No fume in the TIG process?

TWI Bulletin, January 1985

by Janet Moreton

Mrs Janet Moreton, MA, is a Principal Research Chemist in the Chemical Laboratory of the Materials Department.

Experiments to determine fume emission rates for TIG welding show that they are low compared with MIG welding - a result which could have important implications when considering control of welding fume.

It has been said for many years that tungsten inert gas (TIG) welding produces minimal fume. Comparative measurements were made between metal inert gas (MIG) and TIG welding as early as 1956 by Vorontsova et al, [1] who stated that ten times more fume was found in the welder's breathing zone with MIG than with TIG welding.

It is now known to be essential to assess the fume potential of a process by means of a fume emission rate measurement of the Swedish Fume Box type. [2] However, as far as is known, no measurements of the fume emission rate of TIG, or hot wire TIG [3] processes have been made in the Swedish Fume Box. It has been shown that it is possible to apply the Swedish Fume Box technique [2] to both flux cored wire and to MIG welding. Therefore it was considered feasible to run the TIG arc in the Swedish Fume Box, the only expected problems being in assessing the very low weights of fume produced.

Experiments were conducted with mild steel and stainless steel TIG consumables, and also with a manual version of the hot wire TIG process. Emphasis was given to use of stainless consumables, because of the concern which has been expressed [4-6] over the toxicity of chromium in fume obtained during welding of stainless steel.

Experimental

Experiments were carried out with both TIG and hot wire TIG processes.

Firstly, normal TIG welding was performed with a 1.6mm mild steel consumable to BS 2901, Part 1A 1983 17, and with a 1.6mm stainless steel consumable to BS 2901, Part 2 1983 316S92. Secondly, manual hot wire TIG welding was carried out with a 1.2mm stainless steel consumable to BS 2901 Part 2 1983 316S96.

Details of both the TIG and the manual hot wire TIG experimental set-up are given in Table 1. In both cases argon shielding gas, and a W-2%Th nonconsumable 6o° vertex angle electrode were used. In the second series of tests, manual hot wire TIG equipment was employed, because of its convenience for use with the Swedish Fume Box, and because only with manual operation is the welder close to any fume evolved over a significant period.

Table 1 Equipment and welding conditions (non-consumable electrode W-2%Th, 60° vertical angle; shielding gas Ar 8 litre/min)

* mean welding current, including wire heating (40A, 2.5V)

Histograms showing the magnitude of TIG fume emission rates in relation to MMA, MIG and FCW processes: a)mg/sec;

b) mg/g.

Automatic hot wire TIG equipment is more common, but the manual set-up would be expected to give the same order of magnitude results.

Tests were made to the requirements of the Swedish Standard [7] using horizontal-vertical fillet welds on 250mm long: a) 13mm thick mild steel; and b) 10mm thick stainless steel. Full details of the Swedish Fume Box tests are given in ref. [2] . For the TIG and hot wire TIG tests, the front aperture was replaced by a rubber curtain, containing two adjacent hand holes for insertion of the consumable and the W-2%Th electrodes.

For each wire, one test was made to collect a fume sample suitable for analysis, on a Whatman 41, 240mm type filter paper. A further three tests were made with glass fibre papers (type GF/C, 240mm), which were suitable for gravimetric tests. The glass fibre filters were weighed before and after welding, the weight difference giving the amount of fume collected. Test plates were also weighed before and after welding, to give the weight of deposited metal.

Arcing times (about 3min) were measured electronically. Sampling was continued for a further minute after cessation of the arc to ensure evacuation of all the fume from the box. From these data, it was possible to calculate the fume emission rate (FER) in terms of

a) mg fume/sec of arc time;
b) mg fume/g of deposited metal.

Chemical analyses of all plates and wires used are given in Table 2. Table 3 shows the CrVI amounts in the welding fume, the sample being too small per unit area of filter for a total analysis. Fume emission rates are reported in Table 4, and analytical methods for chromium described in ref. [8] .

Table 2 Chemical analyses of plate and wire, wt%

Table 3 Chromium VI in welding fume

Table 4 Fume emission rate results

Results and discussion

The welding fume emission rate (FER) for a typical TIG process, using either mild steel or stainless steel consumables, and for manual hot wire TIG has been found to be very low ( Table 4). The FER for TIG (0.1 mg/sec, tests 7832/38/39) is at the limit of detection for the Swedish Fume Box technique, governed by the minimum weight of fume which can be weighed reliably on a 240mm diameter filter. (The filter weight is about 2.5g and the minimum weight of fume collected is 0.01g). Typical FERs for TIG, and hot wire TIG, expressed as mg/sec are thus 30 to 100 times lower than typical figures for the MIG process. [9] This is even smaller than the factor of 10 difference predicted by Vorontsova et al, [1] based on breathing zone measurements. Compared with the range of FER values from MMA, MIG and FCW processes, FERs from TIG are 20 to 500 times lower, as measured in mg/sec.

If one considers the FER measurement in the units of mg fume per g of deposited metal, then the values for TIG are closer to the lowest figures for MIG and MMA. The high metal deposition rate for hot wire TIG increases the weight of metal deposited by a factor of three (in this case) compared with TIG, and thus reduces the FER from 2 to 0.8 mg/g.

However, it is the rate of fume emission per unit time which is more pertinent to a consideration of the amount of fume potentially available to be breathed by the welder. Thus the amount of fume evolved during the TIG process (normal or hotwire) has been proved to be negligible compared with MIG, MMA and flux cored wire (FCW) welding.

Chemical analysis shows the Cr VI amounts to be <0.1 %, compared with average figures of 0.3% Cr VI in TIG fumes obtained by Hansen [10] on background samples. Because so little fume is evolved in total the amounts of Cr in the fume released into the workshop atmosphere are also negligible.

In the face of the present Occupational Exposure Limited (OEL) for total chromium of 0.05 mg/m 3 for hexavalent chromium (currently recommended 8hr time weighted average value [11] ) the low fume TIG process or the relatively new hot-wire TIG process is much preferred, in terms of the low particulate fume emission.

In the present work, there has been no measurement of the pollutant gases generated by TIG welding. The most important of these is ozone, which cannot be measured readily in the Swedish Fume Box, as the generation rate is affected by the enclosing walls of the box, and the ozone is destroyed by passage through filter paper, or contact with surfaces. Undoubtedly, ozone remains a problem during TIG welding, and it is well known that the characteristic sharp smell can be detected during TIG operations.

Worrall [12] identifies TIG and MIG welding of stainless steel using Ar gas shield as two of the critical process/material combinations for ozone production. It was considered [12] that local exhaust ventilation was an essential part of ozone control strategy, provided that the exhaust duct (of adequate diameter and flow rate) was mounted above the arc.

Summary

It has proved possible to measure particulate fume emission rates (FERs) of the TIG and manual hot wire TIG welding process in the Swedish Fume Box. These tests confirmed that the FER values for TIG are very low, and typically 30 to 100 times less than those for MIG welding, as measured in mg fume/sec of arcing time. FERs for TIG are 2-5 times less than MIG as measured in mg fume/g deposited metal, but for manual hot wire TIG, the equivalent figure in mg/g is 4-10 times lower than MIG.

If it should prove necessary in future to limit still further the amounts of chromium in fume (particularly the more toxic Cr VI) emitted during welding, then this work points to the possibility of replacing MIG with hot-wire TIG. The latter gives similar deposition rates and similar ozone emission, but has the distinct advantage of reduced FERs with consequent reduction of potential welder exposure to chromium in stainless steel welding fume.

Acknowledgement

B Males, G Hutt, M Millson and M Churchley in the Arc Welding Department and N Jenkins, G J Carter and S E Day in the Chemical Laboratory gave advice and assistance.

References