RESEARCH PAPER

Reducing potential errors in the calculation of cooling ratesfor typical arc welding processes

A. Hälsig1 & S. Pehle1 & M. Kusch1& P. Mayr1

Received: 13 December 2016 /Accepted: 27 March 2017 /Published online: 6 April 2017# International Institute of Welding 2017

Abstract To reach the goal of weight reduction, modernthermo-mechanically processed micro-alloy steels are increas-ingly used to replace carbon-manganese structural steels. Theprocessing window for welding these newer materials is muchsmaller, so it is important to accurately determine the rate ofheat input, the t8/5-value (cooling rate) and also to ensure thatthe desired cooling rate is achieved in production. Variationsin the welding process, welding parameters welded joint con-figuration, welding position, and layer structure change therate of heat input into the component. At the same time, thearcing efficiency is affected by configuration and position. Incombination, these parameters can affect the cooling behaviorby more than 60%. Various welding processes and parametersare analyzed and the potential errors are discussed. Followingthis, the impact of these errors is illustrated with reference topractical measurements. The summation of the possible errorsshows that it is difficult in practice to achieve the desiredmechanical and metallurgical characteristics, as well as theo-retically predicting these values, and also calculating or sim-ulating properties such as distortion, microstructure, or resid-ual stresses. The work presented here, together with recom-mendations for adjustment of published efficiency values aswell form factors to calculate the t8/5-value (cooling rate), isexpected to make a significant contribution to improve thequality of welded joints.

Keywords (IIW Thesaurus) Arcwelding . Cooling rate .

Efficiency . Energy input . Measurement instruments

1 Introduction and state of the art

1.1 Calculation and influence of the cooling time

Modern high-performance materials have an enormous poten-tial for lightweight construction. During production ofthermo-mechanically processed micro-alloy steels, their spe-cific material properties are adjusted by applying specialtemperature-time regimes and mechanical processing. It isnot possible to replicate this treatment during welding. Theweld thermal cycle inevitably leads to a degradation of thematerial properties. As a consequence, these materials canbe welded only in a very narrow parameter window. In prac-tice, this is specified by a defined t8/5-value [1, 2]. To predictthe cooling rate that will be achieved as a result of the weldingprocess, in current practice, a distinction is made dependingon the welded sheet thickness in two-dimensional (Eq. 1) andthree-dimensional (Eq. 2) heat dissipation [3].

t8=5 ¼ 4300−4; 3T0ð Þ*105* E*ηð Þ2d2

*1

500−T0

� �2

−1

800−T0

� �2" #

*F2 ð1Þ

t8=5 ¼ 6700−5T0ð Þ*E*η*1

500−T 0ð Þ2 −1

800−T0ð Þ2" #

*F3 ð2Þ

List of symbols:t8/5 [s] Cooling time between T = 800 °C and T = 500 °CE [kJ/mm] Energy by unit length

E ¼ U*I=vw ð3Þ

Recommended for publication by Commission XII - Arc WeldingProcesses and Production Systems

* A. Hälsigandre.haelsig@mb.tu-chemnitz.de

1 Technische Universität Chemnitz, Chemnitz, Germany

Weld World (2017) 61:745–754DOI 10.1007/s40194-017-0462-9

η Thermal efficiency of the welding processU [V] VoltageI [A]CurrentPW [W] Welding powertW [s] Welding timevW [mm/s] Welding speedd [mm] Thickness of the plateF2 Form factor for two-dimensional heat dissipationF3 Form factor for three-dimensional heat dissipationT0 [°C]Starting temperature of the componentCalculation of the t8/5-value requires the efficiency of the

process, the form factors with respect to the weld geometry(F2, F3), plate thickness, preheat temperature T0, and weldingperformance and welding energy E. In particular, there is ahigh potential for error in the values that are assigned to effi-ciency and form factor. There is also scope for significant errorwhen measuring the welding power. These effects are ana-lyzed in more detail below.

1.2 Influence of the efficiency

Thermal efficiency is used in welding procedures to cal-culate the energy per unit length of weld. For this pur-pose, values published in standards [1, 2, 4] as well asdata from published literature are used. These values ofefficiency are however fixed, strictly defined for eachwelding process and independent of welding parameters.A classification into different power ranges or subgroupsof the respective welding process does not normally takeplace. Recent research [5] has shown that the rate of heatinput into the component greatly varies as a function ofthe welding parameter settings and physical conditions [3,5]. Thus, for example, GTAW welding has a span in theabsolute values of about 15% between different outputsetting values. An analysis of the literature evaluatingthe average efficiency of the GTAW-process [6–9] an

average value of about η = 0.72 is described. Comparedto the efficiency described in the standard of η = 0.60 [1],calorimetric analyses conducted by the current authorsshow an average value of η = 0.75.

For the GMAW-process, the standard [1] as well as theliterature [7, 10–13] describe an average efficiency of aboutη = 0.80. Recent calorimetric research [5] shows a significantinfluence on the efficiency depending on the metal transfermode. The actual efficiency varies between η = 0.65 andη = 0.90.

Using incorrect values of efficiency in the calculation of thet8/5-value (Eqs. 1 and 2) will clearly result in significant errorsthat can be readily avoided.

1.3 Influence of the part geometry

According to DIN EN1011-2 [4], aspects of the part geom-etry such as sheet thickness, welding seam geometry, andpartially buildup sequence are considered by using correc-tion factors. Just like calculating the t8/5-values even withthe correction factors by the seam geometry F2 and F3 areseparated by the welded sheet thickness and the ratio of heatinput in two- or three-dimensional heat dissipation(Table 1). The correction factors for all types of seam geom-etries are referenced to bead on plate welds, for which avalue of 1.0 is set.

In particular, when using fillet welds, the details are veryvague. Differentiation between opening angle, gap width, orany other geometric parameters are not normally consid-ered. The possible influence of the welding position on therate of heat input and cooling time is also ignored. Therehave been few studies concerned with the influence of thewelding seam geometry [10, 14] and welding position [15]on such factors.

Recent studies [5, 16] measuring levels of emissions, heatconduction, and convection energy of the arc have shown that

Table 1 Correction factors in relation to the seam geometry [4]

746 Weld World (2017) 61:745–754

up to 15% of the welding electrical power are transferred bythe welding arc to the environment. If the arc is enclosedthrough the weld geometry or by the welding position, a partof this otherwise Blost^ energy will be absorbed by the com-ponent. The proportion and the influencing factors for theadditional rate of heat input into the component are unknownat present.

1.4 Influence of the welding power

Several methods exist for calculating the welding electricalpower input [17]. Various values for the welding power canbe determined for the same welding process, which has a highpotential for error or highly imprecise results. The correctmethod of calculation, the welding power for all processes,and conditions are shown in Eq. 4.

Pw ¼ 1

tw

*

∫tw0 U tð Þ*I tð Þdt ð4Þ

To summarize, it can be stated that the calculation of thet8/5-values using Eqs. 1 or 2 is affected by three significantuncertainties. Firstly, the efficiency is fixed depending onthe welding process, without taking different process statesor metal transfer modes into account. Secondly, there areundefined variations in the selection of a correction factorfor the weld geometry. The welding position is completelydisregarded. Additionally, the measurement and calculationof the welding electrical power has a high potential for error,whereby deviations of up to 70% have been determined [18,19]. These sources of error can prevent the accurate estima-tion of heat input into the component. Consequently, in in-dustrial practice, the heat input is usually determined exper-imentally by thermocouple-based measurements of the realt8/5 values.

2 Measurement and experimental technology

2.1 Determination of the heat input—calorimetry

A difference-temperature calorimeter with water as calorime-ter medium (Fig. 1) has been developed to perform compara-tive studies on various welding processes and process condi-tions with defined welding positions and alterable jointgeometries.

In an upwardly open, thermally insulated vessel, thewelding sample is clamped at a defined angle. The weldingtorch is mechanized and moved parallel to the sample sur-face and the water level raises simultaneous with thewelding torch position. Afterwards, the resulting tempera-ture difference ΔT before and after welding is analyzed,accounting for the various mass and materials of all

components. For example, the 1870 g clamping system ismade of copper and the 390 g pumps and pipes are made ofplastic. Since all parts in the calorimeter vessel (1 ... n)absorb heat, their warming must be taken into account(Eq. 5). From the ratio of introduced amount of heat andthe expended electrical work, the efficiency of the weldingprocess is finally calculated (Eq. 6).

QCal ¼ ∑n

1m*c*ΔTð Þ ð5Þ

ηeff ¼QBM þ QCal

PW*tW

¼ QBM þ QCal

tW* 1tE*∫tE0 U

*I*dtð6Þ

List of symbols:Qcal [J] Measured heat value in the calorimetern Number of specific parts in the calorimeter vesselM [g] Mass of the specific part in the calorimeter vesselc [J/(g*K)] Specific heat capacity of the specific part in the

calorimeter vesselΔT [K] Temperature difference in the calorimeter vesselQBM [J] Measured heat in the base material (welded

component)tE [s] Time of the last measurement (end time)To examine the influence of various welding positions, the

device is rotatable mounted (Fig. 2). The rotation from posi-tion PA to PC takes place along the longitudinal axis of theweld. Basically, a rotation of 360 ° is possible, which is repro-ducible to 2 ° accuracy by using an integrated angular scale.Due to the continuously rising water level used by this meth-od, the welding positions PD and PE (overhead) are not fea-sible because the water would impede the welding process.

To analyze the influence of different welding seam geom-etries and types of joint, the clamping system is flexible(Fig. 2). The following welding seam geometries are analyz-able with the adapted calorimetric system:

– bead on plate– lap joint

Fig. 1 Calorimeter method at TU Chemnitz

Weld World (2017) 61:745–754 747

– I-, V- and HV-shape joint– fillet weld/ corner joint.

2.2 Welding parameter for GTAW and GMAW

Arc welding processes are categorized into two groups. Theenergy transfer of processes with non-consumable electrodestakes place via the arc. Using processes with consumableelectrodes the energy transfer is via the arc and also the moltenfiller material. The filler material changes the emissions, heatconduction, and convection energy of the arc as well as the arctemperature. Because of these differences, representatives ofboth groups have been analyzed in this project. Process pa-rameters for the GMAW (ISO 4063:2010 reference number

B135^) and GTAW processes (ISO 4063:2010 reference num-ber B114^) are listed in Table 2.

2.3 Measurement of the cooling rate - t8/5-value

The measurement of actual cooling rate was carried out byusing K-type thermocouples (T = −200...1250 °C, wire diam-eter 2 × 0.6 mm). Measuring the cooling rate was performedby immersing the thermocouple in the still liquid weld poolimmediately after the welding process. Since the measurementposition in relation to the weld was not perfectly consistentdue to manual immersion of the thermocouples, multiple ther-mocouple measurements were performed per weld, as shownin Fig. 3.

Fig. 2 Principle of the turningand clamping function

Table 2 Process parameter for the analyzed GMAWand GTAW processes

Reference parameter GMAW B135^ Reference parameter GTAW B14^

Wire feed speed 10 m/min Welding current 200 A

Arc type standard spray arc Polarity negative

Gas type 82% Ar, 18% CO2 Gas type Argon

Gas volume 15 l/min Gas volume 12 l/min

Distance contact Tube 18 mm Distance of electrode 4 mm

Welding speed 70 cm/min Welding speed 30 cm/min

Torch position neutral in the gap torch position neutral in the gap

Filler wire G3Si1 (1.5125) Diameter of electrode 3.2 mm

Diameter of filler wire d = 1.2 mm Angle of electrode 20°

Base material EN 10025–2 S235JR(1.0038) Base material EN 10025–2 S235JR(1.0038)

Thickness base material 5 mm

Welding position PA

748 Weld World (2017) 61:745–754

2.4 Measurement procedure

To assess representative results, at least three measurementsper parameter setting have been performed with the calorimet-ric system and with the immersing thermocouples in the weldpool. Based on this set of experiments, standard deviationshave been calculated.

3 Discussion of potentials for errors

3.1 Deviations due to efficiency

The efficiency of welding processes is dependent on a numberof process parameters and it cannot be accurately defined as astatic value. For this reason, modification of the standardswhich use this factor may be necessary. [20, 21] Using theGTAW process as an example, DIN EN1011-1 specifies ηeff=0.60. However, a comparison between the work reported bythe authors, and existing published results [5] on the efficiencygave identical results. The average arc efficiency of theGTAW process varies around ηeff = 0.75. The arc efficiencydecreases with increasing welding current, increasing elec-trode distance and increasing positive polarity of the electrode.

By increasing the gas flow or changing the gas mixture com-position by addingmore helium to the argon shielding gas, thearc efficiency of the GTAW welding process increases. Inaddition, some dependence on the base material was detected.A dependence of the absolute rate of heat input into the com-ponent by the component thickness was not determined. [20,21].

Similarly the effective efficiency of the GMAW process isnot a global value of ηth = 0.80, as specified in DIN EN 1011-1 [1]. Rather, a classification according to the type of arc isadvisable [20, 21]:

– Short arc ηeff = 0.85– Pulsed arc ηeff = 0.77– Spray arc ηeff = 0.70

The difference between thermal efficiency and effectiveefficiency is the used amount of heat that is uses as inputparameter. For the thermal efficiency, the whole heat inputinclusive of heat losses due to spatter, spillings, metal vapor-ization, and heat losses of the welding arc is calculated inrelation to the welding electrical energy. By using exact mea-sured values for the real transferred heat into the welded part(by calorimetry) in relation to the measured electrical power,the result is more precise and the term effective efficiency wasintroduced. [20, 21].

Factors that reduce the effective efficiency are [20, 21]:

– Increasing the welding current– Increasing the welding voltage

Factors that increase the effective efficiency are [20, 21]:

– Increasing the amount of shielding gas– Increase in the helium or CO2 content in the shielding gas– Increase in the contact tube to workpiece distance

Fig. 3 Positioning of the thermocouples in the weld seam

Fig. 4 Analyzed welded jointconfigurations

Weld World (2017) 61:745–754 749

It was further noted that modern heat-reduced GMAWshort arc processes have a slightly higher effective efficiencythan classical standard GMAW processes. However, the moredecisive parameter is the required welding power, which isconsiderably lower in heat-reduced GMAW processes. Thus,the absolute rate of heat input into the component, calculatedby the multiplication of the welding power with the efficiencyof the process and welding time, decreases. [20, 21].

3.2 Influence of the welded joint configurations

The influence of different welded joint configurations (Fig. 4)on the efficiency and the rate of heat input of GTAW andGMAW processes have been analyzed. Due to the synchro-nously rising water level within the calorimetric measurementsystem, the I- and V-shaped joints need to be protected fromwater in the weld zone. Cover plates were welded on thebottom side of the joint configurations, with thicknesses be-tween 1…5 mm depending on the transferred energy input perunit length.

The rate of heat input (Eq. 7) is calculated by the multipli-cation of the average welding power (Eq. 4) with the efficien-cy of the welding process (Eq. 6).

Pinput ¼ PW*ηeff ð7Þ

In contrast to the GTAWprocess, energy transfer within theGMAW process takes place via the welding arc as well as bythe short circuit and by the energy transfer of the molten fillermaterial (GTAW just by the welding arc). This is the mainreason why the influence of the welding seam geometry hasa greater influence on the GMAW process. Figure 5 showsthat the actual rate of heat input into the component can beincreased up to 25% when the welded joint configuration ischanged. When changing between similar joint geometrieswith the GTAW process, the measured rate of heat input in-creased by only 7% (Fig. 6).

Since the welding power remains virtually unchanged, theincrease of the rate of heat input into the component must beexplained by the increase in efficiency. In comparison

Fig. 5 Influence of the weldedjoint configuration on theefficiency of GMAW (Parameter:GMAW, see Table 2; BM: EN10025 t = 5…15 mm, PA-position)

Fig. 6 Influence of the weldedjoint configuration on theefficiency of GTAW (Parameter:GTAW, see Table 2)

750 Weld World (2017) 61:745–754

between a bead on plate weld and a V-seam weld the efficien-cy of GMAW increases by more than 15%. This indicates thatincreasingly more energy of the welding process, especiallythe welding arc, is absorbed by the weld edges in the form ofthermal conduction and thermal radiant energy. This increaseis much higher than in GTAW (about 5%).

When considering (Fig. 7), the measured cooling time (t8/5, from T = 800 °C to T = 500 °C) of different GMAW jointconfigurations reduces to less than half when the joint designchanges from a bead on plate weld to a HV-shape joint.

As the results of the increased efficiency show, a majorinfluencing factor is the change of surface area that is affectedby the welding arc when using different joint configurations.On one hand, the heat absorption into the component can takeplace into several directions. On the other hand, linking of the

welding process in direction of the center of the component isbetter. The energy transport is not only a heat transfer modefrom the top surface to the center but also due to weld flanksinto the component. This causes a reduction of the t8/5-value.This is why for example with the HV-shape joint, the energycan distribute better into the component in comparison to thebead on plate weld. This is the reason why the t8/5-value isonly 50% of the bead on plate weld.

The actual information contained in the standards [4] indi-cates, for example, a potential bandwidth of the form factor F2of 0.67...0.90 for fillet welds. As a result of several samplecalculations using the values from the standard [4], it was notpossible to accurately calculate the t8/5-value for the experi-mental cases in this study. For this reason, Table 3 describes aproposal for a necessary adaption of valid regulations [4] in

Fig. 7 Influence of the weldedjoint configuration on the coolingrate t8/5 of GMAW (Parameter:GMAW, see Table 2; BM: EN10025 t = 5…10 mm, PA-position)

Table 3 Proposal for the adaption of form factors (F2 and F3) into one common form factor F in the standard [4]

welding seam geometryForm factor F2 and F3

DIN EN 1011-2 adaption

bead on plate 1.00 1.00

interlayer 0.90 0.85

fillet weld - edge 0.67…0.90 0.60

fillet weld - T-Joint 0.45…0.67 0.50

Weld World (2017) 61:745–754 751

order to achieve a simplification and systematization. Thesuggestion is that a classification into two- and three-dimensional form factors is not necessary.

3.3 Influence of the welding position

When the welding position is changed from PA- to PC-posi-tion, there is an increase in both the efficiency and the percent-age of welding power which is converted into real heat in thecomponent (rate of heat input). In this specific case, the dif-ference of the efficiency between the PA and PB (60 °) posi-tion is more than 12%, as shown in Fig. 8. For GTAW, thisdifference is about 5%.

The experiments show that the increased heat is increasing-ly absorbed by the component and weld flanks. In this case,the rate of heat input in the component increases by 25%. Anexplanation of this fact can provide the analysis of heat flowbehavior of the component in various welding positions.Naturally, a warm medium is moving upwards. Due to theadditional gas flow of the welding process, the arc energy istransported not only in the form of radiant energy, but also

convectively by the shielding gas and atmosphere. Comparedto the PA position, the contact area between the componentand heated gas increases, so more energy can be absorbed bythe entire component.

Despite the unchanged component geometry and constantwelding parameters, the rate of heat input increases whenchanging the welding position from PA to PC. At the sametime, the cooling time increases from t8/5 = 2.9 s to t8/5 = 4.4 s when the welding position changes from PA toPC-position (see Fig. 8). Since the rate of heat input in thecomponent increases by 15% and in addition the cooling timeincreases more than 50%, it seems that the warming was main-ly locally improved. The effect is a better warming orBreheating^ of the joining zone by the encapsulation of thewelding process due to the changed welding position.

In the literature and in the actual standards [4], no correc-tion factors exist with respect to the welding position.However, Fig. 9 illustrates that the welding position has asignificant influence on the cooling time of the component.For this reason, it is recommended to introduce a correctionfactor with respect to the welding position in addition to the

Fig. 8 Influence of the weldingposition on the efficiency ofGMAW (Parameter: GMAW, seeTable 2; BM: EN 10025 t = 8mm,fillet weld- edge)

Fig. 9 Influence of the weldingposition on the cooling rate t8/5 ofGMAW (Parameter: GMAW, seeTable 2; BM: EN 10025 t = 8mm,fillet weld-edge)

752 Weld World (2017) 61:745–754

correction factors with respect to the welded joint configura-tion. A first proposal for the necessary adaption or integrationof a position factor can be seen in Table 4.

3.4 Influence of interlayers

The investigations in relationship to the layer structure werecarried out exclusively with the GMAW process.

Figure 10 clearly shows that as the layers build upwards inthe predetermined V- welding seam geometry, there is a de-crease in efficiency and thus the proportion of welding powerwhich is converted into heat within the component. Comparedbetween root and top layer, the efficiency of η = 0.82 (rootpass) drops to η = 0.70 (top layer).

This confirms the results of previous work that the thermalconduction and radiant energy from the welding arc can bebetter absorbed by the weld edges if the weld zone is enclosedby the component or welded joint configuration. In directcomparison to the results shown in Fig. 5, the root layer hasthe same efficiency value as the V-shape joint (t = 15 mm andα = 60 °), and the top layer produces an identical result to thebead on plate weld.

4 Summary

The results show that the energy and power of heat input intothe component depends on a number of parameters and cannotbe accurately calculated by using the details given in the cur-rent standards [4].

Practical measurement and calculation methods of weldingpower were analyzed and their potential for errors determined.Deviations of up to 70% were determined in the current work[18, 19].

Depending on the welding process parameters, deviationsin efficiency up to 20% can occur [20, 21]. In addition,welding processes with non-consumable electrodes have anefficiency up to 25% higher than described in the currentstandards [1, 2]. It has also been found that welding processeswith consumable electrodes should be categorized dependingon the type of arc (short arc ηeff = 0.85, pulsed arc ηeff = 0.77,spray arc ηeff = 0.70) [20, 21]. The static value of η = 0.80 [1,2] should be adjusted accordingly.

The welding seam geometry has a significant influence onthe efficiency, rate of heat input into the component also thecooling time. For example, when using GMAW and compar-ing the efficiency between a bead on plate weld ηeff = 0.70and an HV-seam ηeff = 0.85, the relative proportion of the rateof heat input into the component increases by more than 20%.However, the cooling time from T = 800…500 °C (t8/5-value)decreases to less than half. As a result, the adaption of thecorrection factors regarding to the welding seam geometriesis necessary, as demonstrated in Table 3.

The welding position also has a significant effect: UsingGMAW, the efficiency increases by about 12% when chang-ing the welding position from PA- to PC-position. In this case,the rate of heat input to the component increases by 25%. Atthe same time, the cooling time (t8/5- value) increases bymorethan 50%. For this reason, it is recommended to introduceanother correction factor (Table 4) in the standards [4] withrespect to the welding position, in addition to the correctionfactors with respect to the welded joint configuration.

Finally, an analysis of the layered V-joint multi-pass struc-ture showed a decrease in efficiency from the root pass to thetop layer, and thus, a decrease in the proportion of the weldingpower that is converted into heat within the component. Theresults were in good correlation to the individual efficienciesobtained in single-pass welds.

Table 4 Proposal for theadaption of the positionfactor in the standard [4]

Welding position Position factor

PA 1,0

PB 1,3

PC 1,5

Fig. 10 Influence of theinterlayer on the efficiency ofGMAW (Parameter: GMAW, seeTable 2; BM: EN10025 t = 15 mm, V-shape jointα = 60 °, PA-position)

Weld World (2017) 61:745–754 753

Acknowledgements This work was part of research project IGF Nr.15.749B/ DVS-Nr. 03.108 of the research coalition BDeutscher Verbandfür Schweißen und verwandte Verfahren e.V.^ (DVS) and was promotedby the program for industrial alliance research (IGF). The financial sup-port by BGerman Federal Ministry of Research and Technology^ via theconsortium BAiF^ is gratefully acknowledged.

References

1. DIN EN 1011-1: BSchweißen – Empfehlungen zum Schweißenmetallischer Werkstoffe – Teil 1: Allgemeine Anleitungen für dasLichtbogenschweißen^

2. SEW 088, Beiblatt 1:1993-10: BSchweißgeeignete Feinkornbaustähle;Richtlinien für dieVerarbeitung, besonders für das Schmelzschweißen^

3. Jaeschke , B. : DVS-Merkb la t t 0973 BÜbers ich t derProzessregelvarianten des MSG-Schweißens^. In: Schweißenund Schneiden 66, Heft 9, DVS Media, Düsseldorf 2014.

4. DIN EN 1011-2: Schweißen – Empfehlungen zum Schweißenmetallischer Werkstoffe – Teil 2: Lichtbogenschweißen vonferritischen Stählen

5 . Hä l s ig , A . (2014) Ene rge t i s che Bi l anz i e rung vonLichtbogenschweißverfahren. In: Dissertation, UniversitätsverlagChemnitz, ISBN 978-3-944640-10-5

6. Mishra, S., DebRoy, T. (2005) A heat-transfer and fluid-flow-basedmodel to obtain a specific weld geometry using various combina-tions of welding variables. J Appl Phys 98:044902 (p. 1–10), ISSN:1089–7550

7. Fahrenwaldt, H. J.; Schuler, V. (2009) Praxiswissen Schweißtechnik-Werkstoffe, Prozesse, Fertigung. In: GWVVerlageGmbH, 3. edition,Wiesbaden

8. Cantin GMD, Francis JA (2005) Arc power and efficiency in gastungsten arc welding of aluminum. Sci Technol Weld Join 10(5):200–210

9. Lapin I, Kosovich V, Potapov A, Savinov A (2003) Energy char-acteristics of the alternating current arc with right-angled pulses inwelding aluminum alloy. Weld Int 17(3):217–220

10. Pépe, N.; Yapp, D. (2008) Measurements of process efficiency for arange of MIG/MAG welding process. In: Conference proceedings:18.-19. Cranfield, UK; ISBN 978–1–903761-07-6

11. Liskevich, O.; Quintion, L.; Scotti, A. (2012) Intrinsic errors onCryogenic Calorimetry Applied to Arc Welding^. In: IIW-Document, No. XII-2061/212–1218-12/IV-1101-12, Denver, USA

12. Egerland S, Pépe N, Colegrove PA, Yapp D, Leonhartsberger A,Scotti A (2011) Measuring the process efficiency of controlled gasmetal arc welding processes. Sci Technol Weld Join 16(5):412–417

13. Joseph A, Harwig D, Farson DF, Richardson R (2003)Measurementand calculation of arc power and heat transfer efficiency in pulsedgas metal arc welding. Sci Technol Weld Join 8(6):400–406

14. Probst R, Hartung F (1982) Energetische Betrachtungen zumMAG-Schweissen von Dickblechen. Schweißtechnik Berlin32(8):345–348 ISSN: 0036-7192

15. Tušek J, Markelj F, Barbič J, Jež B (2003) Influence of type ofwelded joint on welding efficiency. Sci Technol Weld Join 8(3):157–164

16. Hälsig, A.; Kusch, M.; Mayr, P. (2013) Kalorimetrische Betrachtungdes Energieeintrages ins Bauteil. In: DVS Congress, Essen, DVS-Verlag, S. 353–361, DVS-Berichte Band 296, ISBN 978–3–87155-614-2

17. Bernstein, H. (2012) Elektrotechnik/Elektronik für Maschinenbauer– Grundlagen und Anwendungen. In: Vieweg+Teubner Verlag, 2.edition, ISBN 978–3–8348-1606-1

18. Pehle S, Hälsig A, Kusch M, Mayr P (2016) Fehlerpotenziale undMaßnahmen. Der Praktiker, DVS Media GmbH 7:209–295 ISSN0554-9965

19. Pehle S, Hälsig A,Mayr P (2016) Ermittlung von Fehlerpotenzialenbei der Schweißleistungsberechnung energiedynamischer MSG-Prozesse im unteren Leistungsbereich. DVS-Berichte Band 314:68–71 ISBN 978-3-945023-55-6

20. Hälsig A, Kusch M, Mayr P (2012) New findings on the efficiencyof gas shielded arc welding. Welding in the World 56(11–12):98–104. doi:10.1007/BF03321400

21. Hälsig A, Mayr P (2013) Energy balance study of gas-shielded arcwelding processes. Welding in the World 57(5):727–734. doi:10.1007/s40194-013-0073-z

754 Weld World (2017) 61:745–754