Practical measurement of voltage and current in arc welding

TWI Bulletin, April 1985

by John Street

John Street is Principal Research Engineer in the Control Engineering Department.

This practical guide to measurement of arc current and voltage, also discusses sources of error and methods of dealing with them so that their effects are minimised.

Arc voltage and welding current are two of the major variables that need to be controlled in fusion welding processes. Where current and voltage meters are incorporated into arc welding equipment their accuracy may be checked in situ by specially designed reference sources. ^[1] If integral meters are not fitted, which is often the case in industry, the important welding parameters may be set up and checked with instrumentation developed to be used without disturbing the welder or the welding equipment. ^[2,3]

All physical measurements are subject to error, however, so it is still possible to obtain apparently conflicting results from different instruments when measuring identical parameters. Welding engineers or investigators need to be aware of the sources of error, be able to assess their magnitude, and know how to deal with them when presenting experimental results.

A first step in the right direction is to measure the variables correctly. A simple and practical guide to measuring arc voltage and current is given below.

Arc voltage

Where to measure voltage

True arc voltage is the potential difference between the opposite ends of the arc. The value normally but incorrectly described as arc voltage is that measured by connection to the workpiece and electrode at conveniently accessible points ( Fig.1).

Measurement at the welding set adds any circuit voltage drops to the true arc voltage, and is to be avoided at all costs. (The errors can easily exceed 10%, and may be greater than 25%)

How to measure voltage

Manual metal arc (MMA) welding

Ideally, one lead should be connected directly to the electrode core wire. In practice, connection to the electrode gripper or to a sharp conducting probe pushed through the cable insulation will add only a small, constant, error. The other lead is attached to the workpiece, preferably away from the workpiece connection. Take care to make good electrical contact by removal of oxide, grease, etc.

MIG and submerged arc (SA) welding

A brush in light electrical contact with the consumable wire makes the ideal 'live' connection but in practice the contact may have to be made at the contact tip or body of the welding gun.

TIG welding

Direct contact with the tungsten electrode may be made via a small screw in a threaded hole in the torch body, taking care to insulate exposed conducting surfaces. For MIG, TIG and SA welding, the 'earthy' connection is as described for MMA. Do not allow high voltages for are starting to reach the instrumentation.

Welding current

Where to measure current

Welding current should be measured in the cable which runs most directly between the welding set and the arc. ( Fig.2)

Where this is not possible, measure current in the return cable between workpiece and welding set ( Fig.3) taking care not to include current appropriate to any other welding set.

When measuring here, beware unintended current paths which may result in misleading readings. For instance, leakage current may flow through safety earth connections attached to the workpiece ( Fig.4)

Return current paths may also exist through traverses, work handling devices and cables attached between the workpiece and other welding sets. Leakage currents can be a large proportion of the welding current (as much as 50A in 170A has been observed) and may vary unpredictably during welding.

How to measure current

Resistive shunt and meter, for DC (Fig. 5) - Take care to use a matched combination of meter and shunt, ensuring that the shunt does not come into electrical contact with earth or any other electrical circuit. Do not use two or more shunts in parallel for high current, or two or more shunts in series for low currents.

Current transformer and meter, for AC only (Fig.6) - The current transformer is inductively coupled to the welding cable which is passed through its centre.

1. A matched meter must be used with it. It must never be used without a matched meter or other load, or it may be destroyed.
   

   
   
2. Hall-effect current transducer, for AC and DC - This type of transducer may be bolted into the welding circuit or clipped over a welding cable, according to type. It requires electronic circuitry to produce a signal for the current display, but gives electrical isolation from the welding circuit. Correct choice gives a high frequency response if required.

What indicating instruments to use

DC measurement

The current or voltage to be measured may be displayed on a moving-coil meter, on a digital meter or on a chart recorder.

Voltage

Voltage measurements require a suitable series resistor to obtain an appropriate full scale range. This is often integral with the instrument used, and may be switch selected for range ( Fig.7).

where ΔZ is the rms of the absolute errors.

b) If Z = AB or Z =

then

where ΔZ is the rms of the percentage errors.

In real life, if one of the errors that apply to a measurement is more than twice any other, then the largest error can be taken as the overall error.

Sometimes the errors in A and B are not independent of each other and more involved manipulation is needed to account for them.

Dealing with errors

Avoid parallax when reading meters and remember that the stated accuracy of meters and chart recorders is a percentage of full scale deflection (fsd) at reading, e.g. a meter scaled to 1000A fsd, of 2.5% accuracy, may have an error of ± 25A at reading. It is therefore important not to measure small values with respect to fsd, but to read in the upper half of the scale.

Magnetic or electric fields from welding sets/cables can adversely affect readings [4] so care should be taken in positioning instruments. Instrument tolerances are quoted at 20°C under standard test conditions. A change of temperature of 10°C may contribute a further 1.5% to the error of the readings of a 1.5% accuracy meter, so this must be guarded against, as must operation at an angle other than that for which the instrument was designed.

Quantities which are to be compared with each other should always be quoted with an error band, e.g. 94 ±4 Ω (or ±4%). Thus 94 ±1 Ω differs significantly from 97 ± 1 Ω, while 94 ± 4 Ω and 97 ± 4 Ω are probably not different 'within experimental error'.

The error band is particularly important when comparing a measured quantity with a theoretical calculation, a 'known' value or with another independently measured value.

An unnecessary number of decimal places should not be stated in a final report. For example, if a voltage measured as 5.32V is subject to an error ± 0.02V, it should be rounded to 5.3V. The errors themselves do not have to be quoted accurately; one significant figure is usually sufficient. They can be expressed as either absolute or as percentage errors.

The Welding Institute makes a range of instrumentation for arc welding. Members with specific enquiries should contact J A Street.

References