Using shielding gases for improved productivity arc welding of titanium

Dave Howse joined TWI in 1994 and currently carries out work on a wide range of arc welding and cutting processes within the Arcs section of the Arc, Laser and Sheet Processes Department.

Mike Gittos is a principal Metallurgist in the TWI Materials Department. His main areas of research have related to aspects of joining aluminium, copper, titanium and nickel alloys. He is currently responsible for electron microscopy services at TWI and, in 1993, he was awarded the A F Davis silver medal by the American Welding Society.

As the use of titanium becomes more common in welded fabrication, the search to find the correct welding process becomes critical. Dave Howse, Richard Wiktorowicz (Air Products) and Mike Gittos report on an investigation into the effects of the use of shielding gases for improved productivity of welding.

Recent reductions in the price of titanium have made the use of this material an attractive and viable proposition for offshore applications. Compared with steels, titanium alloys offer the potential for significant reductions in weight and improved corrosion resistance. However, the cost of welding titanium fabrications is relatively high, and they are susceptible to atmospheric contamination during welding. High standards of gas shielding must be maintained in order to prevent embrittled weldments and, combined with the inherent low deposition rate of the TIG welding process used for site welding, this results in low joint completion rates.

In other applications, such as the welding of stainless steel or aluminium, it has been demonstrated that the use of shielding gas mixtures containing helium can significantly increase the joint completion rate [1] . This can reduce the fabrication costs by more than 50% and similar benefits may be possible for the welding of titanium alloys.

This article reports on an assessment of the productivity, soundness and mechanical properties of welds made in titanium alloy using the TIG and MIG welding processes with a range of shielding gas mixtures carried out by TWI.

Approach

The material selected for the welding trials was a Ti-6Al-4V titanium alloy. This particular alloy is the most widely used titanium alloy, accounting for more than 80% of the titanium tonnage in the world. It is used mostly in the aerospace industry, but other uses include medical, defence and automotive applications [2] . 2mm thickness material was used for the TIG welds, and 5mm thickness material was used for the MIG welds. Both of the plate materials conformed to ASTM B 265-94 grade 5. The welding wire used for the MIG welds was 1.0mm diameter and was the extra low interstitial grade of Ti-6Al-4V.

The gases used were 100% Argon, Astec™ 30 (Ar-30%He), Astec™ 50 (Ar-50% He), Astec™ 75 (Ar-75% He) and 100% helium.

TIG Welds

The TIG welds were autogenous and made on the 2mm thickness material which had been prepared by shearing to give a square edge preparation. These welds were carried out using an OTC inverter Accutig power source and a ST-12 welding torch. The torch was modified to include a trailing shield. For experimental convenience, each different gas evaluated was used for shielding through the torch, through the trailing shield and for the back purge. However, the use of argon for the trailing shielding gas would not be expected to change the results obtained here. The welds were carried out using a welding current of 70 amps and constant electrode to workpiece distance. The travel speed was adjusted for the different shielding gases to give fully penetrating welds.

A procedure was first defined for the argon shielding gas. Subsequently, shielding gases containing increasing amounts of helium were then evaluated and any benefits in terms of increased welding speed were noted. A weld was judged to be visually acceptable if it gave a bright surface appearance with no apparent oxidation. These welds were then also examined by surface crack detection and radiographically tested. Tensile tests were taken from these welds, as were macrosections. In the absence of a standard specifically designed for titanium, all of the inspection was carried out in accordance with BS EN 288: Part 3: 1992.

Oxygen and nitrogen analyses were taken from a selected weld and parent material.

The scope of work carried out on the TIG welds is summarised in Table 1.

Table 1. The scope of work carried out for the TIG welding procedures

MIG welds

The MIG welds were made with pulsed parameters on the 5mm thickness material that had been prepared with a 70° inclusive vee preparation and a 2mm root face. These preparations were 'wet' machined and cleaned with acetone prior to welding. The power source used was a Fronius TPS 450 with a WFVR wire feed unit and a Fronius AW502 torch. The torch was modified to include a trailing shield attachment. For each different shielding gas evaluated, the gas was used for shielding through the torch, through the trailing shield and for the back purge.

The shielding gases and methods of inspection and testing were similar to those used for the TIG welds.

Once all five of the welds had been made, a further weld was made using argon at the end of the programme. It was necessary to alter the welding parameters slightly for each of the helium gas mixtures. The second procedure using argon gave a more stable welding condition that gave increased welding speed above that of the original condition.

Again oxygen and nitrogen analyses were taken from a selected weld and parent material.

The scope of work carried out on the MIG welds is summarised in Table 2. This table also includes the mean welding current and voltage used to make the welds.

Table 2. The scope of work carried out for the MIG welding procedures

Results and Discussion

TIG welds

The first batch of TIG welds passed visual and liquid penetrant inspection. The surfaces of the welds were 'bright' metal with no surface discolouration which indicated no atmospheric contamination. The radiographic examination of these welds showed internal porosity that failed the acceptance limits of BS EN 288. These welds were used to make tensile tests. The tensile results were acceptable, mostly failing in the parent material.

It was decided to repeat these welds in order to reduce porosity to an acceptable limit. These welds were made using the modified preparation and cleaning procedure detailed in Table 1. In addition, the backing purge flow rate was reduced from around 6.0 l/min to 2.5 l/min.

The welds produced acceptable visual, radiographic, liquid dye penetrant, tensile and macro results.

Photographs showing the typical visual appearance of the weld cap and root surfaces are shown in Fig 1.

In terms of productivity, Fig 2 summarises the change in welding speeds possible for additions of helium in argon. It can be seen that the relationship between welding speed and % helium addition is approximately linear. The 100% helium shielding gas giving a welding speed of 180mm/min compared to 100mm/min for the 100% argon shielding gas.

The oxygen or nitrogen levels in the welds indicated that no significant atmospheric contamination occurred during welding. The results of the oxygen and nitrogen analyses are shown in Table 3.

Table 3. Oxygen and nitrogen analysis of selected weld metal and parent plate

This work has emphasised the need for stringent joint preparation prior to welding and control of backing purge flow rate in order to control porosity. It was also apparent that the composition of the shielding gas does not have any effect on the level of internal porosity.

From the results of the tensile tests, it was noted that, although some of the welds fractured in the weld area, all of the results were acceptable.

MIG welds

All of the MIG welds made passed visual and liquid dye penetrant inspection. The majority of the welds were 'bright' metal with no surface discolouration, which indicated no atmospheric contamination. The weld made with 75% helium had a 'straw' colouring. However, it was considered that the degree of contamination was not sufficient to necessitate repetition of the weld. The radiographic examination of these welds was acceptable to the limits of BS EN 288. The tensile results were acceptable.

Photographs of the two weld caps showing visual appearance of the surface weld surfaces are shown in Figs 3 and 4.

In terms of productivity, Fig 5. summarises the change in welding speed for additions of helium in argon. It can be seen that the relationship between welding speed and % helium addition is approximately linear up to 50% helium addition in argon. Above 50% helium addition, the arc started to spread and this negated the advantage of the hotter arc ^[1] in terms of increased productivity. This arc spreading tends to give less peaky weld bead than the welds made with 100% argon.

It should also be noted that the MIG welds required a less stringent cleaning procedure in order to make welds that passed the internal porosity acceptance requirements of BS EN 288:1992.

The potential of the MIG process for increasing joint completion rates for titanium alloy welding has been demonstrated. Helium additions to the shielding gas are beneficial, both in terms of increasing the welding speed, and improving stability and thereby decreasing weld spatter.

Conclusions

• The welding speed for TIG welds made on 2mm Ti-6Al-4V with argon shielding can be increased by approximately 20% for each 25% addition of helium up to a maximum of 80% increase in welding speed for 100% helium shielding.
• The welding speed for pulsed MIG welds made on 5mm Ti-6Al-4V with argon shielding can be increased by 50% for a 50% addition of helium to the shielding gas. Helium additions of greater than 60% helium did not shown any further gains in productivity.
• Helium additions tended to increase the stability of the arc for pulsed MIG welding, giving less spatter compared to the 100% argon shielding.
• The addition of helium to the shielding gas in the TIG or MIG welds does not cause any deterioration in mechanical properties of the welded joint.
  

References