Flux Composition Dependence of Microstructure and Toughness of Submerged Arc HSLA Weldments
CaF2-CaO-SiC>2 system fluxes produce good quality niobium microalloyed HSLA weldments with very low oxygen content
BY C. B. DALLAM, S. LIU, AND D. L OLSON
ABSTRACT. Twenty eight reagent grade the weld metal microstructure transition heat input and cooling rate, solidification fused fluxes from the CaF2-CaO-Si02 sys with oxygen level was not so clear. characteristics, and reheating thermal tem were used to produce bead-on-plate In spite of the essentially similar optical cycles (in multiple pass welds) contribute and double-V-groove submerged arc microstructure and similar chemical com to the final properties of the weld joint. welds on a quenched-and-tempered nio position (other than oxygen), the Adjacent to the fusion zone is a thermally bium HSLA steel. An E70S3 welding wire mechanical properties of the various affected region (heat-affected zone, i.e., was used with two different heat welds were observed to be very differ HAZ) within which the base metal micro- inputs —namely, 1.9 and 3.3 kj/mm (48.3 ent. Toughness data (upper shelf energy structure is altered by the high tempera and 83.8 kj/in.) A niobium microalloyed and transition temperature) were found ture of the molten weld pool. Martensite steel was selected because of its fine to correlate with weld metal oxygen or other low temperature transformation grained microstructure, high yield content. The upper shelf energy products may be formed impairing the strength, and high toughness at low tem decreased with increasing oxygen level in toughness of these regions. peratures. Fluxes from the CaF2-CaO- the weld metal. As noted below, under separate head Si02 system were selected because of ing, the presence of oxygen can influence their low oxygen potential, and their Introduction and Background weld metal microstructure and proper ability to produce low oxygen (80-450 ties. With this in mind, the behavior of ppm) welds. Quantitative metallography HSLA steels were developed to CaF2-CaO-Si02 flux systems was studied and chemical analysis were performed on achieve high yield strength and at the with the purpose of reporting on weld the welds. The chemical behavior of this same time maintain a reasonable level of metal performance when using these flux system has been characterized with toughness with a minimum of alloying. fluxes on a niobium microalloyed steel. respect to manganese, silicon, niobium, Due to the possibility of increased design and sulfur. loading and strength to weight ratio, Some Background The lower heat input welds showed more and more structural applications predominantly fine microstructure of using HSLA steels are being seen. Some Typical Microstructure of C-Mn Steel acicular ferrite. At high oxygen content, a examples are line pipes for gas and oil Weldments transportation and off-shore structures. higher percentage of grain boundary fer Several different microstructures may The physical metallurgy of these microal rite (ferrite veining) was observed. By be obtained in the weld metal of low loyed steels for the optimization of their reducing the oxygen in the weld metal, carbon microalloyed steels. They are microstructure and properties has the amount of acicular ferrite was grain boundary ferrite (BF), side plate already been treated extensively in the increased. With further reduction of weld ferrite (SPF), acicular ferrite (AF), upper literature (Refs. 1-4) and are not discussed metal oxygen, the main microstructural bainite (AC), and micro-constitutents such in this paper. feature, instead of acicular ferrite, as pearlite, cementite, and martensite. became bainite. Using higher heat input. Most applications of HSLA steels Figure 1 shows some of the main involve structures where welding is used. constituents of a C-Mn steel weldment. A Two major consequences of this fabrica comparison of the various classifications Based on paper presented at the 64th Annual tion process are the deterioration of the of low carbon, low alloy steel weld metal A WS Convention held in Philadelphia, Pennsyl base metal properties due to the welding microstructure is shown in Table 1. vania, during April 24-29, 1983. thermal cycles and the introduction of a solidification structure which is heteroge C B. DALLAM, S. LIU, and D. L. OLSON are Factors Affecting Weld Metal Microstructure with the Center for Welding Research, neous (compared to the base metal) both Department of Metallurgical Engineering, Col chemically and microstructurally. In the F.-jctors affecting weld metal toughness orado School of Mines, Golden, Colorado. fusion zone, the weld metal composition, have been studied and acicular ferrite
140-s | MAY 1985 was found to be the constituent respon sible for the high toughness (Refs. 6, 7). >vru
Table 1—Classification and Terminology of Microstructures in Low C-Low Alloy Steel Weld Metal (Ref. 5).
Dube, CA. Aronson, H.I. Cochrane, R.C.* Widgery, D.J. Abson, D.J. Japanese Researchers'3' Others'6'
Allotriomorphic (polygonal) Proeutectoid ferrite Grain boundary ferrite Proeutectoid ferrite. Grain Proeutectoid ferrite. Grain ferrite* boundary ferrite. boundary ferrite. Polygonal ferrite Polygonal ferrite, Block ferrite - - Polygonal ferrite Polygonal ferrite Ferrite islands Primary and secondary Lamellar component Ferrite with aligned M-A-C Ferrite sideplate Ferrite sideplate ferrite sideplates* (product) Lath-like ferrite Lath ferrite Upper bainite Intragranular ferrite plates* • Acicular ferrite Acicular ferrite Acicular ferrite Acicular ferrite Needle-like ferrite Fine bainitic ferrite Fine grained ferrite Massive ferrite - - Granular ferrite — Pearlite Pearlite Ferrite-carbide aggregate Pearlite Lath martensite Martensite Martensite Martensite Martensite Twinned martensite M-A consitutent M-A constituent M-A constituent Retained austenite High-C martensite Lath ferrite Upper (occasionally lower) Upper bainite bainite
(a) Nakanishi, M., Watanabe. I. I.. Kojima, T.. Mori, N. et al, Kikuta, Y., et al. (b) Pargeter, R. I.; Watson, M. N.. Harrison, P. L. Farrar, R. A.; Levine, E„ Hill, D. C, Sawhill |r., |. M., Choi. C L., Hill, D. C, Clover, A. C, et al., Pacey, A. ).. Kerr, K. W., Abson, D. I.. Dolby, R. E., and Hart, P. H. M.
WELDING RESEARCH SUPPLEMENT 1141-s £- — High ,- Oxygen ' 1 AF \| High \ A 3i Hordenability / / I 1 AC /' \ I j 1
M CoO CoF, 0.10 020 0.30 0.40 0.50 0.60 0.70 0.80 0.90
NCaF? ~ log (t) Fig. 2 — Schematic continuous cooling transformation diagram of Fig. 3 —Partial ternary diagram of CaF2-CaO-Si02 system showing the nominal a low carbon low alloy steel weldment showing the effects of compositions of the fluxes used (open circles first set of welds; filled composition on weld metal phase transformations. In the dia circles—both first and second set of welds) gram, "M" represents martensite, "AC" designates upper bainite (aligned ferrite with carbide), "AF" signifies acicular ferrite, and "BF" represents blocky ferrite
welds (200 ppm to 500 ppm) produced inclusions control the austenite grain rite nucleation is also being investigated tough acicular ferrite structures. Kikuta ef growth by pinning of the grain bound (Ref. 23). al. (Ref. 16) reported the same trend for aries, following the Zener equation of Screes of Oxygen in the Submerged Arc electroslag weldments. precipitate-boundary interaction. They W ding Process Cochrane and Kirkwood (Ref. 10) stud also found that acicular ferrite directly ied the effects of weld metal oxygen on nucleates from the inclusions. The four main sources of weld metal acicular ferrite formation by producing a Keville (Ref. 14) and Pargeter (Ref. 20) oxygc can be identified as the flux, the series of weld metals of similar chemical showed that the inclusions' size, distribu base metal, the electrode welding wire, composition, with the exception of oxy tion, and type (composition) can be cor and the at1 osphere. In general, the oxy gen level. Intermediate weld metal oxy related to the microstructure and the gen and nitrogen pickup from the atmo gen content (200-300 ppm) gave a pri toughness of the weld metal. Keville (Ref. sphere can Le considered as contamina marily acicular ferrite microstructure, 21) also indicated that the inclusions are tion and i ;itutes only a minor part of while side plate ferrite predominated at of complex nature and that distinct geo the weld metal oxygen (Ref. 24). Howev high oxygen contents (500 ppm). They metrical shapes could be associated with er, it may become more significant in concluded that weld metal oxygen varia the type of welding flux used. A change welds made with high basicity fluxes. tion affected the inclusion size distribu in inclusion shape was always observed The lower viscosity of high basicity tion and the surface energy of the inclu with a change in the chemical composi fluxes may be the cause of the increase in sion/matrix interface which resulted in tion of the inclusion. veld metal oxygen pickup (Ref. 24). The different microstructures. Abson, Dolby The types of inclusions were also cata t ase metal and welding electrode con and Hart (Ref. 9) observed the same kind logued using particle analyzer scanning tain approximately 100 ppm oxygen of microstructural evolution as Cochrane electron microscopy (Ref. 22). The idea each; the amount is determined by the and Kirkwood. They also postulated that that the differential thermal contraction r, telting process. The amount that the oxygen-rich inclusions can directly nucle of the inclusions and the austenite matrix base metal and electrode contribute is ate acicular ferrite. In the absence of during cooling of the weldment may appreciably less than that generated from these inclusions, lath type structures provide favorable conditions for the fer the welding flux. The major components (bainite) will form. For higher oxygen systems (>500 ppm), Indacochea and Olson (Ref. 17) obtained predominantly grain boundary ferrite and blocky ferrite (intragranularly Table 2—Chemical Compositions of Base Metals and Welding Electrode, % nucleated coarse ferrite grains). Some evidence of the direct nucleation of the Plate I Plate II E70S3 electrode acicular ferrite phase on inclusions was observed. Increasing weld metal oxygen C 0.14 0.09 0.10 Mn 1.38 1.40 1.15 content, and thus the amount of inclu Si 0.41 0.24 0.53 sions, should move the transformation P 0.005 0.013 0.025 curve in Fig. 2 to shorter delay times. S 0.030 0.004 0.025 Harrison and Farrar (Ref. 18) reported Nb 0.042 0.033 that the austenite to ferrite transforma Al 0.041 0.024 tion temperatures can be altered by the Cr 0.11 0.13 interaction of inclusions with austenite Ni 0.14 0.13 grain boundaries. A high inclusion con Mo 0.06 0.04 centration tended to reduce the austenite Cu 0.18 0.20 0 U> 83 185 grain size, favoring grain boundary ferrite 0 N 90 105 and side plate ferrite formation. Ferrante and Farrar (Ref. 19) determined that the (a) ppm.
142-s I MAY 1985 2.0 Table 3—Welding Conditions for the 1st and 2nd Set of Experiments I / .125 1 t \ .625 1st 2nd 1 / \ .125 1 Potential, volts 30 28 Current, amperes 500 520 4.0 Travel speed, mm/min 480 265 Fig. 4 —loint geometry of the double-V- Heat Input, kj/mm 1.87 3.29 grooved welds; measurements given in inches of a flux are generally oxides, and a tration, and contact angle for interfacial ticular element is made up of contribu reasonably high oxygen potential of the energy calculations. tions from the welding electrode, the flux would be expected. Therefore, the Metallographic analysis and Rockwell base metal, and the weld metal-flux reac main source of oxygen in the submerged hardness measurements were performed tions. The chemical behavior will be arc welding process is the flux. Some on each weldment. Chemical analyses expressed in terms of a quantity called commercial flux systems based on FeO, were performed using a Baird Atomic delta (A); this is the difference between MnO and SiC>2 may contribute up to Spectrovac Model 1000 Emission Spec the composition of a particular element 1000 ppm of oxygen to the weld metal. trometer. Weld metal oxygen, nitrogen, determined analytically and the amount The requirement of low weld metal sulfur, and carbon contents were deter of that element which could be present in oxygen to achieve high toughness (high mined using Leco Analyzers. The chemi the weld if no elemental transfer from the volume fraction of acicular ferrite) leads cal interaction between the molten flux weld pool to the flux or vice versa had to the reformulation of fluxes attempting and weld pool was also studied. occurred. The effects of dilution with the to reduce the oxygen transfer. One way 2. Double-V-groove welds were base metal were also taken into account. is to substitute non-oxygen carriers (such made on plates machined from the plate A negative value of delta means that, as fluorides) for some of the oxides in the II. The dimensions of the welding plates during welding, a particular element has fluxes. Decreasing the amount of oxygen were 4X11X0.625 in. (100 X 280 X been transferred from the weld metal to in the flux, a reduction of weld metal 16 mm). Specific fluxes, indicated in Fig. 3, the slag; a positive delta would indicate oxygen would be expected. A second were used for this second set of welds. the element going from the molten flux way is to replace weak oxides (e.g., silica The joint design is shown in Fig. 4; the to the weld metal. and ferrous oxide) partially by stronger welding conditions are shown in Table The chemical composition and the ele oxides such as calcium oxide and magne 3. mental delta values of the first set of sium oxide. Since the Ca-O bonds are The welds were radiographically welds are presented in Tables 4A, B and much stronger than the Si-O bonds, the examined before standard Charpy speci 5. The delta quantities of the elements degree of dissociation of calcium oxide in mens were prepared. Approximately were then plotted as a function of flux the molten flux would be less than that of nine Charpy bars were machined out of composition. Negative delta values were silica, resulting in a lower weld metal each weld. The Charpy V-notch tough seen for manganese and niobium —Figs. oxygen content. Another possible alter ness measurements were done using a 5 and 6. These delta quantities were also native is the addition of metal powders Tinius Olsen Charpy machine, with the insensitive to flux compositional changes. (e.g., aluminum, titanium) to the fluxes to testing temperatures ranging from —100 This observation implies the activities of reduce the weld metal oxygen pickup to 120°C (-148 to 248°F). At least eight manganese and niobium are nearly con (Refs. 11,25,26). temperatures were tested for each weld. stant for these systems. Fractographic analyses were also done High temperature activity data of MnO using AMR scanning electron micro Experimental Procedure (Ref. 28) in the MnO-CaO-Si02 system scope. indicate that the MnO activity is almost Twenty eight reagent grade fused Carbon extraction replicas were made constant with 40% Si02 content in the fluxes were produced from the CaF2- on some of the welds in order to exam flux providing the conditions of a con CaO-Si02 system, using nominal compo ine the inclusions and other second stant delta manganese in the weld metal sition at every 10% increment on the phases present in the welds. The replicas for that system. Comparing the two ternary phase diagram —Fig. 3. The were examined using the Philips 200 curves in Fig. 5, the same behavior would details of the flux preparation were transmission electron microscope. be expected for welds made with CaF2" reported in a previous paper (Ref. 27). In an attempt to characterize the crys CaO-Si02 fluxes, resulting in the constant Two quenched-and-tempered niobium talline nature of the inclusions, an acid delta manganese. A constant delta quan microalloyed steel plates of similar grade extraction was performed. Some drilling tity in this case also suggests that small were used to make the welds, and a %i shavings from the weld were chemically flux variation will not significantly alter the in. (2.38 mm) diameter E70S3 welding decomposed with hydrochloric acid weld metal chemical composition. electrode was used. The chemical com (10:1). To accelerate the process of dis The delta sulfur quantity is plotted as a position of the plates and the welding solution, the solution was heated to close function of the ratio CaO/Si02 in Fig. 7. electrode are given in Table 2. to boiling with an argon cover gas. After The deviations were all negative, indicat Two sets of experiments were per the extraction, the fluid was filtered ing weld pool desulfurization. With formed: through a Millipore filtration set-up (0.025 increasing CaO-Si02, an increasing sulfur 1. Bead-on-plate welds were made on micron pore size). The filtered particles loss was observed. This is related to the 4X10X0.5 in. (100X250X13 mm) were examined using a Philips x-ray dif- oxygen potential of the molten flux in plates (machined from plate I) with each fractometer and spectrometer. contact with the molten weld pool and one of the fluxes. Instead of using the flux can be explained by the desulfurization hopper, a flux bin was set up along the reaction: torch path. The welding process parame Results and Discussion ters are shown in Table 3. The welds 5(metal) + f
WELDING RESEARCH SUPPLEMENT 1143-s Table 4A—Chemical Composition of the 28 Welds Produced Using a Heat Input of 1.87 kj/mm, Wt-%,W
Flux O Mn Nb Cr Ni Al Mo Cu
1 0.10 122 116 0.97 0.47 0.016 0.030 0.012 0.06 0.05 0.008 0.02 0.11 2 0.10 140 91 0.98 0.52 0.018 0.031 0.014 0.07 0.05 0.003 0.02 0.14 3 0.11 58 92 1.03 0.36 0.022 0.032 0.022 0.06 0.05 0.010 0.02 0.06 4 0.11 183 101 0.99 0.49 0.020 0.028 0.017 0.07 0.06 0.005 0.02 0.14 5 0.10 215 104 0.87 0.47 0.019 0.025 0.008 0.06 0.05 0.003 0.02 0.09 6 0.09 87 96 0.92 0.31 0.011 0.027 0.011 0.06 0.05 0.004 0.02 0.07 7 0.12 206 92 0.87 0.59 0.021 0.026 0.013 0.06 0.05 0.003 0.02 0.11 8 0.10 222 102 0.96 0.47 0.022 0.026 0.014 0.06 0.06 0.004 0.02 0.12 9 0.10 154 100 1.04 0.43 0.014 0.030 0.016 0.07 0.06 0.004 0.02 0.15 10 — 229 116 0.75 0.35 — — 0.025 - — — 0.02 — 11 0.11 266 78 0.92 0.56 0.022 0.025 0.015 0.06 0.06 0.003 0.02 0.13 12 0.10 234 87 0.96 0.51 0.022 0.030 0.019 0.07 0.07 0.005 0.02 0.10 13 0.10 231 86 1.01 0.46 0.021 0.028 0.017 0.07 0.06 0.005 0.02 0.09 14 0.10 213 83 1.05 0.43 0.013 0.029 0.019 0.07 0.06 0.006 0.02 0.10 15 0.10 88 117 0.97 0.20 0.014 0.020 0.008 0.05 0.04 0.003 0.02 0.14 16 0.11 326 81 0.77 0.61 0.023 0.028 0.014 0.07 0.06 0.003 0.02 0.10 17 0.09 265 95 0.87 0.48 0.021 0.027 0.013 0.06 0.05 0.003 0.02 0.11 18 0.10 248 195 1.00 0.49 0.015 0.030 0.015 0.07 0.07 0.005 0.02 0.11 19 0.11 107 93 1.03 0.39 0.011 0.031 0.015 0.07 0.06 0.003 0.02 0.10 20 0.09 301 85 0.93 0.49 0.022 0.028 0.015 0.08 0.06 0.004 00.02 0.10 21 0.11 268 98 0.90 0.49 0.025 0.031 0.015 0.08 0.09 0.005 0.03 0.08 22 0.12 215 160 1.00 0.53 0.012 0.027 0.014 0.07 0.07 0.006 0.02 0.08 23 0.10 250 82 1.00 0.34 0.011 0.027 0.012 0.07 0.05 0.004 0.02 0.07 24 0.09 318 97 0.92 0.47 0.022 0.027 0.016 0.07 0.06 0.004 0.02 0.08 25 0.09 293 114 0.98 0.43 0.020 0.030 0.016 0.07 0.06 0.004 0.02 0.07 26 0.10 154 100 0.98 0.38 0.017 0.030 0.015 0.06 0.05 0.003 0.02 0.07 27 0.09 353 97 0.89 0.44 0.022 0.029 0.015 0.06 0.05 0.005 0.02 0.05 28 0.09 457 98 0.98 0.46 0.019 0.028 0.019 0.07 0.05 0.006 0.02 0.06
(a) Oxygen (O) and nitrogen (N) given in ppm.
(%S) the sulfur distribution ratio between the being partitioned more to the slag. This K' = (%FeO), (2) metal and the slag, and (%FeO)t is the translates into a negative delta sulfur [%S] total iron oxide content in the slag. quantity and agrees with the trend where K' is known as the equilibrium The increase of K', with increasing shown in Fig. 7. By increasing CaO activi index of the sulfur reaction, (%S)/[%S] is basicity in Fig. 8, indicates that the sulfur is ty in the flux, the sulfur will be removed
Table 4B—Flux Compositions for the Table 5—Delta Quantities of Manganese, Silicon, Niobium, Sulfur for the 28 Welds Produced Various Weld Number Designations Using a Heat Input of 1.87 KJ/mm, %
Flux Number %CaF2 %CaO °»Si02 Wei AMn ASi ANb AS
1 100 0 0 1 -0.23 -0.03 +0.002 -0.013 2 90 0 10 2 -0.29 +0.05 -0.008 -0.011 3 90 10 0 3 -0.23 -0.11 -0.001 -0.006 4 80 0 20 4 -0.28 +0.02 -0.004 -0.008 5 80 10 10 5 -0.38 -0.01 -0.010 -0.009 6 80 20 0 6 -0.35 -0.16 -0.010 -0.018 7 70 0 30 7 -0.43 +0.14 -0.014 -0.005 8 70 10 20 8 -0.33 -0.01 -0.012 -0.003 9 70 20 10 9 -0.24 -0.03 -0.008 -0.012 10 70 30 0 10 -0.49 -0.13 +0.008 0.000 11 60 0 40 11 -0.38 +0.11 -0.012 -0.005 12 60 10 30 12 -0.34 +0.06 -0.009 -0.004 13 60 20 20 13 -0.28 0.00 -0.009 -0.005 14 60 30 10 14 -0.22 -0.04 -0.003 -0.013 15 60 40 0 15 -0.28 -0.18 -0.010 -0.013 16 50 10 40 16 -0.54 +0.16 -0.015 -0.005 17 50 20 30 17 -0.43 +0.03 -0.015 -0.006 18 50 30 20 18 -0.30 +0.04 -0.012 -0.012 19 50 40 10 19 -0.25 -0.07 -0.008 -0.017 -0.004 20 40 20 40 20 -0.36 +0.03 -0.010 -0.004 21 40 30 30 21 -0.36 +0.02 -0.005 -0.015 22 40 40 20 22 -0.26 +0.05 -0.007 -0.019 23 40 50 10 23 -0.26 -0.13 -0.008 -0.004 24 30 30 40 24 -0.39 +0.02 -0.014 -0.007 25 30 40 30 25 -0.31 -0.02 -0.010 -0.014 26 30 50 20 26 -0.31 -0.08 -0.010 -0.005 27 20 40 40 27 -0.40 -0.02 -0.010 -0.006 28 10 50 40 28 -0.30 0.00 -0.005
144-slMAY 1985 10 1 1 1 1 —I I 0.0 30 -l 1 1 1 1 1 1 r
40% Si02 0.020 40 % Si02 - 0 8 "5 0.010 — i 0.6 - - 1 ° z MnO-Si0 -CaF 2 2 ^ - 0.010 o - 0 020 0.4 0 / - 0.030 o o J I I I 1_ J L o 0 10 20 30 40 50 60 70 80 i CaF IN FLUX (w/ol 0.2 0 2 c o o <1 o o T ' i r T- 1 1 ! 20 % 0.020 Si02 -
CaO-SiOz- CaF2 0.010 -
-0.2 0 Z o D 0 < 0.010 - u D / p. 0 o O -0.4 - U • • 0.020 - - D 0.0 30 1 i i 1 1 1 1 1 1 i i i t -0.6 20 30 40 50 60 70 20 30 40 50 60 W , in FLUX ( /o) CaF2 IN FLUX (W/.l Fig. 6 —Delta niobium plotted as a function of weight- Fig. 5 —Delta manganese plotted as a function of weight-percent CaF2 for two different flux systems percent CaF2 for different Si02 contents in the CaO-CaF2- SI02 flux system
more readily. This can occur by reducing Fig. 9. When the SiO? content is approxi that region (%Si02/%CaO greater than the silica content in the flux for the mately equal to the CaO content, no one), there was an increase of silicon in
CaF2-CaO-Si02 system. silicon transfer was observed. This corre the weld pool indicating that silicon was Silicon had both positive and negative sponds to the zero delta silicon region transferred from the flux to the weld. delta values (gained or lost) as shown in (shaded region indicated in Fig. 9). Above When %Si02/%CaO is less than one, a
400 1 TYPE OF DATA -0.006 A BASIC 0-H A O.S.M. • 0 BASIC 0-H •0.008 • EQUILIBRIUM • 300
•0.010 <* A o A u_ . 1 3* • 0.0 I 2 <" "l 200 01 3 0 A • 0.0 I 4 . 0 A A A co O A A * A6. •0.0 16 100 0 "**• 8 , O A • 0.0 I 8 8 A*° A A 0.020
p %CaO + 1.4 x %MoO B CaO " %Si02 + 0.84 x %P205 Si02 Fig. 8 — Variation of sulfur equilibrium index (K'). %CaO+ 1.4(%MgQ)
Fig. 7—Delta sulfur plotted as a function of basicity index ~ %Si02 + 0.84 (%P205) '
CaO/Si02 Data were collected from various sources using basic open hearth furnaces
WELDING RESEARCH SUPPLEMENT 1145-s 40%Si02 40% SI02 Welds made with silica free fluxes showed the lowest oxygen contents (<100 ppm). Increasing the silica content in the flux, the weld metal oxygen was found to increase. To a lesser degree, an increase in CaO content also increased the oxygen content in the weld metal. Traditionally, weld metal oxygen and mechanical properties had been related to the basicity index. Tuliani et al. (Ref. 31) CaO CaF, showed that higher basicity fluxes 10 20 30 40 50 60 70 80 90 resulted in lower weld metal oxygen, sulfur, and silicon. Weld metal toughness %CaF 2 was also said to improve with increasing Fig. 9 —Partial ternary diagram of CaF2-CaO-Si02 system showing the composition regions where basicity index (Ref. 32). However, it is still the weld metal gains or loses silicon. Notice the region of no silicon change, ASi = 0 questionable whether the parameter basicity index expresses any fundamental relationship. The flux basicity index (B.I.) 0.30 0.70 equation proposed by Tuliani (Ref. 31) is:
B.I. = [CaO + CaF2 + MgO + BaO + Sr O + K20 + Na20 + Li20 + ¥l (MnO + FeO)]/[Si02 + '/2 (Al203 + Ti02 + Zr02)] (3) Eagar (Ref. 24) modified this relationship by omitting the CaF2 term. Using the equation developed by Eagar (Ref. 24), the weld metal oxygen contents were plotted as a function of basicity index — Fig. 12. The data points showed large scattering without following the usual 0.10 trend as indicated by the solid curve proposed by other authors (Refs. 24, 31). However, if the weld metal oxygen con CaO CaF2 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 tent were plotted using all the terms in equation (3), the data points fell smoothly in the band as seen in Fig. 13 and report
Fig. 10- 1450°C (2642 °E) high temperature activity data ofSi02 in the CaF2-CaO-Si02 system (Ref.ed in the literature (Ref. 24). This compli 30) cates the interpretation of the relation ship between the basicity index and weld metal oxygen. It is also known that weld silicon loss was seen. The thermodynamic trodes. The silicon content, in the present metal oxygen content increases with an activity data (Ref. 30) shown in Fig. 10 case, can be controlled by simply adjust increasing heat input. explains the shaded region of no silicon ing the flux composition. The concept of change, since the silicon isoactivity lines controlling the weld metal composition Microstructure lie parallel to this shaded region. by the proper combination of flux and If it is desirable for the composition of filler metal (i.e., welding electrode) is very A definite trend of microstructural vari the weld to be equal to the base metal important, especially in cases where weld ation was observed within the series of composition, manganese and niobium metal chemistry must be kept in narrow weld metals. For the purpose of analysis, can be introduced into the weld metal by compositional ranges. the following discussion will be based either adjusting the flux composition or The results of the oxygen analysis on upon three particular weldments. Figure by alloy additions to the welding elec- the weld beads are shown in Fig. 11. 14 shows the light micrographs of the three weldments. Their oxygen contents are 350, 250, and 140 ppm, respectively 40%SiO 40% Si02 (each containing similar manganese and silicon contents). The two main features seen in the high oxygen weld metal are grain boundary ferrite and acicular ferrite. The grain boundary ferrite occurred in the form of continuous veining along the prior aus tenite grain boundaries, and constituted around 20 to 30% of the whole struc ture—Fig. 14A. CoO CaF. 10 20 30 40 50 60 70 80 90 By reducing the oxygen in the weld metal to 250 ppm, a refining of the % CaF2 microstructure was seen. There was a
Fig. 11-Partial ternary diagram of CaF2-CaO-Si02 system showing the variation of weld metal significant decrease of grain boundary oxygen as a function of flux composition ferrite veining. Quantitative metallogra-
146-s | MAY 1985 1000 1000 (ref.24) after T Eager (24)
o Ca0-CaF2-Si02 E a. 800 - 800 a.
E Q. 2 O. LU
o 600 UJ 600 O C>D- X LU o C>5- X o < 400 i- 400 UJ 5 LU o _l Ul 200 5 200 T%- O O
0*->- V-1 I V. BASICITY INDEX BASICITY INDEX Fig. 12 - Variation of weld metal oxygen content with basicity index Fig. 13 — Variation of weld metal oxygen content with basicity index omitting the CaF2 term including the CaF2 term
phy showed that acicular ferrite consti content ranged from 200-250 ppm corre reached with respect to the applicability tuted more than 90% of the microstruc sponding to a microstructure approxi of CaF2-CaO-Si02 flux system. There is a ture—Fig. 14B. Decreasing the oxygen mately 90% acicular ferrite. flux compositional region which showed further to 140 ppm, a finer acicular ferrite very good performance during welding; was not obtained; instead long and Physical Properties this appears as the shaded area in Fig. 15.
aligned ferrite laths were formed with an Fluxes on the CaF2-CaO side of the terna The weld penetration, interfacial ten aspect ratio of approximately 10:1 to ry system would not be advisable sions, and arc stabilities were quantita 12:1. Carbide precipitation could be because of its lack of a silicate network. tively measured and analyzed. The results seen between the ferrite plates and were Weld pool protection and the weld have been reported in a previous paper thus classified as bainite —Fig. 14C. surface quality were observed to be (Ref. 27), and are not discussed here. The results depicted in Fig. 14 suggest poor. The region of the CaF2-Si02 side that, for a given flux system, there is an should be avoided because of the two Recommended Fluxes from the optimum level of weld metal oxygen liquids region. Flux behavior may be Fluorspar-Lime-Silica System producing the maximum amount of acic unpredictable. Welds made in the upper ular ferrite. In the CaF2-CaO-Si02 flux From the results of the first set of left region should be avoided in the system, the optimum weld metal oxygen welds, important conclusions can be welding of HSLA steels because the
Fig. 14 — Microstructural variation with weld metal oxygen content: A - 350 ppm of oxygen, mixed microstructure of grain boundary ferrite and acicular ferrite; B- 260 ppm of oxygen, microstructure predominantly acicular ferrite; C- 107 ppm of oxygen, microstructure predominantly upper bainite
WELDING RESEARCH SUPPLEMENT 1147-s 40% Si02
2 LIQUIDS REGION
CaO 60 70 80 90
%CaF0
Fig. 15 — Partial ternary diagram of CaF2-CaO-Si02 system showing as the shaded region the flux compositions which resulted in good quality welds
"higher" weld metal oxygen content oxygen welds. However, no significant enced by the presence of inclusions (low compared to many welds made difference in the microstructure could be (population, spatial distribution, and with other flux systems) would yield a found to distinguish the welds. In spite of shape). Due to the very low solubility of larger amount of grain boundary ferrite the similar microstructure of the different oxygen in BCC iron, the total amount of and side plate ferrite than desired for the welds, their toughness measurements oxygen in solution is negligible. At room maximum toughness. Regions to the left showed quite different results. The upper temperature, oxygen exists in the com side ought to be avoided because of the shelf energies and the ductile brittle tran bined form as inclusions (e.g., oxides, steep ridges in the melting temperature sition temperatures (DBTT) of the weld silicates, aluminates, oxysulfides, etc.). of the fluxes. ments are given in Table 7. Based on this argument, weld metal oxy With an increasing weld metal oxygen gen content could be used as an estimate Toughness content, the upper shelf energy indicating the amount of inclusions in the absorbed at fracture decreased. This rela weld metal as a correlating parameter. From the indicated region in Fig. 9, nine tionship is shown in Fig. 16. In the upper Fundamentally, however, inclusion con fluxes could be selected along the zero shelf region, the Charpy specimens frac tent should be the more appropriate delta silicon quantity line for the second ture by ductile failure mode, absorbing a variable to be compared. set of welds. The chemical composition maximum amount of energy which is Transmission electron micrographs of of the second pass (the untempered related to the magnitude and distribution selected carbon extraction replicas are bead) of this second set of weldments of strain at notch root. It is mainly influ shown in Fig. 17. High weld metal oxygen was determined and presented in Table 6. This criterion of flux selection proved Table 7—The Upper Shelf Energies (USE) and the 100 Joules Transition Temperatures (TT10oj) of the Nine Welds Produced Using a Heat Input of 3.3 kj/mm adequate, because with the exception of oxygen, the welds showed similar com Flux position. A higher heat input of 3.3 kj/ Weld USE, joules TTiooj,°C mm (83.8 kj/in.) was utilized to achieve 1 5 215 -3 deeper penetration such that the Charpy 2 8 220 -3 specimens were made entirely of weld 3 9 195 + 15 metal. Due to the slower cooling rate of 4 13 190 + 18 these weldments, the microstructural 5 17 200 -3 variation was not as clear as the lower 6 18 195 -10 21 190 +20 heat input welds. High magnification light 7 25 185 +6 micrographs showed that slightly finer 27 165 + 18 acicular ferrite was associated with lower
Table 6—Chemical Composition of the Second Set of Weldments—Heat Input 3.3 kj/mm, %
Weld Flux Mn Si Nb Cu O N
1 5 0.085 1.20 0.40 0.038 0.19 205 108 2 8 0.081 1.09 0.46 0.022 0.12 240 80 3 9 0.088 1.27 0.41 0.023 0.14 195 74 4 13 0.085 1.17 0.46 0.021 0.12 391 86 5 17 0.079 1.11 0.47 0.022 0.12 373 72 6 18 0.086 1.25 0.42 0.025 0.13 264 88 7 21 0.085 1.14 0.44 0.022 0.12 345 76 25 0.088 1.19 0.44 0.023 0.12 338 81 27 0.079 1.12 0.51 0.019 0.12 432 75
(a) Oxygen (O) and nitrogen (N) given in ppm.
148-s|MAY 1985 •• iai iamr*^"Tt ^ aa> ^ Jfc«* _ E70S-3 Lukens Frostline ~ 220 3.2kJ/mm, 16mm plate *
> 200 CD rr Ul z Ul * * 180 I cn CE Ul Q. => 160 180 260 340 420 500 OXYGEN (ppm) Fig. 76 - Variation of the upper shelf energies (USE) of the Charpy specimens with weld metal oxygen content showing a decreasing USE with increasing oxygen samples showed higher inclusion content weld metal oxygen can be used to indi than the lower oxygen welds. Finer parti cate the amount of inclusions in the weld Fig. 19 —Scanning electron micrographs of the cles and larger size variation were also metal for correlation purposes. Charpy specimens fracture surfaces at 97 "C seen to be associated with the higher To further analyze the inclusions effect (206.8 "F) high testing temperatures showing oxygen welds. Particles' sizes ranged on the weld metal toughness, the highest ductile type failure with dimples: A — 432 ppm from 0.05 to 1 micron,* and various and lowest weld metal oxygen samples oxygen weld metal; B - 205 ppm oxygen weld metal geometrical shapes were also observed. were chosen. Their Charpy V-notch This all seems to support, to a certain energy transition curves as a function of extent, the previous statement that the temperatures are shown in Fig. 18. Corre dimples, but in the amount of strain sponding scanning electron micrographs associated with each dimple. The micro- of the fracture surfaces at the upper shelf void fracture mechanism can be 3 *1 micron = 10~ mm = 0.00004 in. and lower shelf regions are shown in Figs. described as being composed of three 19 and 20, respectively. stages. The first stage is the initiation of a At the upper shelf regions, dimples microvoid which occurs by the decohe- could be seen in the fracture surface of sion of a particle from the matrix material, both the samples, indicating ductile fail or by the fracture of a second phase ure by the microvoid fracture mecha material. The second stage of the micro- nism. The difference between the two void model is the growth of the dimple samples is not so much in the number of which is independent of inclusion parti- ® • - 205 ppm OXYGEN If- j •V - 432 ppm OXYGEN ** * p Fig. 17 —Carbon extraction replicas of weld 0 20 40 metal showing the effects of weld metal oxy TEMPERATURE (»C) gen content on the distribution of inclusions: Fig. 18-Charpy V-notch energy curves as a function of temperature for the highest and lowest A — 432 ppm oxygen weld metal; B — 205 ppm weld metal oxygen content welds produced with 20%CaF2-40%CaO-40%SiO2 and 80%CaF2- oxygen weld metal 10%CaO-10%SiO2 fluxes WELDING RESEARCH SUPPLEMENT 1149-s The energy difference observed resulted in an improvement of the weld between the high and low oxygen welds metal toughness. may be explained by the presence of 7. High weld metal oxygen welds small regions of ductility between the flat showed high inclusion contents, but the fracture facets. They were more fre inclusions were observed to be finer in quently observed in the low oxygen size. welds. This indicates that the crack has to 8. In the lower oxygen content welds, change directions during propagation a larger strain was found to associate throughout the material, and while doing with the upper shelf ductile dimples lead so, a small amount of ductility is seen. ing to a higher energy absorbed at frac This is related to the finer grain size (thus, ture. more grain boundaries) observed in the 9. The lower shelf failure was charac lower oxygen weld metal. terized by cleavage fracture with small A minimum transition temperature was regions of ductility between the flat frac found in the range of 200-300 ppm of ture facets which account for the energy oxygen. This agrees with the upper shelf difference between high oxygen and low energy data shown previously. The same oxygen welds. behavior has been observed and report ed by Devillers et al. (Ref. 23). When Acknowledgments analyzing these toughness data and com paring with the metallographic and frac- The authors acknowledge the research tographic results, the optimum weld met support of the United States Army Research Office, the fellowship support al oxygen content for the CaF2-CaO-Si02 system was determined to be in the of one of the authors by the Conselho range of 200-300 ppm. Nacional De Pesquisa e Desenvolvimento (Brasil), and the materials and equipment Fig. 20 —Scanning electron micrographs of the support of Lukens Steel Company and Charpy specimens fracture surfaces at —55°C Particle Extraction Hobart Brothers Company. (—67°F) low testing temperatures showing In an attempt to study the chemical cleavage type failure: A - 432 ppm oxygen References weld metal; B — 205 ppm oxygen weld metal and crystalline nature of the particles, an acid dissolution technique was used. The 1. Cohen, M„ and Hansen, S. S. 1979. collected residue was analyzed using x- Microstructural control in microalloyed steels. ray spectrometry and diffractometry. ASTM STP 672, pp. 34-52. cles. The third stage is the coalescence of Peaks for several elements including man 2. Meyer, L, and DeBoer, H. 1977. (Jan). the individual microvoids. ganese, silicon and niobium were seen. HSLA plate metallurgy, alloying: normalizing, As the material between the micro- However, no diffraction pattern could be controlled rolling. Journal of Metals: 17-23. voids becomes necked, the process obtained. This seems to suggest that the 3. DeArdo, A. J., and Brown, E. 1977. (Jan). Hot rolling behavior of austenite microalloyed depends essentially on the properties of inclusion particles could be amorphous with vanadium and nitrogen, journal of Metals: the matrix only. Due to the many fine even though they showed definite facets. 26-29. particles in the high oxygen content The negative results using the x-ray tech 4. Boyd, J. D. 1977. (March). Microstructure welds, the stress-strain behavior of the niques could also be due to the extremely and toughness in high-strength linepipe steels. matrix material between the larger inclu small quantity of residue obtained being Symposium on Toughness Characterization sions is modified. The total amount of insufficient for the characterization. and Specifications for HSLA and Structural uniform strain to failure is reduced. Steels, pp. 216-235. Atlanta, Georgia: Metallur Therefore, the material with more parti gical Society of AIME. Conclusions cles (high oxygen content) has a lower 5. Committee of Welding Metallurgy of total uniform strain to failure, and not The results of this investigation led to Japan Welding Society. 1983. Classification of microstructures in low-C low alloy steel weld much strain will be associated with the the following conclusions. metal and terminology. IIW DOC IX-1282-83. ductile dimples. 1. The CaF2-Ca0-Si02 flux system is 6. Dolby, R. E. 1976. Factors controlling This is reflected in a lower absorbed found to produce good quality niobium weld toughness — the present position, part energy at failure —Fig. 19A. In the lower microalloyed HSLA steel weldments with II—weld metals. The Welding Institute oxygen weldments, the inclusions con very low oxygen content (100-460 Research report 14/1976/M. tent is lower. Fewer microvoid nucleation ppm). 7. Clover, A. G., McGrath, J. T., and Eaton, sites would be expected, requiring more 2. Manganese and niobium showed N. F. 1977. Fracture toughness of submerged energy to fracture the material. This negative delta values during welding, arc weld metal. Symposium on Toughness results in a large degree of plastic defor indicating a loss from the weld pool to Characterization and Specifications for HSLA and Structural Steels, pp. 143-160. Atlanta, mation at the ductile dimples —Fig. 19B. the slag. Georgia: Metallurgical Society of AIME. On the other hand, brittle failure in the 3. There is a region of zero delta 8. Glover, A. G., McGrath, J. T„ Tinkler, M. lower shelf region is more microstructure silicon where weld pool chemistry can be J. and Weatherly, G. C. 1977. The influence of dependent. It is influenced mainly by the easily controlled. cooling rate and composition on weld metal grain size and precipitation hardening of 4. The negative delta sulfur values microstructure in a C/Mn and HSLA steel. the matrix. An examination of the two indicated that the CaF2-CaO-Si02 flux Welding lournal 56 (9): 267-s to 273-s. fracture surfaces revealed flat fracture system was effective in sulfur control. 9. Abson, D. )., Dolby, R. E., and Hart, P. H. facets — Fig. 20. The size of these facets is 5. The optimum weld metal oxygen M. 1978. The role of non-metallic inclusions in at the order of 20-30 microns, corre content for this flux system ranged from ferrite nucleations in carbon steel weld metals. Intl. Conf. on Trends in Steel and Consumables sponding to the allotriomorphic ferrite 200-300 ppm, corresponding to a micro- for Welding, paper 25. The Welding Institute, grain size, suggesting that the cracks tend structure of approximately 90% acicular London. to follow the path which gives the largest ferrite. 10. Cochrane, R. C, and Kirkwood, P. R. uninterrupted route. This is also observed 6. The refinement of the weld metal 1978. The effect of oxygen on weld metal by Ebden and Weatherly (Ref. 33). microstructure due to oxygen decrease microstructure. Intl. Conf. on Trends in Steels 150-s|MAY 1985 and Consumables for Welding. The Welding austenite to ferrite phase transformation in welding. Int. Conf. on "Trends in Steels and Institute, London. high-strength low-alloy (HSLA) steel weld met Consumables for Welding." Paper 21, pp. 11. North, T. H., Bell, H. B., Koukabi, A., and als, lournal of Materials Science 16: 2218- 249-263. London: The Welding Institute. Craig, I. 1979. Notch toughness of low oxygen 2226. 26. Koukabi, A., North, T. H., and Bell, H. B. content submerged arc deposits. Welding 19. Ferrante, M„ and Farrar, R. A. 1982. The 1978. Flux formulation, sulphur, oxygen and lournal 58 (6): 343-s to 353-s. role of oxygen rich inclusions in determining rare-earth additions in submerged arc welding. 12. Ricks, R. A., Howell, P. R., and Barrite, the microstructure of weld metal deposits. Intl. Conf. on "Trends in Steels and Consuma G. J. 1982. The nature of acicular ferrite in lournal of Materials Science 17: 3293-3298. bles for Welding," Paper 22. London: The HSLA steel weld metals, lournal of Materials 20. Pargeter, R. J. 1981. Investigation of Welding Institute. Science 17: 732-740. submerged arc weld metal inclusions. The 27. Liu, S., Dallam, C. B„ and Olson, D. L. 13. Kayali, E. S„ Corbett, J. M., and Kerr, H. Welding Institute research report no. 151/ 1982 (Mar). Performance of the CaF2CaO- W. 1983. Observations on inclusions and acic 1981. Si02 system as a submerged arc welding flux ular ferrite nucleation in submerged arc HSLA 21. Keville, B. R. 1983. Preliminary observa for a niobium based HSLA steel. AWS, ASM, welds, /ournal of Materials Science Letters: 2: tions of the type, shape, and distribution of and ORNL Joint Intl. Conf. on Welding Tech 123-128. inclusions found in submerged arc weldments. nology for Energy Applications, Gatlinburg, 14. Keville, B. R., and Cochrane, R. C. 1981. Welding lournal 62 (9): 253-s to 260-s. Tenn. The influence of inclusion morphology on the 22. Ahlblom, B., Bergstrom, U., Hannerz, N. 28. Abraham, K. P., Davies, M. W., and microstructure and toughness of submerged E., and Werlefors, I. 1983 (Nov). Influence of Richardson, F. D. 1960. journal of Iron and arc weldments. Conf. Proceedings, "Steels for welding parameters on nitrogen content and Steel Institute 196: 82-89. line pipe and pipeline fittings." London: The microstructure of submerged-arc weld metal. 29. U.S. Steel Corporation. 1971. The mak Metals Society. First Intl. Conf. on The Effect of Residual, ing, shaping, and treating of steels. 9th edition, 15. Ito, Y„ and Nakanishi, M. 1976. Study Impurity, and Microalloying Elements on Weld pp. 392. on Charpy impact properties of weld metal ability and Weld Properties. The Welding Insti 30. Sommerville, I. D., and Kay, D. A. R. with submerged arc welding. The Sumitomo tute, London. 1971. Activity determination in the CaF2-CaO- Search 15:42. 23. Devillers, L., Kaplan, D„ Marandet, B., Si02 system at 1450C. Met. Trans 2 (6): 1727- 16. Kikuta, Y., Araki, T., Honda, K., and Ribes, A., and Riboud, P. V. 1983 (Nov.). The 1732. Sakahira, S. 1980 (Oct.). The metallurgical effects of low concentrations of source ele 31. Tuliani, S. S., Boniszewski, T., and Eaton, properties of electroslag weld metal using ments on the toughness of submerged-arc N. F. 1969 (Aug.). Notch toughness of com CeF3 addition wire. Conf. Proceedings, Weld welded C-Mn steel welds. Conf. Proc. "The mercial submerged arc weld metal. Welding ing research in the 1980's, Session B, pp. Effects of Residual, Impurity, and Microalloying and Metal Fabrication: 27. 131-136. Osaka, Japan: Japan Welding Elements on Weldability and Welded Proper 32. Garland, I. G., and Kirkwood, P. R. 1976 Research Institute. ties," P11-P1-11. London: The Welding Insti (Apr.) A reappraisal of the relationship 17. Indacochea, J. E., and Olson, D. L. 1983. tute. between flux basicity and mechanical proper (Dec). Relationship of weld metal microstruc 24. Eagar, T. W. 1978. Sources of weld ties in submerged-arc welding. Welding and ture and penetration to weld metal oxygen metal oxygen contamination during sub Metal Fabrication: 217:224. content. /. Materials for Energy Systems 5 (3): merged arc welding. Welding journal 57 (3): 33. Ebden, ]. R., and Weatherly, G. C. 139-148. 76-s to 80-s. (1983). Weld metal fracture modes in a series 18. Harrison, P. L., and Farrar, R. A. (1981). 25. Craig, I., North, T. H., and Bell, H. B. of HSLA steel weldments. Canadian Metallurgi Influence of oxygen-rich inclusions on the 1978. Deoxidation during submerged arc cal Quarterly 22 (2): 149-155. WRC Bulletin 298 September 1984 Long-Range Plan for Pressure-Vessel Research—Seventh Edition By the Pressure Vessel Research Committee Every three years, the PVRC Long-Range Plan is up-dated. The Sixth Edition was widely distributed for review and comment. Up-dated problem areas have been suggested by ASME, API, EPRI and other organizations. Most of the problems in the Sixth Edition have been modified to meet current needs, and a number of new problems have been added to this Seventh Edition. The list of "PVRC Research Problems" is comprised of 58 research topics, divided into three groups relating to the three divisions of PVRC; i.e., materials, design and fabrication. Each project is outlined briefly in a project description, giving the title, statement of problem and objectives, current status and action proposed. Because of budget limitations, PVRC will not be able to investigate all of these problems in the foreseeable future. Therefore, the cooperation and efforts of other groups in studying these areas is invited. If work is planned on one of the problems, PVRC should be informed in order to avoid duplication. Publication of this bulletin was sponsored by the Pressure Vessel Research Committee of the Welding Research Council. The price of WRC Bulletin 298 is $14.00 per copy, plus $5.00 for postage and handling. Orders should be sent with payment to the Welding Research Council, Room 1301, 345 E. 47th St., New York, NY 10017. WELDING RESEARCH SUPPLEMENT 1151-s AN INVITATION TO AUTHORS Gentlemen: The American Welding Society will hold its 67th Annual Convention and AWS Welding Show in Atlanta, Georgia, during April 13-18, 1986. One of the most important events of our 67th Annual Convention will be its Professional Program. It is indeed a pleasure to invite you as Authors to be participants in the Professional Program of our 67th Annual Convention. On this occasion, the Society is offering an opportunity to Authors to bring the results of outstanding work on their part to the attention of our entire membership, the welding industry, and the nation's metalworking industries. To this end, the Society's Technical Papers Committee will be happy to receive your application for participation in our 67th Annual Convention Professional Program; the Committee is inviting 500-word summaries for, basically, two categories of papers: 1. Applied Technology — unusual industrial or field applications of welding, new process or significant equipment developments, surfacing, unique welding case histories, education, safety and health, cost studies —also related topics such as nondestructive testing of weldments as well as maintenance and repair. 2. Research Oriented — results of significant laboratory research and/or development projects, welding metallurgy, weldability studies, weld cracking or fracture, new test methods, arc physics —also related topics such as fracture mechanics. To apply for participation in our 67th Annual Convention Professional Program, please complete both sides of the Author Application Form on the facing page. Also, please prepare a 500 to 1,000 word summary of what you intend to say in your paper and mail with the completed form to AWS. The Technical Papers Committee will screen all Author applications and summaries (also the manuscripts of completed papers if included), and we expect to notify Authors during late October or early November concerning acceptance. Completed Author Application Forms and accompanying 500-word summaries must be mailed by August 1, 1985, to ensure consideration for the 67th Annual Convention Professional Program. Please note that, as it screens Author Applications and summaries, the Technical Papers Committee will be looking for: (1) the newness of information and need for the information in its field, (2) technical accuracy, (3) clarity of presentation, and (4) adaptability for oral presentation (a paper consisting basically of complicated tables would not be suitable for oral presentation). Sincerely yours, f^OD.lC^^L P. W. Ramsey Executive Director May 1, 1985 AMERICAN WELDING SOCIETY P.O. Box 351040, Miami Florida 33135 152-s|MAY 1985 COMPLETE BOTH SIDES AND AUTHOR APPLICATION FORM MAIL WITH500-1000 WORD -FOR- SUMMARY TO: 67th Annual AWS Convention American Welding Society P.O. Box 351040 Atlanta, Georgia, April 13-18,1986 Miami, Florida 33135 Date Mailed Author's Name Check how addressed: Mr. • Ms. D Title or Position Dr. • Other Company or Organization Mailing Address City State Zip Telephone: Area Number If there are to be joint Name Authors, give name(s) Address of other author(s) Name Address PROPOSED TITLE (10 words or less): SUBJECT CLASSIFICATION: Classify your paper by placing a check mark in the two appropriate boxes: Applied Technology • Research Oriented • Other • /An Original Contribution • A Review • Progress Report • SUMMARY: • Type, double spaced, a summary of not less than 500—but preferably not more than 1000 — words and attach to this form. • Be sure to give sufficient information to enable the Technical Papers Committee to obtain a clear idea of content of the proposed paper; confine background to about 100 words. Also be sure to emphasize Results and Conclusions since this material, together with information on what is NEW, will have a very important bearing on the decision of the Technical Papers Committee. • If complete manuscript is available, in addition to summary, please attach two copies of this form. • Application Form and summary must be postmarked not later than August 1, 1985, to ensure consideration for inclusion on the 67th Annual AWS Convention Professional Program. MANUSCRIPT DEADLINES: • All manuscripts shouldbe in the hands of the Technical Papers Committee not later than March 1,1986. If received prior to November 30, 1985, every effort will be made to publish them in advance of meeting. • It is expected that the Committee's selections will be announced sometime in October or November 1985. • If your paper is made a part of the program, which of the following manuscript deadlines will you be able to meet? November 30, 1985 D January 31, 1986 • March 1, 1986 • PRESENTATION AND PUBLICATION OF PAPERS: • Has material in this paper been previously presented in meeting or published? YesD NoD When? Where? • Following presentation at the 67th Annual AWS Convention, would you accept invitations to present this paper before , WS Sections? Yes • No D • Papers accepted for presentation become the property of the Society with original publication rights assigned to the Welding Journal. RETURN TO AWS HEADQUARTERS. MUST BE POSTMARKEDUOl LATER THAN AUGUST 1,1985, TO ENSURE CONSIDERATION. Author's Signature Author(s) of Proposed Paper: SUMMARY CODE (For AWS Headquarters Use Only) Title of Paper (may be abbreviated) TO AUTHOR(S): Please briefly answer questions in spaces provided below—also please note that answers given below are required in addition to 500 to 1,000 word Summary. Please use typewriter. (1) WHY WAS THE WORK DONE? (2) WHAT WAS DONE? (3) WHAT WAS FOUND? (4) WHAT ARE YOUR MOST IMPORTANT CONCLUSIONS? (5) TO WHOM DO YOU THINK YOUR PAPER WILL BE IMPORTANT? HOW? OTHER COMMENT (Optional) RETURN TO AWS HEADQUARTERS. MUST BE POSTMARKED NOT Be sure to attach 500-1,000 word Summary when LATER THAN AUGUST 1, 1985, TO ENSURE CONSIDERATION. mailing completed Author's Application Form to AWS Headquarters.