Aluminum Welding: short circuit vs. spray transfer

By Kix

Date 03-10-2009 19:40

I wouldn' accept it with any year of D1.2 on 1/4" and 1/2" material. That sucker has got to be hangin on by a thread. Tell him to show you his Welding procedures qualification test records and his WPS's that proves his recipe for that process and thickness of material works. I bet he doesn't have any... You have to use the code that the contract specifies. If no certain year is specified in the contract, then they can tell you what code year they will use and then you can accept or deny.

By jarsanb (***)

Date 03-10-2009 20:22

I have to agree. I wouldn't even accept a WPS w/PQR. 'IF' is pretty much guarenteed.

By designrider (*)

Date 03-10-2009 21:27

Thank you for the feedback. Here is a bit more detail regarding the project that may shed new light on things.

The weld inspector is onsite, he says after looking at the welds and the fabricators equipment settings he says he doesn't believe short circuit was actually used. The voltage is set too high, plus the welds do not show globule transfer. He asked the fabricator to set up for a weld as he did on the framework, and the fabricator set up for spray transfer. The inspector thinks the fabricator just doesn't know the correct weld terminology (ie the difference between short circuit and spray transfer). However, the issue still stands, the WPS documents the fabricator has is outdated and shows "short circuit" specified. I am considering having the fabricator set up and perform a sample weld exactly as he says he did while the inspector documents all the details. Then produce that information into a new WPS and see if it is approvable (ie. a retroactive WPS).

The issue is that this is a large project, upwards of $50,000 in weld fabrication and materials. To reject this entire project will likely put the fabrication company out of business. Therefore, I ask, is there anyway to determine the quality of the welds to eliminate the doubt of whether short circuit or spray transfer was actually used? I know there are multiple forms of NDT, but which one(s) would give substantial evidence of the adequacy of 3/16"-1/4" fillet welds?

Ultimately, I do not want to adopt excess liability for this project just because a fabricator chose not to stay up to date with AWS code. However, I would rather not put him out of work if there is a relatively simple solution. Any advice is appreciated.

By Lawrence (*****)

Date 03-10-2009 23:35 Edited 03-10-2009 23:42

First off, as Ray recommended, you should get hands on and review the actual PQR's that support the WPS's..

If all of the electrical values (Voltage, WFS) are spray transfer values in the PQR and the WPS's generated from that PQR also represent spray transfer values than it's possible that somebody simply *accidentally* filled in a blank with the word "short circuiting" transfer (a typo)

If this is not the case and there is any question at all than you must think of your own liability first.

Even if everything checks out in the PQR/WPS's... it would be wise to do extensive NDT and also select random fillets to cut and etch so you can be sure fusion issues were properly addressed.

If somebody is going to go out of business for doing the work wrong... It should be the folks who did the work wrong... But it would be tough for somebody to go out of business for a typo.

If the PQR and WPS show short circuiting parameters and the on site inspector visually recorded spray parameters being used in production, than the welder was not compliant with the WPS... (scrap it, there was no process control at all) Diddn't the on site inspector have the WPS available to him when he was watching the work?

By designrider (*)

Date 03-12-2009 03:12

Your words are wise, indeed the deeper we dig the more we confirm that "there was no process control at all"...Ouch.

By Kix

Date 03-11-2009 13:43 Edited 03-11-2009 14:30

So far we know this guy has WPS's, but does he have the PQR's to back the WPS's? Remember that there is no prequalified status in D1.2 and you joints must be qualified. If he does have all the PQR's for the WPS's, then it was probably just a typo. Double check the WPS to what they are actually doing. You may in fact have a WPS for Short Circuit transfer for thinner material for a previous job. The guys on the floor sound like they know what they are doing and know to setup for spray on that thickness of material. In other words they might just be working without a proper WPS which happens all the time. IN this case, just have the CO qaulify some fillet procedures per (SEC 3 AWS D1.2) with the settings the guys are using on the floor. This will show you that the welds they were producing on your parts more then likely are quality welds beneath the surface. If they cannot qualify a procedure using the settings the guys were using on the floor, well, we'll just say you might have a problem.
You have to watch out for these kind of fabricators. I run into this a lot with the whole "Oh yeah, I have WPS's and WQTR's so I can build your parts". Then you go over there to review them and they don't have what you need. You see it a lot in carbon steel fabrication where someone came in and wrote up a bunch of prequalified WPS's and everything you need to be done has to be qualfied. Then they ask you, "how do you do that" and all you can do is scratch your head. This code work gets pretty harry sometimes and can get a company that doesn't know what they are doing into a lot of trouble.

By designrider (*)

Date 03-12-2009 03:09

Thanks again for the comments; your opinions are in line with ours. The issue is more than just a typo. It stands that their WPS is truly out-of-date. It is dated 1998, and has not been updated. The inspector was not present during all welding because continuous inspection was not required for the weld sizes. Thus, I don't think it can be adequately verified whether or not they actually used a spray transfer procedure (in conflict with their own WPS). In addition to all this, the issue was further complicated by the fact that the not all welders performing the work were qualified for all the weld positions used. At this point, we have rejected the project. We are requiring the contractor to replace the frames or produce "undisputable evidence that the existing fabricated frames can be verified as code compliant", we will negotiate a solution from there. They can spin their own wheels on this if there is a salvageable solution, but at this point I have my doubts.

If they propose a solution to grind out ALL welds, and replace them using a approved WPS and qualified welders, should I be suspicious of the effects on the base metal? In other words, will an excessive amount of re-work damage the mechanical properties of the 6061-T6 sections?

By Kix

Date 03-12-2009 12:48

Yes, If you keep putting the heat to and welding on the 6061-T6, you will be loosing some of the mechanical properties of 6061-T6. Just thought I'd add one more thing, WPS's do not go out of date. It just needs all the other things to back it up, like, WPQR's and what the welders are actually building in the shop. So in other words what your saying is, they showed you a WPS for something else.

By bozaktwo1 (***)

Date 03-12-2009 14:54

Check D1.2, section 7. I would certainly question the integrity of the base metal. Heat input from welding is a concern, as is aging of the metal. Re-welding will anneal the HAZ (probably already annealed to some extent), and widen it. Unless you can get the parts heat treated, there will always be a question about how efficient these weld joints are.