Pulsed MIG - a closer look at productivity, spatter and fume

TWI Bulletin, September/October 1996

Paul Anderson joined TWI in 1990 as a Research Engineer in the Arc Welding Department. He has since moved to Business Development, with emphasis on obtaining European funding for collaborative R&D.

Graham Carter joined TWI in 1965, and for the past 12 years has been actively involved in work related to welding fume, its generation and control. He is the UK representative on CEN/TC121/SC9, WGl and WG2.

For continuous current MIG welding of mild steels, the effect of the shielding gas composition on the economics of welding and the level of fume generated is currently well documented. Paul Anderson and Graham Carter of TWI worked with Richard Wiktorowicz and David Young of Air Products to produce comparable data for pulsed MIG welding.

Pulsed MIG welding can offer a number of potential advantages over continuous current MIG welding, including:

• increased travel speed and higher deposition rate, compared with the dip/short circuit metal transfer mode
• lower heat input and reduced distortion, compared with the spray metal transfer mode
• less spatter and fume
• better bead appearance.

To achieve optimum performance from pulsed MIG welding, consideration should be given to the wire feed speed, travel speed, peak and background values of the welding current, peak time of each pulse and the pulse frequency. The optimum setting for each of these parameters will depend on the material to be welded and the filler wire, the power source, and the shielding gas. In practice, the base material is normally specified by the job, and the power source is determined by availability. Hence, when attempting to improve process performance, a fabricator may change only the shielding gas or the filler wire. ^[1]

A wide range of shielding gas compositions is commercially available for pulsed MIG welding of mild steels. These are argon-based, and typically contain a small proportion of oxidising gas to assist arc stability; ie 2-25% CO ₂ and/or 0-5% O ₂ . Welding with more than 25% CO ₂ in the shielding gas is not normal due to high levels of globular spatter and fume. This article reports on an evaluation of a range of shielding gas compositions. Comparative data are provided to show how the proportion of carbon dioxide in the shielding gas affects the process performance, the economics of welding and the fume emission rate during pulsed MIG welding of mild steel.

Process performance

The amount of weld spatter and the metal deposition rate obtained when welding with shielding gases containing various proportions of CO ₂ were determined for the same joint configuration: a 6mm leg length fillet weld in 12mm thick mild steel. An E70S-3 filler wire was used in combination with four different shielding gas mixtures. The optimised welding procedures are shown in the Table. It should be noted that CO ₂ shielding gas is not generally used in practice for pulsed MIG, it is only included for comparative purposes. The welding power source was a programmable inverter.

Table: Welding parameters

The welder commented that, even with the optimised welding conditions, the arc was slightly unstable when using the Ar-2% CO ₂ shielding gas, and with CO ₂ shielding gas, there was severe arc instability. The welder's opinion was that, for this application, argon shielding gas containing 8 or 15% CO ₂ was the easiest to use and provided the most stable and reproducible welding characteristics.

Welding using 100% CO ₂ shielding gas resulted in high levels of globular spatter, Fig.1, whereas the use of Ar-2% CO ₂ resulted in lower levels of fine spatter, which is related to the arc instability. A smaller amount of fine spatter was generated when using Ar-8% CO ₂ . Very low levels of spatter were generated when using the Ar-15% CO ₂ shielding gas.

All of the welds exhibited satisfactory penetration for this application. While comparisons are not straightforward due to the different welding conditions and resultant weld bead sizes, the weld beads produced with the Ar-CO ₂ shielding gas mixtures had shallow, wide penetration profiles; whereas the weld bead produced using 100% CO ₂ shielding gas had a more convex profile.

The highest deposition rates were associated with welding procedures and shielding gas compositions which resulted in good arc stability and low levels of spatter. For the test sample, the highest deposition rate was achieved with Ar-15% CO ₂ shielding gas, followed by Ar-8% CO ₂ and Ar-2% CO ₂ shielding gases. The lowest deposition rate was achieved with CO ₂ shielding gas, Fig.2. The deposition rate achieved during pulsed MIG is 35% higher than that achieved using the dip/short circuit mode of metal transfer, for comparable joint configurations. ^[2]

Overall, the preferred welding process performance was achieved using the shielding gas containing 15% CO ₂ , which is consistent with current industrial practice.

Economics of welding

The comparative cost of welding using different shielding gases is influenced by the deposition rate and any post-weld cleaning or dressing required to remove spatter. The reduction in labour cost when using a higher productivity, lower spatter shielding gas is typically much more than the additional gas cost.

In the test samples with a labour cost (including overheads) of £30/hr, the lowest cost was incurred when using Ar-15% CO ₂ shielding gas, followed by Ar-8% CO ₂ , where spatter removal was not necessary, Fig.3. Higher costs were incurred when using Ar-2% CO ₂ or CO ₂ shielding gas due to lower deposition rates, higher amounts of spatter and the subsequent need to dress the weld.

Fume generation

Personal exposure is variable, depending upon factors such as work situation and welding duty cycle. Consequently, unless personal exposure is measured under strictly controlled conditions, it is not a suitable parameter for measuring the effects of welding variables (such as the welding process, operating parameters and the consumables) on the quantity of fume emitted. The effects of welding variables on fume emission rate are normally investigated using fume box techniques, where the parameter of interest is the only variable. ^[6] Although emission rates measured by fume box techniques cannot be used directly to assess worker exposure, it is expected that conditions giving low emission rates will result in lower exposure than conditions giving higher emission rates used in the same working situation.

To compare the effect of different shielding gas mixtures on the fume emission rate, fume box techniques based on a draft European standard Health and safety in welding and allied processes were used. ^[7] Part 1 of the standard was used to determine the particulate fume emission rate, Part 2 for evaluation of the emission rate of carbon monoxide.

Part 3 will provide a method for measurement of the emission rate of ozone, but is not yet complete. Ozone was measured, therefore, by measuring personal exposure under strictly controlled conditions.

Particulate fume

Particulate fume emission rate can be evaluated by measuring the particulate weight in terms of milligrams of fume emitted per second (mg/sec), Fig.4, or milligrams of fume emitted per gram of deposited weld metal (mg/g). The unit mg/sec is generally recognised as providing the best indication of exposure, whilst the unit mg/g can be used to determine the amount of fume generated in the performance of a job. Interpretation of the results was the same using either unit of measurement. Providing there is sufficient oxidising potential in the shielding gas to achieve good arc stability, the emission rate of particulate fume increases as the proportion of carbon dioxide in the shielding gas increases. This is in good agreement with work which investigated flux-cored arc welding. ^[1,8]

Composition of the shielding gas gave rise to only small differences in composition of the particulate fume.

Gaseous fume

Carbon monoxide

The emission rate of carbon monoxide was measured using the units millilitres per second (ml/sec). There was a clear increase in the emission rate of carbon monoxide as the proportion of carbon dioxide in the shielding gas increased, Fig.5. However, it is considered unlikely that CO will be present in hygienically significant quantities except when 100% CO ₂ shielding gas is used. Although the figures are not directly comparable because of differences in the welding parameters, results showed the same order of carbon monoxide relative to carbon dioxide in the shielding gas, as measured in published work for breathing zone measurements. ^[9]

Ozone

It has been shown that for MIG in the dip/short circuit metal transfer mode, there is a reduction in concentration of ozone as the proportion of carbon dioxide in the shielding gas increases. ^[10] However, for pulsed MIG, the shielding gas composition had no effect on the ozone concentration in the breathing zone, Fig.6.

Summary

For the pulsed MIG welding of mild steel with E70S-3 wire, and Ar-2% CO ₂ , Ar-8% CO ₂ , Ar-15% CO ₂ and CO ₂ shielding gas mixtures, the following conclusions may be drawn:

1. Ar-15% CO ₂ shielding gas gave the highest deposition rate, the lowest cost per unit weight of metal deposited and very little spatter. A satisfactory weld was also achieved with Ar-8% CO ₂ shielding gas.
2. Providing there was sufficient oxidising potential in the shielding gas to achieve good arc stability, the level of particulate fume and carbon monoxide increased as the proportion of carbon dioxide in the shielding gas increased.
3. The shielding gas composition had no effect on ozone concentration in the breathing zone.
4. In agreement with normal practice, these trials confirmed that CO ₂ shielding gas is not suitable for pulsed MIG welding. This shielding gas gave the lowest deposition rate, the highest cost per pound of metal deposited, the highest emission of carbon monoxide and high levels of globular spatter.

References

Specialists from TWI sit on the CEN/TC121 subcommittee nine and its working groups, contributing expert knowledge to help draw-up the new standards.