Effects of Shielding Gas on Gas Metal Arc Welding Aluminum
Contrary to the conclusion of previous investigators, argon may be the optimum shielding gas for the gas metal arc welding of aluminum
BY W. R. REICHELT, J. W. EVANCHO AND M. G. HOY
ABSTRACT. Argon, helium, and mix Introduction to the relative ease with which the tures of these gases are used for shield shielding gas forms the plasma; the ing in gas metal-arc welding of alumi Shielding gas, as an important vari higher the ionization potential, the num. Considerable work has been able in gas metal-arc welding of alumi more difficult it is to initiate an arc. done previously by other investigators num, serves three primary functions: It High ionization potentials can also to define the effects of these gases on provides a plasma for commutation of result in poor arc stability; helium, penetration, arc stability and porosity. current; it protects the weld pool from therefore, welds with a less stable Most of this work, however, was per reacting with the air environment; and arc. formed by maintaining an "optimum" when welding using direct current re Thermal conductivity is important, arc gap. Consequently, for various verse polarity (DCRP), it can provide because a gas with good thermal con compositions and flow rates studies, the cleaning action which partially ductivity will help to conduct heat weld parameters were also varied. removes aluminum oxide from the alu into the workpiece. Degree of thermal To further understand the effects of minum plate. conductivity has been reported to shielding gas on gas metal-arc welding The physical properties and charac affect the shape of the weld bead and aluminum, a study was conducted teristics of both argon and helium, the the condition of the metal next to the whereby all weld settings were preset two principle shielding gases used for weld. Helium's high thermal conduc at constant values and effects of welding aluminum, are compared in tivity relative to that of argon is one shielding gas composition and flow Table 1. The ionization potential of factor why welds made with helium rate on arc gap, voltage, and current, in argon is much lower than that of show a broader weld puddle. Density addition to penetration, arc stability, helium; ionization potential is the of helium is one-tenth that of argon. and porosity were evaluated. Shielding voltage required to move an electron Because of this difference in density, gas composition and flow rate both from an atom, thereby turning the higher flow rates are generally re affected depth of penetration. Maxi atom into an ion. Comparison of ioni quired when welding with helium. mum penetration was obtained with zation potentials gives an indication as Argon possesses excellent cleaning pure argon at 150 cfh, whereas mini mum depth of penetration was asso Table 1—Properties of Shielding Gas ciated with a 25% argon/75% helium mixture at 200 cfh. Argon Helium Effects of shielding gas on depth of penetration correlated directly with Ionization potential 15.8 eV 24.6 eV Arc initiation Good effects of shielding gas on wattage. Poor Arc stability Good Poor Gas compositions high in argon also Thermal conductivity 0.406 x 10 3.32 X 10- produced best arc stability, and the (cal/sq. cm/cm/°C/s) minimum amount of micro-porosity. Density (relative to air) 1.38 0.137 Results of this work suggest that, con Cleaning action Good Poor trary to the conclusions of previous investigators, argon my be the opti mum shielding gas for gas metal-arc Table 2—Advantages/Disadvantages of Helium and Argon welding of aluminum. Advantages Disadvantages
Helium • Higher arc voltage and assumed • Poor cleaning Paper presented under sponsorship of the greater heat input (deep penetra • Poor arc initiation and stability Aluminum Alloys Committee of the Weld tion and high welding speeds) • Cost ing Research Council at the AWS 60th • Requires greater flow rates Annual Meeting in Detroit, Michigan, dur • Broad weld root width • Narrow weld root width ing April 8-12, 1979. Argon • Good arc initiation and stability • More effective shielding W. R. REICHELT, J. W. EVANCHO and M. G. • Low cost HOY are with the Joining Division of Alcoa • Good cleaning Laboratories, Alcoa Center, Pennsylvania.
WELDING RESEARCH SUPPLEMENT I 147-s Fig. 1—Grooved 5083 plate. Groove dimensions: VA in. (6.3 mm) Fig. 2—Commercially available equipment utilized during weld evalua deep and Yi in. (12.7 mm) wide tion. Shown are the welding torch, arc viewer, track wire filler metal feed controller
action. Helium on the other hand pro result, several "rules of thumb" have travel speed—12 ipm (5 mm/s). vides no cleaning. evolved as to the benefits of the vari These parameters resulted in an arc Because of the differences in physi ous shielding gases or the mode of length of 0.148 in. (4 mm). Subsequent cal properties and characteristics of using the various shielding gases with welds were made maintaining these helium and argon, each shielding gas out substantial theoretical verification. machine settings as shielding gas com has its advantages and disadvan In this regard, most references on arc position and flow rate were varied. tages-Table 2. A higher arc voltage welding indicate that shielding gas Initially, three welded samples were associated with helium results in a flow rate should be selected by reduc made for each condition of shielding greater heat input and supposedly a ing flow rate until an unstable arc is gas composition and flow rate. How deeper penetration. This greater heat observed and then increasing the flow ever, uniform results permitted the input also permits welding at higher rate slightly for a safety margin. Anoth number of samples to be reduced to speeds. The main advantage asso er statement is that helium provides one per shielding gas variable. ciated with the high heat input, how greater penetration. ever, is the broad weld root width Previous in-house process develop Phase II—Arc Gap Control obtained when welding with helium. ment work has suggested that shield The procedure for Phase II was simi Although the greater heat input is a ing gas flow rates may be just as lar to that of Phase I. In this phase, major advantage for helium, this gas significant as shielding gas composi however, a constant arc gap of 0.150 also has several disadvantages. Helium tion in affecting weld penetration and in. (4 mm) was maintained first by does not assist in cleaning the oxide weld cross-section geometry. To quan adjusting current. The procedure was from aluminum plate, and its high tify the effects of shielding gas compo then repeated and arc gap was ionization potential results in difficult sition and flow rate, a program investi adjusted by voltage (filler metal arc initiation and poor arc stability. A gating both argon and helium was feed). greater flow rate is also required with conducted. helium because of its lower density. Accurate shielding gas flow rates were maintained through the use of Probably the greatest disadvantage of Procedure thermal mass meters. Voltage and cur this shielding gas is related to its cost. rent were recorded using a Brush Helium continues to be more costly Effects of shielding gas composition recorder, and arc gap was observed by than argon. and flow rate on weld penetration and means of an arc viewer mounted on Argon has many advantages over weld cross-section geometry were de the travel carriage. helium. Because of its low ionization termined by automatic GMA welding All welds were made on a grooved potential, it demonstrates good arc in the flat position. Shielding gas com 5083-0 plate, 1X9X12 in. (25.4 initiation characteristics and excellent positions ranging from pure argon to x 228 X 305 mm), using 3/32 in. (2.4 arc stability. Argon is also effective in pure helium, including various mix mm) diameter 5183 electrode. The assisting cleaning of the oxide from tures, were evaluated at flow rates groove was VA in. (6.3 mm) deep and Vi aluminum plate. Because of its higher ranging from 50 cfh (23.5 liters/min) to in. (12.7 mm) wide—Fig. 1. Shielding density, argon provides a more effec 300 cfh (141 liters/min). The program gas was commercial welding grade tive shield from the environment at was conducted in two phases as argon and helium. lower flow rates than are required for described below. helium; also, most importantly, argon All welds were cross-sectioned at the midpoint perpendicular to the is cheaper than helium. The main dis Phase l-No Arc Gap Control weld. The cross-section was polished advantage attributed to argon is the Initial weld parameters were deter and etched; and weld penetration, narrow weld root width that results mined using a shielding gas composed width at varying depths,* and crown when using this shielding gas. of 50 cfh (23.5 liters/min) helium and height were measured. In Phase II, Many of these statements about 50 cfh (23.5 liters/min) argon. Wire welders' comments concerning arc argon and helium as shielding gases filler metal feed speed, current and characteristics were also docu travel speed were adjusted to produce have resulted from qualitative evalua mented. tions over several years. However, only a smooth welding arc and a sound limited quantitative data on the effects weld. The weld settings determined by of shielding gas composition or flow an experienced welder's observations *Weld width measured approximately VA in. rate are available in the literature. As a follow: voltage-35 V; current-404 A; (6.4 mm) from bottom of weld
148-sl MAY 1980 Equipment
Commercially available equipment was utilized for this investigation—Fig. NO ARC GAP CONTROL 2. This equipment included a water- cooled welding torch with a 3/4 in. (19 mm) gas cup, a wire filler metal feed controller, a Gulco KAT track, and a drooping volt-ampere characteristics power supply.
Results
Phase I—No Arc Gap Control When arc gap was allowed to vary, both shielding gas composition and flow rate significantly affected weld penetration—Fig. 3. For pure argon, maximum penetrations were asso ciated with 150 cfh (70.5 liters/min) flow rates. Flow rates greater than or less than 150 cfh (70.5 liters/min) resulted in lesser penetrations. Mini mum penetrations were associated with shielding gas compositions be tween 60 and 80% helium at approxi 100 mately 150 cfh (70.5 liters/min) to 200 ARGON - CFH cfh (94 liters/min) flow rates. Increas Fig. 3—Effect of shielding gas on weld penetration ing or decreasing the flow rate or increasing or decreasing the percent age of helium in the shielding gas resulted in increased penetration. In containing 70% helium—Fig. 4. taining 50% or more argon. As the no case, however, was penetration as Effects of shielding gas on weld compositions became richer in he great as the level obtained with 150 cfh cross-sectional area were complex and lium, flow rate had less effect; howev (70.5 liters/min) pure argon. are shown in Fig. 5. For shielding gas er, variations in composition between Shielding gas flow rate had little compositions rich in argon, flow rate 70 and 90% helium did affect cross- effect on weld root width. However, had a significant effect on cross- sectional area. shielding gas composition significantly sectional area. Maximum cross-sec These results are further illustrated affected weld root width. Root widths tional areas of 1 square inch (645 mm2) in Fig. 6, which shows weld cross- less than 0.250 in. (6.3 mm) were pro was obtained at flow rates between section profiles as affected by shield duced with pure argon and increased 100 cfh (47 liters/min) and 150 cfh ing gas composition and flow rate. Of to 0.4 in. (10 mm) with compositions (70.5 liters/min) for compositions con particular interest is the effect of flow
90% He 80% 60% 90% He 80% 70% 60%
NO ARC GAP CONTROL
50% 150
Jj \.850 SQ. IN. // / E 100 / \A.900 \ s => Zl ^V.950 VAA III / \\j.OOo\ \ Y\
50 j.ooo \ J>«r\ \ \
20% 20
.850^-- 10%
50 100 150 50 100 150 ARGON - CFH ARGON - CFH Fig. 4—Effect of shielding gas on weld root width Fig. 5—Effect of shielding gas on weld cross-sectional area
WELDING RESEARCH SUPPLEME NT I 149-s 200 NO ARC GAP CONTROL
U
00
uJ I
50
100 150 200 ARGON-CFH Fig. 6—Effect of shielding gas on weld cross-section geometry
rate on increasing the weld root width improvement over the narrow weld Phase II—Arc Gap Control when pure argon is used. Weld root spike typically associated with pure profile for 150 cfh (70.5 liters/min) argon. Wide weld widths generally Arc Cap Controlled by Current. argon is similar to profiles using an associated with pure helium or com When arc gap was adjusted by current argon-helium gas mixture. A flow rate positions high in helium are most evi to 0.150 in. (4 mm), penetration was of 150 cfh (70.5 liters/min) pure argon dent at the top of the weld. However, primarily affected by shielding gas provides adequate weld width and this wide weld bead provides no sig flow rate; shielding gas composition deep penetration and is a significant nificant benefit. had essentially no effect—Fig. 7.
80% 60% 90% He 80%
/ \ ARC GAP - 0.150" r / #\ / / J \ ADJUSTMENT BY )\ / ^y \CURRENT .350 IN. \ / y
\.350 IN. I I / X ! 1 O 100 Jy^/
I / / 30%
50 JS \ ^^250 IN^\
20%
\i //yy^ .200 IN. _\ —* \"
50 100 50 100 ARGON - CFH ARGON - CFH Fig. 7—Effect of shielding gas on weld penetration Fig. 8—Effect of shielding gas on weld root width
150-sl MAY 1980 Increasing the gas flow rate beyond 90% He 80% 70% 60% 100 cfh (47 liters/min) decreased weld penetration. Both shielding gas flow rate and composition affected weld root 50% width—Fig. 8. Mimimum weld root widths were associated with pure £L argon shielding at 50 cfh (23.5 liters/ o _l min). Maximum weld root widths UJ were associated with pure helium at > flow rates exceeding 100 cfh (47 liters/ min). X Flow rate had a significant effect on o 40% DC weld cross-sectional area for shielding < UJ gas compositions rich in argon—Fig. 9. CO Increasing flow rate decreased weld E cross-sectional area. On the other _i hand, shielding gas compositions rich UJ x z in helium were not affected by flow 30% rate, although changes in helium con UJ centration did affect the weld cross- 0. o sectional area. Maximum cross-sec _l 2 UJ tional areas of 1 square inch (645 mm' ) 20% were obtained with pure helium. > UJ The results are further illustrated in Q X Fig. 10. The spike typically associated o with welds made using pure argon 10% Q. X o > I00 UJ o -^ X _j DC < UJ jjj 50 CO ARC GAP =0.150 in. ADJUSTMENT BY ao. _J UJ 0 CURRENT > UJ o >*. x o WELDING RESEARCH SUPPLEMENT I 151-s I 90% He 80% FUZZY \ ARC \ VIOLENT ARC - UNDULATING PUDDLE UNDULATING 100 ~ VIOLENT / PUDDLE ARC / SMOOTH ARC 50 V I V I 100 150 ARGON -CFH Fig. 11-Effect of shielding gas on arc characteristics—current adjustment SO 100 Fig. 12— Effect of shielding gas on weld penetration ARGON - CFH flow rates beyond 125 cfh (58 liters/ as percentage of helium was in sectional areas are associated with min) resulted in a rough, undulating creased. pure argon at flow rates less than 100 arc. Violent arc characteristics were Weld root width was affected pri cfh (47 liters/min)—Fig. 14. These associated when pure helium was marily by shielding gas composition. results are dramatically illustrated in employed. However, shielding gas flow rate Fig. 15. The tremendous benefit of Arc Cap Adjusted by Wire Feed. slightly affected root width, especially pure argon on weld penetration is Contrary to the results obtained when at flow rates exceeding 150 cfh (70.5 obvious from Fig. 15. Of particular arc gap was adjusted by current, when liters/min). Minimum weld root interest is the tendency towards a arc gap was adjusted by wire feed, widths were associated with pure wider weld root width for pure argon shielding gas flow rate did not affect argon, and maximum weld root widths at 100 cfh (47 liters/min) than is weld penetration significantly. Pene were associated with pure helium, as normally associated with pure argon tration, however, was significantly shown in Fig. 13. shielding. Weld root width increased affected by shielding gas composition Both shielding gas flow rate and considerably when pure helium was — Fig. 12. Maximum penetrations were composition affected weld cross-sec employed; however, penetration de obtained for pure argon and decreased tional area slightly. Maximum cross- creased significantly. Benefits of ad- 90% He 80% 70% 60% 90% He \ . / / /\ ARC GAP - 0.150" I \/ / \ ARC GAP - 0.150 V / \ ADJUSTMENT BY 450 IN ADJUSTMENT BY / // N^ WIRE FEED 150 / \ WIRE FEED y 50% 'y / \ ^y / \ 1 Y\400 IN. / 1/ N, V \^^/ r\ >v^/*Ns>^600 SQ. IN. 100 ,_ w / ' <-> 100 .500 IN.\ /SoO IN s 3 \ i\/y / 30% / *5*^S^\ ** ^*< y' ^\\^sv600 SQ" IN*\ SO V^"x<250 IN. 50 4 \ 1 \s ^yy^ \ \~S<--"'''' .650 SQ. IN.\ yc~~~~ i. y^y~- \ -200 IN. -«^- ,-.700 SQ. IN!^\ _\X—V* —^T" J \ \i \i X \ A 50 100 50 100 150 ARGON - CFH ARGON - CFH Fig. 13-Effect of shielding gas on weld root width Fig. 14—Effect of shielding gas on weld cross-sectional area 152-sl MAY 1980 150 a. X O _i y !00 >UJ O ADJUSTMENT BY 0. o —i WIRE FEED UJ 0 > x o CL O _i i i i i UJ > UJ SLUDGE I VIOLENT ARC • UNDULATING a FUZZY I PUDDLE y ARC SLIGHT \ x \ UNDULATING o 100 - J PUDDLE j " WELDING RESEARCH SUPPLEMENT I 153-s 1 1 1 I I I I I I i i ARC GAP - 0.150" ARC GAP = 0.150* m CURRENT ADJUSTMENT 0 CURRENT ADJUSTMENT • • O WIRE FEED ADJUSTMENT O WIRe FEED ADJUSTMENT - .600 "• " • - o°° o o • • o • • - 8? ft o °o%§* 0 % .200 I I i I I i U i i I 40 50 60 70 80 90 60 70 3 3 JOULES X 10 JOULES x 10 Fig. 19— Effect of heat input and method of arc adjustment on penetra Fig. 18—Effect of heat input and method of arc adjustment on weld cross-sectional area tion I I I 1 1 1 I i I ARC GAP = 0.150" CURRENT ADJUSTMENT o • ARGON HELIUM WIRE FEED ADJUSTMENT .6 - o - o 100% - — 100% A• 50% 50% 25% 75% J5 - 0 o A - - \cA* CD • • • O • C\ rP\ 3 - O 9 - - ^ b - J? ^Jgn- \ ?: 2 •• - ^Cp ^ • o I i I 1 I I i i I 60 70 JOULES x 103 WIDTH (IN) Fig. 20—Effect of heat input and method of arc adjustment on weld Fig. 21—Effect of shielding gas composition on weld penetration/ root width weld root width ratio however, are not consistent with those section geometry then are attributed ment decreased weld root width as previously reported in the literature. to the effects of changing shielding gas heat input increased. Conversely, in Depending on how arc gap was composition and flow rate on welding creasing heat input increased weld adjusted, much deeper penetration current. This is further illustrated in root width when arc gap was main could be obtained when pure argon Figs. 18,19 and 20. tained by current adjustment. was employed. More important, how The effect of heat input on weld True improvements in weld bead ever, is the effect of flow rate. Flow cross-sectional area is indicated in Fig. geometry are associated with increas rate not only affects weld penetration 18. Regardless of shielding gas compo ing weld root width while maintaining but weld root width and, by employ sition and method of arc gap adjust a constant penetration or increasing ing proper controls, a desirable weld ment, weld cross-sectional area in penetration. Such improvement has cross-section geometry can be ob creased as expected with increasing normally been attributed to the use of tained when using pure argon. heat input. The method of arc gap helium mixtures in shielding gas. Fig Reasons for the effects of shielding adjustment did, however, affect the ure 21 shows the ratio of weld penetra gas composition and flow rate on weld weld cross-section geometry. As indi tion to weld root width as a function cross-section geometry are not ob cated in Fig. 19, increasing heat input of shielding gas composition. Greater vious. Contrary to what had been increased weld penetration when ad weld root widths for a given penetra reported previously, greater penetra justment in arc gap was made by wire tion were obtained when pure helium tion did not result from a higher arc feed adjustments. Current adjust was employed; however, greater voltage associated with pure helium ments, however, resulted in a constant depths of penetration were obtainable with pure argon. Most interesting, shielding. Penetration, however, was weld penetration even though total however, is that additions of helium to affected by the welding parameters, heat input was affected substantially. shielding gas up to a level of 75% particularly current (Fig. 17), regardless The opposite effect is noted for weld helium did not significantly affect ratio of voltage (wire feed speed) em root width—Fig. 20. Increasing heat of weld penetration to weld root ployed. The differences in weld cross- input by wire filler metal feed adjust 154-sl MAY 1980 width. row weld root widths resulting from 2. The manner by which shielding argon shielding, proper selection of gas composition and flow rate affected shielding gas flow rate and method of weld penetration and cross-sectional area is dependent on method of arc Conclusion arc adjustment can optimize weld root width when pure argon shielding is gap adjustment. Results of this work indicate that employed and produce satisfactory 3. Maximum penetrations are pro selection of shielding gas in gas metal- weld geometries. Consequently, the duced with pure argon or argon-rich o0- arc welding is an important criteria following conclusions are warranted: shielding gas. —I LU which affects not only the arc charac 4. Desirable weld cross-sectional > geometry can be obtained with pure LU teristics but the weld geometry. 1. Shielding gas composition and a Although much concern has previous flow rate can affect weld penetration argon shielding provided flow rate is x ly been expressed regarding the nar and cross-sectional area. optimized. o WRC Bulletin 256 January 1980 Review of Data Relevant to the Design of Tubular Joints for Use in Fixed Offshore Platforms by E. C. Rodabaugh The program leading to this report was funded by 13 organizations over a two-year period. The objective was to establish and/or validate design methods for tubular joints used in fixed offshore platforms. The report is divided into four self contained chapters (1) Static Strength (2) Stresses (3) Fatigue (4) Displacements, wherein detailed cross comparisons of various types of test data and design theories are reviewed. Publication of this report was sponsored by the Subcommittee on Welded Tubular Structures of the Structural Steel Committee of the Welding Research Council. The price of WRC Bulletin 256 is $13.00 per copy, plus $3.00 for postage and handling. Orders should be sent with payment to the Welding Research Council, 345 East 47th St., Room 801, New York, NY 10017. WRC Bulletin 257 February 1980 Analysis of the Ultrasonic Examinations of PVRC Weld Specimens 155, 202 and 203 by Standard and Two-Point Coincidence Methods by R. A. Buchanan, prepared for publication by 0. F. Hedden This report describes two methods of analysis of ultrasonic examination data obtained in a 13-team round-robin examination of three intentionally flawed weldments. The objective of the examinations is to determine the accuracy of independently detecting, locating and sizing the weld flaws, using a fixed procedure. Computer programs to facility comparison of flaw locations with the ultrasonic data, for each of the specimens and both of the methods, are appended to the report. Publication of this report was sponsored by the Subcommittee on Nondestructive Examination of Materials for Pressure Components of Pressure Vessel Research Committee of the Welding Research Council. The price of WRC Bulletin 257 is $11.00 per copy, plus $3.00 for postage and handling. Orders should be sent with payment to the Welding Research Council, 345 East 47th St., Room 801, New York, NY 10017. WELDING RESEARCH SUPPLEMENT I 155-s