Trail Gas-Shielding Save the Day… Again

The aforementioned opportunity to once again apply this jewel of knowledge presented itself during lot certification testing of a special order of ER308L stainless steel TIG cut length. Thanks to our Automation Division’s expertise in engineering welding cells for the Robotic TIG process, we have, in the Specials Department, the capability to test cut lengths in a much more timely fashion than in times past. And, most importantly, we have the luxury of testing a reel of continuous wire on a robot as opposed to having to weld TIG test plates by hand enables us to improve weld quality.

For this particular special order of TIG cut length, our customer had imposed on us the requirement that the ER308L cut length must possess a ferrite number greater than 8 as calculated using the material’s chemistry via the WRC-1992 method. The reason that a fabricator would want to ensure a certain minimum level of ferrite in the weld deposit is certainly understandable. A little bit of ferrite in an austenitic stainless steel weld deposit helps to reduce the chances for “hot-cracking” when the welds are restrained, the joints are large, and when cracks or fissures adversely affect service performance.[2]

Not only did this customer wish for the electrode to have the characteristic of a minimum ferrite number of 8 as calculated by the WRC-1992 method, but they also had the same requirement for the undiluted weld metal as deposited on a ferrite “pad.” A further requirement on the weld deposit was that it must exhibit a ferrite number of 8 as determined through the use of magnetic measurement.

Much to our chagrin, the calculated WRC-1992 ferrite number of the wire was determined to be between 8 and 9. The obvious reason was formulaic: the chromium content of this lot ER308L wire tended towards the low end of the required range of 19.5 – 22.0 percent as required per the AWS A5.9 filler metal specification, while the nickel content was towards the high end of the range of 9.0 – 11.0 percent. The combined effect that the respective percentages of these two elements had on the ferrite number came as no surprise, as chromium is the most famous of ferrite formers and nickel is the most prominent of all austenite formers.

Because we were confronted with this relatively low ferrite number “straight out of the gate”, we were required to take extra precautions during certification. Simply put, there are other elements besides nickel that contribute to forming austenite, thereby reducing the ferrite number. Of primary concern – the one that caused the most consternation – was nitrogen. Our main focus had now become the necessity to prevent pick-up of nitrogen in the weld metal – otherwise there was a very real possibility that the ferrite number would not meet the customer’s requirements.

Regardless of whether or not the right welding procedure, gas nozzle diameter, and gas flow rate were utilized, there was still a chance that the lot of ER308L TIG cut length might not meet our customer’s requirements. This is especially true in light of a statement made in the Appendix of AWS A5.9 that there is generally a certain minimum level of nitrogen absorption during gas tungsten arc welding process – usually a gain of 0.02 percent. [2]

In this situation, this small percentage of nitrogen pick-up would have been just enough to keep the ferrite level below 8. Therefore, every provision allowable to diminish the absorption of nitrogen needed to be employed. So, after reading the background that I’ve outlined here, you can probably see clearly now where I’m going with this.

Heeding our lesson from the past, we made the decision to implement a trail gas shielding device. What better time to use it than now? And, since we had the advantage of being able to test the lot using Robotic TIG, a trail gas shielding device was much less cumbersome than if the test plate had been welded by hand. Ultimately, after going the extra mile to ensure our success, the ferrite level of the deposit as determined through magnetic measurement panned out to be just above 8. Evidently, we needed every last bit of help that we could get.

Upon performing a chemical analysis of the weld deposit, the nitrogen content of the weld metal was actually lower than that nitrogen content of the wire. We were absolutely incredulous. How was it possible that nitrogen could have emanated from the weld puddle? Thankfully, serendipity struck recently as I was reading the most recent edition of the AWS Welding Journal at about the same time that I was writing this column. In his Stainless Q&A column, Damien Kotecki states that nitrogen evolution from the TIG weld puddle is indeed possible and has been proven to occur.[3]

So once again, a trail gas shielding device saved the day. Undoubtedly, there will be more opportunities to pass along this jewel of knowledge once again.

1. Lincoln Electric Internal Application Engineering Report 70618001.
2. American Welding Society A5.9 Specification for Bare Stainless Steel Welding Electrodes and Rods.
3. Kotecki, Damien, Stainless Q&A, American Welding Society Journal, July 2012.