0 -

ACCURRENT

1.8 r~ 1.3

Fig-Mn

6 — Volume fractions of different microstructural components in the columnar as-deposited weld metal of the top passes plotted against the content in the weld metal.

weld metal being lower in N content; the individual results would be more sensitive to local inhomogeneity fluctuations.

The mean transition curves for each type of weld metal, with different Mn contents, are compared in Figs. 16 and 17, and similar transition curves, but for the lower bounds of the scatter bands (LBSB), are compared in Figs. 18 and 19. In each case, there is more spread in the position of the curves with the Mn content for the DCEP weld metal than for the AC weld metal. Evidently, the DCEP weld metal is more sensitive to the variation in the Mn content, for it contains less nitrogen. The AC weld metal, with its higher N content, is less sensitive to the variation in the Mn-content.

Using markers at —40°C (—40°F) in Figs. 16 and 17, and at - 2 0 ° C in Figs. 18 and 19, it can be seen that for the DCEP weld metal the curves are generally shifted toward the lower temperatures and higher energies, i.e., the DCEP weld metal is tougher than the AC weld metal.

To further compare the toughness of the two types of weld metal, the following parameters have been extracted from Figs. 16-19, tabulated in Tables 10 and 11 and plotted:

Figure 20 shows the change in the transition temperature for 100 J energy absorption as a function of Mn content in the weld metal. The values for both the mean Charpy curves and the lower bounds of the scatter bands are plotted. At any Mn level, the DCEP weld metal is always tougher than the AC weld metal, with the average AC data lying close to the minimum data of the DCEP weld metal. For the DCEP weld metal, the transition temperature rose progressively with the Mn content between 1.45 and 1.80%, showing no benefit from the increasing Mn where the microstructure is almost fully reheated (Fig. 9). This is in very good agreement

80

75 _

70 -

a>

I—" UJ

£ 65 _ o

L U J —

tc g 60 ec

<

5 5 5 -

50 _

45

A0

Hi -N o

/ Hi-N \ I # I . OUTLIER V /

I 1.3 1.4

I 1.5 1.6 1.7 1.8

Fig. 7 —Acicular ferrite content in columnar regions of the DCEP and AC deposits plotted together for the two current types against the Mn-content of the weld metal, with the special indication of the outlying points. Outlying points: "Hi-N" —high nitrogen content; "Ni-AC"—high quantity of the aligned ferrite constituent.


OJ 15 -

c m •fi 3

Top surface o H h e plate 1 - 5 5 M n

Electrode DC*

1.80Mn

Fig. 9 —Distribution of different types of microstructural regions along the vertical centerline in the all-weld-metal deposited with DCEP: black — columnar; hatched—reheated coarse-grained; white—reheated finegrained.

Fig. 10—Distribution of different types of microstructural regions along the vertical centerline in the all-weld-metal deposited with AC: black — columnar; hatched—reheated coarse-grained; white—reheated finegrained.


280

240

— --*

o--* °

— I — 1.4

I -

1.6 1.7

650 _

550

500 _

450

400

® DC+ 1 j TENSILE STRENGTH

• AC )

\ e DC+

o AC YIELD STRESS

TENSILE \ / / \ STRENGTH V /!/ \

Mn-CONTENT IN THE WELD METAL, *

F/g. 77 —Relationship between the all-weld-metal Mn content and the maxima, minima and averages in the Vickers hardness (HV10) measured along the vertical weld centerline.

1.3

Mn-CONTENT IN WELD METAL, %

Fig. 12 — The yield stress and the tensile trength of the two types of all-weld-metal plotted against the Mn content: for DCEP, the trends are reasonably consistent; for AC the strength parameters are markedly affected by the weld metal N content (see Table 3), with low values occurring at lower N contents.

1.3 1 .'t 1.5 1.6 1.7


Fig. 13 —The tensile elongation of all-weld-metals plotted against their Mn content: some scatter in the trend appears to be related to the N content of the weld metal. The two nitrogen levels are indicated for the DCEP weld metal only.

Table 10 — Transition Temperatures of 100 J Energy Absorption Extracted from the Mean and Lower Bound of the Scatter Band (LBSB) Curves of the Charpy-V Test Results

Electrode on DCEP Electrode on AC

Temperature Temperature Temperature Temperature Weld for 100 ) for 100 ) Weld for 100 ] for 100 )

Electrode Deposit Metal from mean from LBSB1 Deposit Metal from mean from LBSB1

No. No. Mn, % curve, °C curve, °C No. Mn, % curve, °C curve, °C

3037 3038 3039 3040 3041 3042

960 961 962 963 964 965

1.45 1.55 1.65 1.72 1.70 1.80

-59 -56 -53 -46 -55 -41

-53 -47 -37 -34 -37 -27

966 967 968 969 970 971

1.36 1.45 1.58 1.62 1.64 1.73

-38 -49 -42 -45 -39 -37

-35 -37 -38 -38 -26 -25

l. LBSB = the lower bound of the scatter band curve of the Charpy transition results


400

•o 3 0 0 -

.200-

a100-

P7 3037-960

1.45 Mn

1 — I — I — I — I — r 70 -60 -50 -40 -30 -20

400

S 300-

x >;200-

e-100-

400

300 -

2 0 0 -

100-3038-961

1.55 Mn

~\—i—i—i—r •70 - 6 0 - 5 0 - 4 0 -30 -20

Test temperature(°C )

400

3 0 0 -

200-

1 0 0 -

400

3 0 0 -

2 0 0 -

100-

i—i—r -70 -60 - 5 0 - 4 0 -30 -20

400

i — i — i — i — T -70 -60 -50 -40 -30 -20

"i i i i i r -90 -80 -70 -60 - 5 0 - 4 0 -30 -20

300-

2 0 0 -

1 0 0 -

0

Fig. 14-

- 9 0 - 8 0 - 7 0 - 6 0 - 5 0 - 4 0 - 3 0 - 2 0 - 1 0 0 Test temperature l°C )

-Charpy-V transition scatter bands and mean transition curves for all-weld-metal deposited with electrode DCEP.

10 20

Table 11 — Energies Absorbed at Temperatures of Curves of the Charpy-V Test Results

-29 and - 4 6 C Extracted from the Mean and Lower Bound of the Scatter Band (LBSB)

Electrode No.

3037 3038 3039 3040 3041 3042

Deposit No.

960 961 962 963 964 965

Weld Metal Mn, %

1.45 1.55 1.65 1.72 1.70 1.80

Electrc } d e on DCEP

Energy absorbed a

_29<

from mean curve,

1

358 302 316 230 258 144

Q- JO I)

from LBSB

curve, I

358 188 192 155 133 107

Temperature

- 4 6

from mean curve,

I

260 262 170 98 144 84

q--50 F)

from LBSB

curve, J

216 115 40 55 85 67

Deposit No.

966 967 968 969 970 971

Weld Metal Mn, %

1.36 1.45 1.58 1.62 1.64 1.73

Electrode

Energy

_29<

from mean curve,

I

150 186 155 168 135 123

q -

on AC

absorbed at Temperature

20°F)

from LBSB

curve, I

126 128 136 136 82 88

- 4 6

from mean curve,

I

65 108 80

102 72 70

q - -50°F)

from LBSB

curve, 1

54 64 68 73 30 53


400

x> OJ

X l

o in xi

> Ol i~ 01

c

>-CL L I rO

X

U

3 0 0 -

200-

100 -

3037-966

1.36Mn

I I I I I I I I 90 -70 -50 -30 -10 0

400

| 300-u o ui

X I rd

g;20o-c Hi

5M00 rrj

JZ o

3040-969

1.62Mn

I I I I I 1 I I J - 9 0 -70 -50 -30 -10 0

400

300-

2 0 0 -

1 0 0 -

I I I I - 9 0 -70 -50 -30 -10 0

Test temperature (°C)

400

3 0 0 -

2 0 0 -

100-

I I I I I I I I I - 9 0 -70 - 5 0 -30 -10 0


400

3 0 0 -

2 0 0 -

1 0 0 -

I T I I I I I I I I 90 - 7 0 - 5 0 -30 -10 0

400

3 0 0 -

2 0 0 -

100 -

I I I I I I I I I I I I 90 -70 - 5 0 -30 -10 0 20

Fig. 15—Charpy-V transition scatter bands and mean transition curves for all-weld-metal deposited on AC.

with Evans's (Ref. 2) results for the AWS tests for which the optimum Mn content is 1.3%. In such a test, the beneficial effect of Mn on increasing the acicular ferrite content (Fig. 7) is irrelevant. For the AC weld metal, on the other hand, an increase in the Mn content of up to 1.6% appeared to be beneficial to toughness, even in the almost fully reheated weld metal—Fig. 10.

Figure 21 shows the change in the Charpy energy absorbed at —29°C (—20°F) as a function of Mn content in the weld metal. At any Mn level, the DCEP weld metal is again tougher than the AC weld metal, with the average values for the AC weld metal lying consistently well below the minima for the DCEP weld metal. Again, the toughness of the DCEP weld metal deteriorates progressively with the Mn content from 1.45% upward, but for the AC weld metal, the rise in the Mn content up to 1.6% appears to be beneficial.

Figure 22 shows a similar picture but for

- 4 6 ° C (-50°F). Again, the trends are the same as described above.

Discussion

Basic low-hydrogen all-position electrodes not only must be tested on AC and DCEP for classification purposes (see Table 8 in AWS A5.1-81), but they are also used in fabrication with those two current types. As shown in this work (Fig. 2), this has important practical consequences for both the electrode manufacturer and user. With a given content of Mn powder in the electrode covering, different Mn contents will be recovered in the weld metal with the two different current types, with other welding conditions being approximately similar. The AC weld could be expected to contain about 0.08% Mn less than the DCEP weld metal.

This work shows that in the range of 1.45-1.7% (Figs. 6 and 7), the increasing Mn content promotes acicular ferrite formation in as-deposited microstructure, as

with E7018 electrodes (Ref. 1). At the same time, hardness, the yield stress and the tensile strength increase (Figs. 11 and 12), this being counterproductive to good toughness. However, bearing in mind the "out-of-position" use of E7016-1 electrodes, especially in the roots of deep V-butt joint welds where there may be little reheating and grain refinement, it is understandable why some manufacturers fix the all-weld-metal Mn content at nominally 1.6%.

This need for the "balanced" Mn-content, in view of the opposing and conflicting trends, is further enhanced by the effect of the current type on the nitrogen content of the weld metal. This work confirmed Abson's (Ref. 14) findings that, with AC, the weld metal N content is higher than with DCEP. With the present series of electrodes, the difference was 40 ppm on average, with the DCEP weld metals containing just over 100 ppm N and the AC weld metals containing usually over 150 ppm N.


400

E lec t rode DC + 1.45 Mn

i — i — i — i — i — r

-90 -80 -70 -60 -50 -40 -30 -20 -10


i—i—r

400

0 10 20

Fig. 76 —Mean Charpy-V transition curves of the DCEP weld metal with different Mn contents compared together. These curves were extracted from the plots of scatter bands.

S C O

'S X ! L.

o Ul

X I I I

QJ C OJ > Q. i_ IQ x U

100-

0

E lec t rode AC

- 9 0 - 8 0 -70 -60 -50-40 -30 -20 -10 0 10 20

Test tempera tu re PC)

Fig. 17 —Mean Charpy-V transition curves of AC weld metal with different Mn contents compared together. These curves were extracted from the plots of scatter bands.

400

S O O -

0) X3

o Ul

X ) r0

Qj C dj

r0 _C CJ

1 0 0 -

E lec t rodeDC*

-90

400

10 20 -70 -60 -50 -40 -30 -20 -10

Test temperature (°C I

Fig. 18 —The lower bound curves which delineate the minimum results of the Charpy scatter bands for the DCEP weld metal with different Mn

Elect rode AC

.58 Mn

~i i i i r~~i i i i — i — i — r - 9 0 -80 -70 -60 -50 -40 -30 -20 -10 0 10 20

Test tempera tu re (°C)

Fig. 19 —The lower bound curves which delineate the minimum results of the Charpy scatter bands of the AC weld metal with different Mn contents.


__

*~ 1_

o

-

p

Q

H j£

<

- 2 5 _

- 3 0 _

-35 J

- 4 0 _

-<i5 -

-50 _

-55 -

- 6 0 _

© _ _ _ ©

0 ^ AC

DC+ 1

AC )

l)C+ ",

1 AC

""~°--.

AC \

DC+

DC+

1 1

/ / / S

0 ^

from

f r o m

AVERAGE CURVES

THE LOWER BOUNDS OF THE SCATTER BANDS

* —» ^

/ / /

/ /

J2>—

0 — '

1 / I /

/ ! / /

i / , a s ,

o—o / ^ ^ 4**^ If

P / * nr

0 / ~~^^~^ e

1 1 1

40C

350

3 0 0

250 -

200

150

100

50

o a DOI ( f rom AVERAGE CURVES

• • AC j O 0 DC ) THE L

> from o OAC ) T H E s

OWER BOUNDS OF

E SCATTER BANDS

- 1 -

1.3 1 5 1.6 -~T~ 1.7

1.3 1.4 1.6 CONTENT IN WELD METAL, %


Fig. 20 — Charpy transition temperature for 100j energy absorption plotted against weld metal Mn content in both the DCEP and A C weld metal. Values taken from both the average curves and the lower bound of the scatter bands.

300

150 -

50

0 0 DC+) 1 from AVERAGE CURVES

• -• AC )

0 6 DC+"\

O O AC J

THE LOWER BOUNDS OF

THE SCATTER BANDS

I ' . 3

~~r 1.4

—1 1.5

I

Fig. 2 I—Charpy energy absorbed at —29°C (—20°F) plotted against weld metal Mn content of both the DCEP and AC weld metals. Values taken from both the average curves and the lower bound of the scatter bands.

30 -

= 25

3 .2 3.3

PARAMETER - Mn (%) + N (ppm) — S o

Fig. 23 —Variation in tensile elongation plotted against a parameter incorporating the Mn and N contents in the all-weld-metal deposited with electrode on DCEP.

6 1.7


1.8

Fig. 22—Charpy energy absorbed at -46°C (~50"F) plotted against weld metal Mn-content for both the DCEP and AC deposits. Values taken fromboth the average curves and the lower bound of the scatter bands.


There was a clear indication that the N content was conducive to increased Vickers hardness (Fig. 11) and the yield stress and tensile strength —Fig. 12. In the plots of those parameters against the Mn content, some scatter was observed within a given (e.g., AC) weld metal series. This scatter could be accounted for by the variation in the N content: the weld metal with lower N content within a series was weaker than what would be expected from the trend governed by the Mn content.

In general, the increasing weld metal strength was accompanied by decreasing elongation values. The plot of elongation against the Mn content shows much scatter (Fig. 13), but it appears that the data for all the weld metals (both DCEP and AC) fall into two families: 1) Two DCEP weld metals with N ~100 ppm stand out with relatively high elongations; and 2) All the other weld metals, i.e., the DCEP ones with N —120-140 ppm and the AC ones with N >150 ppm, form a broad scatter band.

Thus, broadly, both the increasing Mn content in the range of 1.3-1.8% (Fig. 13) and N-content are detrimental to elongation. It would be convenient to have a single parameter describing the effect of those two elements. A possibility of finding such a parameter (Mn + N/80) is shown in Fig. 23 for the DCEP weld metal only. So far, no single parameter embracing both weld metal types was found, and it could be that the type of current used leads to other changes which cause further metallurgical complications determining the elongation values.

In this work, conforming to the standard AWS test assembly, the Charpy-V notch sampled the central all-weld-metal region with nearly wholly reheated and much-refined microstructure—Figs. 9 and 10. Therefore, acicular ferrite could hardly play any role in determining the toughness. For electrode DCEP, toughness deteriorates progressively above 1.45% Mn (Fig. 20) and this is in agreement with Evans's (Ref. 2) work on E7018 electrodes, where in the AWS test, the optimum Mn-content was 1.3%. However, when E7016 electrode is used on AC, it would appear that even with nearly all-reheated micro-structure, the Mn content of up to 1.6% may be beneficial.

The degree of change in the position of the Charpy transition curves (Figs. 16-19) with the Mn content in the range 1.4-1.8% is very encouraging from a practical viewpoint. In manufacturing, there is always some natural Mn variation with respect to the target value in the formulation, and in use Mn recovery would vary as a result of welding conditions. As long as the electrodes are formulated to give all-weld with 1.45% Mn on AC and 1.55% Mn on DCEP, consistently acceptable qual

ity can be expected. There should be no problem in achieving the extra-stringent classification of E7016-1 with the average minimum of 27 J at - 4 6 ° C (see Figs. 16 and 17), and the individual minimum of 20 ) at - 4 6 ° C (see Figs. 18 and 19).

Some peculiar microstructural effects associated with high nitrogen content have been noted:

1) In the AC weld metal with the lowest Mn content of 1.36%, the N content happened to be highest at 191 ppm—see Table 3. This weld metal had an unusually high acicular ferrite content (Fig. 6), which was out of line with the usual trend expected.

2) With a given electrode (Fig. 8), the acicular ferrite itself appeared to be finer in the AC weld metal than in the DCEP weld metal, despite the former containing less Mn than the latter.

These observations are relatively scant, but they may be worthy of further inquiry. Being an austenite stabilizer, if nitrogen in solid solution affects the content and morphology of acicular ferrite, its effect on toughness may be more complicated than expected hitherto.

Conclusions

1) With a given Mn powder content in the electrode covering, the Mn recovery in the all-weld-metal was about 0.08% higher with electrode on DCEP than on AC, with other welding parameters being similar.

2) The type of current, DCEP or AC, had no significant effect on Si-recovery in the all-weld-metal.

3) Over the Mn range of 1.35-1.80%, with the two current types (DCEP and AC), the weld metal oxygen content remained approximately constant in the range of 240-280 ppm.

4) Over the range of 1.35-1.80% Mn in the weld metal, the N content fluctuated between 100 and 140 ppm for electrode DCEP and between 150 and 190 ppm with AC. On average, the N content was 40 ppm higher with AC than for electrode DCEP.

5) In the AC weld metal, the columnar grains of the as-deposited structure were not as wide as those in the DC+ weld metal, and their widths remained independent of the Mn-content. In the DC4- weld metal, the columnar grain width decreased with the Mn-content.

6) In the as-deposited structure, the acicular ferrite content increased over the 1.45-1.70% Mn range in the DCEP weld metal and over the 1.45-1.65% Mn range in the AC weld metal, primarily at the expense of the grain boundary ferrite.

7) Some anomalies in the acicular ferrite content vs. the Mn content trend have been observed for the weld metal with extra-high N contents, with the nitrogen

appearing to increase the acicular ferrite content.

8) There was some tendency for the acicular ferrite to be finer in the AC weld metal than in the DCEP weld metal.

9) The fine size in the reheated regions remained approximately constant at 7 pm over the Mn range studied in both the DCEP and AC weld metals.

10) In the all-weld-metal centerline region where the Charpy-V notch was positioned, the microstructure was predominantly (85-100%) of the reheated type.

11) Over the Mn range of 1.4-1.8%, the average hardness at the all-weld-metal centerline increased from about 180 HV10 to about 210 HV10, and, at a given Mn-level, the hardness of the AC weld metal was about 10-20 HV10 higher than that of the DCEP weld metal.

12) Broadly, the tensile strength parameters increased with the Mn-content, but a clear effect of the N-content could be perceived: 1) the AC weld metal was generally stronger than the DC weld metal where the former was much higher in nitrogen, i.e., by some 30-60 ppm; 2) the results for the AC weld metal with about 150 ppm N were close to those of the DCEP weld metal with 120-140 ppm; 3) the DCEP weld metal with only about 110 ppm appeared to lose strength within the general trend.

13) The elongation tended to decrease with the Mn content, but again the effect of N could be perceived, in the direction opposite to that for strength. At N = 100-110 ppm, the DCEP weld metal was more ductile than at N = 120-140 ppm. The AC results clustered around those for the DCEP welds with the higher N, but again the beneficial effect of the lower N values on elongation could be perceived.

14) Charpy transition plots of all the results gave wider scatter bands for the DCEP weld metal than for the AC weld metal.

15) Using a number of different criteria to compare the DCEP and AC weld metals [Le., temperature for 100 J, and energy absorption at - 2 9 ° and -46°C), the DCEP weld metal was always tougher at a given Mn content than the AC weld metal.

16) With the increasing Mn content from 1.45 to 1.80%, the toughness of the DCEP weld metal deteriorated progressively, but this was true only for the weld centerline region, where the microstructure was almost 100% of the reheated type.

17) For the AC weld metal, also with almost 100% of the reheated microstructure in the weld centerline, the maximum toughness was observed for the Mn range of 1.45-1.6%.

18) It would appear that with other factors being equal, for E7106-1 Class electrodes, a higher Mn-content is required on AC than with electrode DCEP to


achieve max imum toughness in the all-weld-meta l test.

19) For electrodes fo rmu la ted to deposit about 1.4% M n on A C and 1.55% M n on DCEP, it should be possible t o consistently achieve the toughness requirements of the extra-stringent E7016-1 Classification: 27 ) average m in imum and 20 J individual m in imum at — 46°C .

20) As in the A W S al l-weld-metal test, the centerl ine microstructure is largely reheated, the Charpy results f r o m such a test cannot be considered adequate for the determinat ion o f the o p t i m u m M n content for max imum toughness w h e r e as-deposited microstructural regions w o u l d be sampled by the no tch , e.g., as in the roo t regions of deep V-but t joint welds.

Acknowledgment

The authors express their thanks to Ing. Juan L. Trot t i for carrying out mechanical tests and metal lographic examinat ion.

References

1. Evans, G. M. 1977. Effect of manganese on the microstructure and properties of all-weld-metal deposits, IIW Doc.ll-A-432-77; also published in: Welding Journal 59(3):67s to 75-s; and Welding Research Abroad, 29(1), Welding Research Council, New York.

2. Evans, C. M. 1978. Comparison of ISO

2560 and AWS AS. 1-69. IIW Doc.ll-C-547-78; also published in OERLIKON Schweissmitteilun-gen, 85(4):4-10.

3. Evans, G. M. 1979. Effect of heat input on the microstructure and properties of C-Mn all— weld-metal deposits. IIW Doc.ll-A-490-79; also published in Welding lournal 61(4):125-s to 132-s.

4. Evans, G. M. 1980. Effect of welding position on the microstructure and properties of C-Mn all-weld-metal deposits. IIW Doc.ll-A-499-80; also published in Welding Research Abroad, Jan. 29(1):58-69, Welding Research Council, New York.

5. Harrison, P., and Farrar, R. 1987. Micro-structural development and toughness of C-Mn and C-Mn-Ni weld metals—Part 1, microstructural and development. Metal Construction 19(7):392R-399R.

6. Farrar, R., and Harrison, P. 1987. Micro-structural development and toughness of C-Mn and C-Mn-Ni weld metals—Part 2, toughness. Metal Construction, 19(8):447R-450R.

7. Evans, G. M. 1984. The effect of heat-treatment on the microstructure and properties of C-Mn all-weld-metal deposits. IIW Doc.ll-A-605-84; also published in Metal Construction, 17(10):676R-682R.

8. Brouwer, C , and Kalb, J. 1975. A low hydrogen electrode for offshore fabrication. Welding and Metal Fabrication 43(4):425-427; see also Philips Welding Reporter, 10(3-4):19-28.

9. Haas, F. 1980. Major role for Philips 56S low-hydrogen electrode in Norwegian offshore industry. Philips Welding Reporter 10(1-2):17-21.

10. Boniszewski, T. The effect of iron powder in basic low hydrogen all-positional electrodes. Paper 15 in "Trends in Steels and Consumables for Welding," The Welding Institute conference, London, Nov.1978, pp.199-215.

11. Taylor, D. S. 1982. The effect of manganese on the toughness of T7016' type weld metal. Welding and Metal Fabrication, 50(9):452-460; also IIW Doc.ll-A-523-81.

12. Morris, A. G. C. 1983. How welder and current type affect Mn deposition from C-Mn MMA electrodes. Welding and Metal Fabrication, 51(9):395-396.

13. Ahlblom, B. 1984. Oxygen and its role in determining weld metal microstructure and toughness—a state-of-the-art review. IIW Doc.ll-A-642-85 and Doc.lX-|-81-84.

14. Abson, D. J. The influence of current type and polarity on the composition, micro-structure and mechanical properties of C-Mn-Ni and C-Mn MMA deposits. Research Report 292/1985, November 1985. The Welding Institute, Abington, Cambridge, England.

15. Davey, T. G., and Widgery, D. J. 1976. A technique for characterization of weld metal microstructures. IIW Doc.ll-A-387-76.

16. Wegrzyn, J., and Apps, R. L. 1968. Effect of nitrogen on fissuring in mild steel weld deposits. British Welding I., November.

17. Verhagen, J. G., Ouden, dne G., Lifkens, A., and Tichelaar, G. W. 1970. Nitrogen absorption by ferritic weld metal during arc welding. Metal Construction, 2(4).

18. Jenkins, N., and Stevens, S. M. 1977. Nitrogen in ferrous welding. The Welding Institute Research Bulletin 18(1).

WRC Bulletin 354 June 1990

T h e t w o p a p e r s c o n t a i n e d in t h i s bu l l e t i n p r o v i d e de f i n i t i ve i n f o r m a t i o n c o n c e r n i n g t h e e l e v a t e d t e m

p e r a t u r e r u p t u r e b e h a v i o r of 2 1 / .»Cr - lMo w e l d m e t a l s .

( 1 ) Failure Analysis of a Service-Exposed Hot Reheat Steam Line in a Utility Steam Plant By C. D. Lundin, K. K. Khan, D. Yang, S. Hilton and W. Zielke

( 2 ) The Influence of Flux Composition of the Elevated Temperature Properties of Cr-Mo Submerged Arc Weldments By J. F. Henry, F. V. Ellis and C. D. Lundin

T h e f i r s t p a p e r g ives a d e t a i l e d m e t a l l u r g i c a l f a i l u re ana lys is of c r a c k i n g in a l ong i t ud i na l l y w e l d e d h o t r e h e a t p ipe w i t h 1 8 4 , 0 0 0 h o u r s of o p e r a t i o n at 1 0 5 0 ° F . T h e s e c o n d p a p e r de f i nes t h e ro le of t h e w e l d i n g f lux in s u b m e r g e d a rc w e l d i n g of 2 V 4 C r - l M o s tee l .

P u b l i c a t i o n of t h i s r e p o r t w a s s p o n s o r e d by t h e S t e e r i n g and T e c h n i c a l C o m m i t t e e s on P ip ing S y s t e m s of t h e P r e s s u r e Vesse l R e s e a r c h C o u n c i l of t h e W e l d i n g R e s e a r c h C o u n c i l . T h e p r i ce of WRC Bu l le t in 3 5 4 is $ 5 0 . 0 0 per c o p y , p lus $ 5 . 0 0 fo r U.S. and $ 1 0 . 0 0 fo r o v e r s e a s p o s t a g e a n d h a n d l i n g . O r d e r s s h o u l d be sen t w i t h p a y m e n t t o t h e We ld ing R e s e a r c h C o u n c i l , 3 4 5 E. 4 7 t h St. , R o o m 1 3 0 1 , N e w Y o r k , NY 1 0 0 1 7 .

W E L D I N G R E S E A R C H S U P P L E M E N T I 3 6 3 - s

WRC Bulletin 341 February 1989

A Preliminary Evaluation of the Elevated Temperature Behavior of a Bolted Flanged Connection By J. H. Bickford, K. Hayashi, A. T. Chang and J. R. Winter

This Bulletin consists of four Sections that present a preliminary evaluation of the current knowledge of the elevated temperature behavior of a bolted flanged connection.

Section I—Introduction and Overview, by J. H. Bickford; Section II—Historical Review of a Problem Heat Exchanger, by J. R. Winter; Section III—Development of a Simple Finite Element Model of an Elevated Temperature Bolted Flanged Joint, by K. Hayashi and A. T. Chang; and Section IV—Discussion of the ABACUS Finite Element Analysis Results Relative to In-the-Field Observations and Classical Analysis, by J. R. Winter.

Publication of this report was sponsored by the Subcommittee on Bolted Flanged Connections of the Pressure Vessel Research Committee of the Welding Research Committee. The price of WRC Bulletin 341 is $20.00 per copy, plus $5.00 for postage and handling. Orders should be sent with payment to the Welding Research Council, Suite 1301, 345 E. 47th St., New York, NY 10017.

WRC Bulletin 349 December 1989

This bulletin contains two reports that evaluate the PWHT cracking susceptibility of several Cr-Mo steels and several HSLA pressure vessel and structural steels.

Postweld Heat Treatment Cracking in Chromium-Molybdenum Steels By C. D. Lundin, J. A. Henning, R. Menon and J. A. Todd

Postweld Heat Treatment Cracking in High-Strength Low-Alloy Steels By R. Menon, C. D. Lundin and Z. Chen

Publication of this report was sponsored by the University Research Committee of the Welding Research Council. The price of WRC Bulletin 349 is $35.00 per copy, plus $5.00 for U.S., or $10.00 for overseas, postage and handling. Orders should be sent with payment to the Welding Research Council, 345 E. 47th St., Room 1301, New York, NY 10017.

WRC Bulletin 353 May 1990

Position Paper on Nuclear Plant Pipe Supports

This position paper recommends design methods which represent the collective "best methods" of the industry. These enhanced design methods are developed by reviewing the issues that have added to the complexity of design and fabrication of piping systems.

Publication of this report was sponsored by the Task Group on Nuclear Plant Pipe Supports of the Technical Committee on Piping Systems of the Pressure Vessel Research Council of the Welding Research Council. The price of WRC Bulletin 353 is $25.00 per copy, plus $5.00 for U.S., or $10.00 for overseas, postage and handling. Orders should be sent with payment to the Welding Research Council, 345 E. 47th St., Room 1301, New York, NY 10017.

364-s I SEP I EMBER 1992

Effect of Manganese and Type of Current on the Properties and

Documents

Transcript of Effect of Manganese and Type of Current on the Properties and