Aluminum - GTAW

Date 07-28-2014 20:09

I was asked for some help in developing a WPS for a school. I don't have a background in GTAW or Aluminum, but I want to help them out and perhaps cut my teeth on this; I need to learn more about AL and SS. I've been doing some studying and have a theory, but I need some expert feedback.

They told me that they are trying to weld 3/16" 5052 with 5356 1/8" filler rod and they are using 1/8" purple tip non-thoriated 99.37% W and 100% Ar. They've tried to qualify a procedure a couple of times now, but they keep failing the tensile test. The joint is B-L2b.

What I haven't gotten yet is their operating parameters, but based on my studies my theory is that they are using too small of filler and tungsten and that they are probably traveling far too slow in an effort to fill the joint. The slow travel speed is increasing the heat input and compromising the strength properties. I think that my suggestion would be an increase in tungsten size and filler rod size, but if I can make the 1/8" work for them I'd like to, they have a lot of material and filler that has been previously supplied.

My best guess at this point is that if they gave some consideration to their travel speed and sped up simply increasing the number of fill passes and monitoring the inter-pass temps may be the answer.

Any feedback will be appreciated - thank you.

By 803056

Date 07-28-2014 21:07

The tungsten is big enough as is the filler metal.

The alloy is nonheat treatable, so time at temperature isn't a major issue, but still the weldment should be permitted to cool between weld beads.

Cleaning, cleaning, cleaning is usually the problem.

Do a search on the Forum. There has been plenty of discussions on the subject.

Best regards - Al

By Distilled (**)

Date 07-28-2014 21:28

I'll do some more research on cleaning - thank you!

By OBEWAN

Date 07-29-2014 12:11

Here is some information to consider - Alcoa (as one example) has two different forms of the same alloy, and the tensile properties are slightly different.

https://www.alcoa.com/mill_products/catalog/pdf/Alcoa_Insert_5052and6061_FINAL.pdf

Also, classic welding theory (for carbon steels at least) teaches that slower travel speeds sometimes cause larger grains - this might result in lower impact test values. I am not sure what the effect would be on tensile strength would be.

By Distilled (**)

Date 07-29-2014 17:23

Thank you OBEWAN - I'll definitely take a look at the link you provided.

By Plasma56 (**)

Date 07-29-2014 07:24

Hey Distilled. I have a theory as well. I just don't know if it holds up under scrutiny. But its based on my experience with the stuff so it might help.
My theory is your thinking out of the box different on how to solve the problem. Am I correct here?
Out of the box thinking different isn't so strange, it's just not what most are taught to do. I applaud you.

I would go larger tungsten and smaller rod however. The larger tungsten would carry current better and the smaller rod melting easier. Lowering heat input and gaining faster travel speeds. Wrap your head around that for a minute.
Try a blunter tip trunication on the tungsten and see what happens to your current requirements.

But that's not normally how its done? And your concerns were with tensile numbers.
It's being over liquified. Welded in a more heated fluid state then necessary, taking longer to cool and loosing some toughness strength as a result.
That's my guess. And it is a guess.

What they have is workable and typical, but you still have the welder factor. And when it comes to aluminum, welders typically have a tendency to be short on patience and high on current use, and weld everything like carbon steels.

Welding Aluminum does have a requirement for cleanlyness, but it's more likely to effect the welder in weldability efforts depending on power source and the welders skill level to the process so yes, it could be an issue effecting arc on time. Cutting fluids, carbon shears, contaminated disks or files, not failing to mention contaminated filler metals and dirty gloves are other risk factors to this concern.

Heat input is not only an effect of travel speed but of current used. A number of factors will affect that variable. Arc length, tungsten size, tip taper or shape, arc length, power source wave forms comes to mind.
You can use to much current and still move to slow. Too little current and move to slow.
The variables are many and you have left a few questions unanswered. But your on the right path buddy. Your thinking.

Good luck with your endeavors.

By Distilled (**)

Date 07-29-2014 17:34

Thank you for the comment Plasma56 - I am an 'outside the box' thinker, I am often invited to sit in on discussions and strategy meetings for that quality alone. It's not always a great trait, sometimes it results in me complicating things for people, but I've learned to keep my mouth shut and filter through some of the thoughts before talking. :wink:

I stopped by the school yesterday and we discussed the procedure for a little bit; the blunt tungsten tip was one idea that surfaced in addition to slightly faster travel speeds, an improved cleaning process, and no preheat. A CWI previously working with them told them to keep the Al temp between 250ºF and 350ºF for the duration of the weld-out. I also found out that they had been given some 3/8" material to use, so they were welding 3/8", then machining it down to 1/4" or 3/16" after welding and prior to mechanical testing. ??? Don't ask me..... :) I suspect that in addition to everything else that might have gone wrong they were probably having issues accessing the joint resulting in a long arc at the root pass(es) - it appeared to me that the root was the initial point of failure when I look at the coupons.

Based on my research and your guy's help I think I'm going to be able to help them solve this problem fairly easily. There are still a few unknowns I'll admit, but I have a much better understanding of what I'm looking for now.

Thanks guys I appreciate the help.

BJ