CuNi to 304L Stainless steel.

By Kix

Date 08-13-2010 19:44

Well, I'm back at it again and having some more problems with this seam. The joint is a piece of 3/8" thick 90/10 CuNi plate welded to 304L Stainless steel 3/8" thick plate. Both are beveled to 30deg and have a 3/32" gap between them with a 1/4" thick 90/10 CuNi backing strip. This seam is welded in the 1G position. The filler rod is Mil-RN82 which is some kind of Inconel rod. Now for the problem, crack, crack, crack.
Here are the steps. 1 we fuse the root into the backing strip with no problems as of yet. We then add filler to the CuNi side kind of buttering the CuNi bevel face and putting some down in the root at the same time. Still no problems as of yet. Next we run a pass next to that one tying into the stainless plate. Boom that's when we see some very small micro cracks. When you puddle up on the cracks they don't even puddle like there was a crack even there. So we weld over it again to see if any of the crack opens up, which it doesn't. After that, when the plate is up to temp, we don't ever see any more cracks. I'm thinking that the 2 heat conductivities of the 2 different base materials are stress cracking the weld on the stainless side until both plates come up to an even temp. What is everyone's thought's?

Kix

By 803056

Date 08-13-2010 20:11

We tried to weld a copper silicon alloy to 304 stainless and the weld blew out of the sample like shrapnel when we tried to bend it.

It didn't surprise me. Copper and iron do not play well together.

My client is trying to weld copper nickel alloy to carbon steel now, but the most recent test plate failed RT due to several transverse cracks. The cracks are apparently subsurface because I didn’t pick them up when I did PT on the welded joint before sending out to RT. I haven’t tried to PT it again since it came back from RT.

Another application involved copper silicon to carbon steel. The unit stresses are very very low in the application. The customer allowed us to use a larger than normal bend radius for the copper silicon alloy to carbon steel. The tensile test results were acceptable. We have not gone that route with the copper nickel because we haven’t been successful with the RT yet.

The thermal coefficient of the two metals are different and the filler metal's CE is different from either base metal. That can be a problem. However, the problem is more fundamental than the thermal coefficients of expansion; copper has very limited solubility in iron, hence the problem.

I haven’t convinced my client yet, but I think the solution is going to involve using a transition layer of nickel alloy deposit on the carbon steel and then welding it to the copper alloy. Two layers of nickel alloy against the steel should keep the iron content in the weld low enough to be successful in making the final weld to the copper alloy.

Best regards - Al

By OBEWAN

Date 08-13-2010 20:49

Copper is a known bad actor with stainless. It leads to severe cracking. My employer was trying to weld pure copper to stainless and we were ripping our hair out with cracking problems until I switched it to a silicon bronze tig braze weld.

By ravi theCobra (**)

Date 08-13-2010 21:12

I have two filler metals : Inco Weld A which is almost " universal " and Haynes 625.

I would sat that your best bet is the Inco - Weld

By Kix

Date 08-14-2010 00:13

The big problem is that the filler I'm using is what was qualified in the procedure. I'm just trying to get a few men through welder quals. Yeah, I'm hearing that the best rod to use on this application is the RN67 which is exactly what we are using. I never did try to butter the stainless side with the ERNiCr-3(RN82) rod first. I just went for the CuNi first because that's the side that always flows like crap when you have both the stainless and RN82 diluted together washing up on the CuNi. I will try this approach tomorrow and see what happens.

By js55

Date 08-16-2010 13:20

Marty hit the nail on the head. Cu does not like Cr. Buttering is the only safe way to go. I'm surprised you have a viable procedure in place already. I would question that. Because unless I am reading this wrong, even if you passed the Section IX quals you cannot be confident there is not cracked welds being generated.

By Kix

Date 08-16-2010 16:21

The Procedure is qualified to NAVSEA techpub 248/278 and i'm right along side of ya about being able to qualify this procedure and do it consistently in production. I'm hoping the RT results for the welder quals will let us know if we were able to eliminate the cracks.

By MMyers

Date 08-17-2010 13:42

"and do it consistently in production."

That would be my concern too. If you're having this much of a problem with getting the welder quals, I'd be tempted to take some side bends out of their samples just to double check that the procedure is good. It's not a fail safe, but it's a quick way to see which way is up (like would a re-develop and qualification of the procedure be a good idea).

By MBSims

Date 08-14-2010 02:12

Dilution is your enemy here. CuNi weld metal (RN67) can tolerate no more than 5% dilution with either iron or chromium. You're going to need to use an alloy which can tolerate higher dilution, but also use welding techniques to minimize dilution. The Inconel (RN82) can tolerate up to 15% copper dilution, and is compatible with the austenitic stainless steel. Use RN82 and place beads against the SS side of the groove first and weld with stringer beads only. Keep the wire in the puddle at all times and do not stop feeding wire. Minimize penetration into the CuNi.