Thermal Efficiency Of Arc Welding Processes

dient layer principle where a voltageoutput is produced that is proportional to the heat flux through the calorimeter walls. When a welded40.0sample is placed in the calorimeter,the integrated voltage-time curveproduced as the substrate cools toequilibrium, multiplied by a calibration constant, yields the total quantity of energy transferred to the workpiece by the welding process. Thevoltage signal from the calorimeterwas measured as a function of timeby a personal computer with a dataO.O50.10.1 O,2acquisition system.The system was calibrated byplacing a calibration heater insideFig. 2 - - Calibration p l o t for the arc w e l d i n gcalorimeter. The slope o f the plot defines the calibra- the calorimeter, inducing a knownvoltage across the heater, and meation constant.suring the resultant heater current.power source was used for each process.At steady state, the heater input power diA separate plasma console unit was usedvided by the output signal of thefor control of the pilot arc, plasma gas,calorimeter yields the calibration conand shielding gas for the PAW process.stant of the calorimeter. Figure 2 showsMotion of the individual torches was procalibration results for a range of inputvided by an automated travel carriage.power and displays the linear response ofThe power source, travel carriage, and allthe calorimeter. The slope in Fig. 2 deauxiliary equipment are controlled by afines the calibration constant and wasTexas Instruments/Siemens programmameasured at 276 WV -1. This value wasble control unit.within 1% of the calibration constantThe PAW and GTAW processes werespecified by the manufacturer.conducted using direct current electrodeTo use the calorimeter for arc effinegative (DCEN) polarity with a 4-mmciency measurements, samples were(0.16-in.) diameter, 2%-thoriated tungstenwelded and then quickly placed in theelectrode and argon shielding gas. Thecalorimeter. The resultant voltage signalPAW torch was designed specifically forwas recorded as a function of time by thesurfacing applications. A large constrictdata acquisition system as the samplesing nozzle was used, which containedcooled to room temperature and the volttwo ports for delivery of powder fillerage signal was reduced to zero. The totalmetal into the liquid pool. When used, theheat content of the welded sample waspowder filler metal was fluidized in anobtained by integrating the voltage-timeargon gas and delivered to the nozzleplot and multiplying the integrated voltports by a calibrated screw feeder. Argonage signal by the calibration constant.was used as the plasma gas.The integration was performed using theThe GMAW and SAW processes werepersonal computer and internal software.conducted using direct current electrodeWeld times were kept below 10 s to minpositive (DCEP) polarity with a 1.14-mmimize heat losses prior to placing the(0.045-in.) diameter 308 austenitic stainsample in the calorimeter. Transfer timesless steel welding wire. Argon shieldingto the calorimeter after welding weregas was used for the GMAW process. Theheld below 3 s. For the arc efficiencyvoltage was measured between the torchmeasurements, a 100-mm square by 25and substrate with a programmable voltmm thick (4-in. square by 1-in. thick)meter. For all the processes, the meaA36 steel substrate was used. Heat lossessured voltage represents the sum of voltduring welding and transfer are causedage drops across the electrode and arc.by evaporation, radiation, and convecCurrent was measured by a calibratedtion. Evaporation and radiation from theshunt placed in series with the currentliquid pool during welding of iron havecarrying cable.been estimated to be on the order of 30and 10 W, respectively (Ref. 9). LossesArc Efficiency Measurementsdue to convection, Pc, are given by50.0Ii" mmmm m v0.0[ O . OArc efficiency measurements wereconducted using a Seebeck arc weldingcalorimeter. This apparatus, first described for arc efficiency measurementsby Giedt, e t a l . (Ref. 8), works on the gra-4 0 8 - s I DECEMBER 1995Pc (T-To)1'25 h Abm(4)where T is the elevated surface temperature, TO is the ambient temperature, h isthe convection coefficient, and Abm is thesurface area of the base metal. The surface temperature will obviously varywith position, but an effective value of700 K can be used, which should yieldan upper bound value of heat losses dueto convection. With To 300 K, Abrn 0.03 m 2, and h 1.6 Wm-2K-1.25 (Ref.10), the rate of heat loss due to convection is approximately 85 W. Thus, thetotal rate of heat loss during welding isapproximately 125 W, which is typicallyon the order of 1% of the total arc power.Therefore, heat losses during weldingand transfer to the calorimeter can basically be neglected. This general conclusion has been reported in other work aswell (Refs. 6,11,12).After determining the total heat content of the weld sample, the arc efficiencyis calculated byEcalrla - Vl' -(5)where V is the voltage, I is the current, tis the welding time, and EtaI is the energycontent obtained from the calorimetermeasurement. Experimental measurements have shown that arc efficiencyvaries only slightly with changes in processing parameters (Refs.6,11,13). Therefore, arc efficiency was measured only asa function of current. With the GMAWand SAW processes, current is increasedby increasing the welding wire feed rate,so current variations also correspond tovariations in welding wire feed rate. Thenominal range of primary parametersused for the arc efficiency measurementsare listed in Table 1. The contact tube-towork distance of the (3MAW process wasadjusted for each current and voltage setting to produce a nominal electrode extension of 12 mm (0.48 in.). The electrode extension of the SAW process wasnot controlled due to the inability to observe the arc and electrode. Instead, thecontact tube-to-work distance was heldconstant at 15 mm (0.6 in.). Fused fluxfrom the SAW process was removed before the samples were inserted into thecalorimeter. For the GTAW process, theelectrode-to-work distance was heldconstant at 6 mm (0.23 in.). The stand-offdistance of the PAW process was heldconstant at 15 mm (0.6 in.) with a plasmagas flow rate of 1 L/min.Melting Efficiency MeasurementsOnce the arc efficiency was characterized for each process, the melting efficiency was investigated by depositingadditional welds under the range of parameters listed in Table 2, holding all

other variables constant at the valuescited above for the arc efficiency measurements. Note that these ranges are essentially identical to the ranges listed inTable 1 used for the arc efficiency measurements. As previously noted, the thermal efficiency measurements of thiswork were generated for use in parameter optimization of surfacing applications, and the parameters selected forstudy were chosen based on their importance in surfacing. To simulate a typicalsurfacing procedure, Type 308 austeniticstainless steel was deposited onto A36carbon steel for each process. The steelsubstrates were 305 mm square by 6.4mm thick (12 in. square x 0.25 in. thick)and each weld was approximately 254mm (10 in.) in length. The PAW processutilized filler metal in powder form as described above. A cold wire feeder supplied a 1.14-mm (0.045-in.) diameterwire to the weld pool for the GTAWprocess. The G M A W and SAW processesalso used a 1.14-mm (0.045-in.) diameter electrode. The ranges listed for eachprocess were determined by preliminaryweld trials. The lower limit to travelspeed for a given arc power was governed by the formation of excessivelywide and deeply penetrating welds. Theupper limit of travel speed was established for a given arc power when theprocess could no longer adequately meltthe substrate and filler metal. Thus, thevalues listed in Table 2 represent a widerange of operable parameters for eachprocess under the conditions described.After welding, each sample was crosssectioned using an abrasive cut-offwheel, polished to a 1-pm finish using silicon carbide paper, and etched in a 2%Nital solution. The individual cross-sectional areas of the melted substrate anddeposited filler metal were then measured using a quantitative image-analysissystem. The cross-sectional area termswere multiplied by the total weld lengthto determine the individual volumes ofthe melted substrate and deposited fillermetal. Melting efficiency was then determined byrlm Elm Vfm Esvs l Vlt(6)Where E J'Cp(T)dT AHf (Cp - specificheat, AHf - latent heat of fusion) represents the energy required to raise the fillermetal (Efm) and substrate (Es) to the melting point and supply the latent heat of fusion, Vfm is volume of deposited fillermetal, and v s is the volume of meltedsubstrate. The pertinent values of E areEfm 8.7 J/mm 3 for 308 austenitic stain-less steel (Ref. 14) and Es 10.5J/mm 3 for carbon steel (Ref. 15).The value used here for 308austenitic stainless steel was actually reported for 304 and 304Laustenitic stainless steel since nodata could be found for 308.However, it has been shown (Ref.16) that the slight variations inchemical compositions amongthese grades of stainless steelshave a negligible effect on thespecific heat.Results a n d D i s c u s s i o n1 m9o,S.7ee.If8 o.4oo ,3b.2kgPAW(}TAWGMAWSAW375I"/3423Cu enl, AArc EfficiencyFigure 3 shows the arc effi- Fig. 3 - - Arc efficiency for the PAW, GTAW, G M A W andSAW processes as a function o f welding current.ciency for each welding processas a function of welding current.56% of the energy transfer is attributed toA clear distinction in the ability of eachthe anode work function (Ref. 9). Howprocess to transfer energy to the workever, the arc efficiencies for 304 and 316piece is evident. The data also show therestainless steel and A36 steel were essenis very little variation in arc efficiencytially equivalent, suggesting that theover the current ranges investigated. Thework functions of these materials are simconsumableelectrodeprocessesilar. (This also demonstrates that the use(GMAW and SAW) exhibit an average arcof an A36 steel anode for the GTAW andefficiency of 0.84 0.04. The GTAWPAW processes and the use of anprocess has an average arc efficiency ofaustenitic stainless steel anode with the0.67 0.05, and the PAW process disG M A W and SAW processes, as doneplays an average arc efficiency of 0.47 here, should have no contribution to the0.03. These values are in good agreedifferences in arc efficiencies displayedment with other arc efficiencies reportedin Fig. 3.) Watkins, et al. (Ref. 13), meain the literature for these processes. Forsured the arc efficiency of the G M A Wexample, Smartt, e t al. (Ref. 11), meaprocess on carbon steel using filler metalsured the arc efficiency of the GTAWfeed rates similar to the present work andprocess. For the range of current that wasreported a nominal value of 0.85. Basedsimilar to the present work, the arc effion the similarities of the G M A W andciency was approximately 0.70. ThisSAW processes, it is not surprising to findmeasurement was reported for a 304that the arc efficiencies for thesestainless steel anode.processes are essentially identical. TheThe anode material can have an effectlow arc efficiency of the PAW process ison arc efficiency since approximatel

The effect of welding parameters and process type on arc and melting efficiency is evaluated BY J. N. DuPONT AND A. R. MARDER ABSTRACT. A study was conducted on the arc and melting efficiency of the plasma arc, gas tungsten arc, gas metal arc, and