Anyone with any experience welding AL-6XN pipe and tubing

Date 01-31-2007 16:39

Unless something is new in the world AL-6XN will weld more like nickel since the weld metal most often used,to my knwledge is ERNiCrMo-3. It will match the strength level of the AL-6XN and will most likely have corrosion properties in excess of that alloy in most services. There has been some talk about the CB in the NiCrMo-3 being a concern wihen mixed with the N in the AL-6X(N) (CB nitrides and microfissuring). But to my knowledge it has not been related to any failures. We welded miles of this stuff with this filler for nukes.

By chuck meadows (***)

Date 01-31-2007 18:02

js55,
I agree with your response. AL-6XN is a registered trademark of Allegheny Ludlum, and is considered a Superaustenitic stainless steel. And, like you said, the 625 filler metals are usually preferred. The concern about the Nb in the 625 is that it possess a high affinity for nitrogen and will deplete the nitrogen in the HTHAZ resulting in a loss of corrosion resistance and toughness.

By js55

Date 01-31-2007 18:16

Thanks for the addition Chuck. And I forgot about the official Nb as opposed to Cb stuff too. I guess I'm just gettin old. And fill me in, to your knowledge have the concerns about toughness and HTHAZ N depletion been confirmed through any failures? The last time I was involved in the welding of AL-6XN was about 8 years ago.

By chuck meadows (***)

Date 01-31-2007 19:32

You know, I don't know of any documented failures of any in-service constructions. Actually, I think that the highest concerns about the Nb in the 625 filler metal is when it is used to join Duplex stainless steels to nickel based alloys. In those cases, the HTHAZ will result in loss of corrosion resistance and toughness, but also a high ferritic zone. I'm not sure that would be the case when welding a pure austenitic like the AL-6XN with a pure austenitic filler like the 625. In these cases there is no ferrite to be considered. As you and I know, nickel is added to steel, both filler metal and base metals, for strength. So, I'm not sure if it is really that much of a failure concern. If it was, it would definitely be in the HAZ, though.

By js55

Date 01-31-2007 19:41

It seems to me that I did read some stuff on ERNiCrMo-3 causing corrosion issues with Duplex. For the very reason you speak. It seems to me about 6 years ago was when the results started leaking out (pun intended) and the 625 recommendations, as fillers, started to fade away as well. Besides, I think these days matching duplex is an easier proposition than matching AL-6XN. And correct me if I'm wrong, more economical than the nickels. As a side, D10 should be coming out with their Duplex doc before too long. The draft is promising.

By chuck meadows (***)

Date 01-31-2007 20:04

Yes, the issues with using a 625, or any nickel alloyed filler metal containing Nb, when joining Duplex to Duplex is the increased corrosion resistance in the weld itself. But, the loss of corrosion resistance and strength along with the high ferritic zones in the HTHAZ need to be considered, too. Duplex's strength is resisting chloride stress corrosion cracking. There is a number of nickel alloyed fillers one "can" use with AL-6XN, but Duplex is pretty limited. I really never liked using 625 with Duplex because of the potential problems. I mean, Nb is used as a stabilizer at high temperatures, but Duplex will form sigma very quickly at high temps. Sort of a Catch-22.

By js55

Date 01-31-2007 20:31

Yes, and the Nb used as a stabilizer in SS is in ratio to carbon content which is far less than the amount used in 625. The high temp properties of the Nb in 625 really only kicks in at temps too high for duplex.

By chuck meadows (***)

Date 01-31-2007 21:37

Absolutely..Duplex is not recommended at temps. above 500F. It is very susceptable to sensitization and sigma. I totally agree with you. Duplex should never reach operating temperatures where niobium would be beneficial.

By GRoberts (***)

Date 02-01-2007 03:57

I agree that ERNiCrMo-3 is the "standard" filler for superaustenitics like AL-6XN. Most of my experience is with welding similar castings. The Nb-N that forms right near the fusion line also embrittles that small region where the N diffuses into the weld. It is normally ductile enough to not cause problems when welding does not recieve PWHT. When it does recieve PWHT, then the Nb really poses a problem. Bend and tensile tests will have problems. In those cases, I prefer ERNiCrMo-10. I actually prefer it in both cases, but it is not essential in the case without PWHT. If you are welding thick walled pipe, you might also want to consider the SMAW process to increase your deposition rate after the root and a couple more layers are completed with GTAW.

By js55

Date 02-01-2007 16:09

10 is a good choice, though if memory serves slightly more hot crack sensitive (maybe the W), but it eliminates the Nb problem.

By M-Squared (**)

Date 02-01-2007 19:13 Edited 02-02-2007 13:40

Although I have not been welding pipe, I have in fact been welding AL-6XN plate in both ¼ inch and ½ inch thicknesses. I have been using ERNiCrMo-10 in lieu of ERNiCrMo-3 as well. The big thing you must remember when welding AL-6XN is that you must use a filler metal that is overalloyed in Mo to maintain the best corrosion resistance. When using a filler metal that is not overalloyed in Mo, microsegregation of Cr and Mo in the weld will occur and reduce the corrosion resistance of the weld. Even thought the dash 3 filler is considered overalloyed in Mo (9% Mo) when compared to the AL-6XN, dilution must be considered. For that reason I chose the dash 10 alloy (13.5% Mo) and it has worked well.

Also, according to documentation from Rolled Alloys the heat input should be kept as low as possible. I used a heat input of about 33 kJ/in. Rolled alloys suggest 40 to 50 kJ/in but no higher that 50 kJ/in to help prevent hot cracking.

I have not used the GTAW process due to it's slow production rates. Instead, for the ½ inch thick plate I used the GMAW-P (Pulse) process with the ERNiCrMo-10 filler and a single-V groove using a 60 degree bevel no root opening with a backgouge. This process worked very well and we achieved travel speeds of 8 ipm and a deposition rate of about 5 lb/hr. Mechanical and RT test were very good as well as bead appearance. For the ¼ in thick plate I opted to look at PAW using the keyhole mode with the addition of the ERNiCrMo filler. I used 100% Ar for both the plasma and shielding gas. The weld was completed in a single pass using a square butt joint, no root opening, and a backing gas, also 100% Ar. This process worked extremely well and we achieved travel speeds of 7 ipm. Mechanical and RT test were excellent, not to mention a sweet looking weld on the back side of the joint.

For welding pipe with a wall thickness of 5/16 inch or less I would highly recommend looking at the PAW process. No joint prep, and welds completed in a single pass! Doesn't get much better than that.

By js55

Date 02-01-2007 22:17

I would suspect that Mo segregation would be greatly reduced with a heat input of 33kj/in. That's chilly. It would require an almost 30% seg for NiCrMo-3 to hit the nominal Mo composition of AL6XN. That's quite a bit. For 10 its over 50%. It would be interesting to see the seg differences between WM and HAZ. But who's got the money for those kinds of tests.

By js55

Date 02-02-2007 14:56

To expand on a point. I think the problem of Mo seg can be exaggerated a little. I may have to go back and research some more but I believe that with the viscosity of AL6XN there is a thin layer of an unmixed remelt BM in which case the Mo will seg as well. If it does then the BM Mo content will be even lower in certain regions than it would be in a weld metal of NiCrMo-3. In the end, I'm not sure that considerations of Mo seg have proven to be problematic in actual services using NiCrMo-3, since the argument would be that if the AL6XN is an overdesign of corrosion calcs then the NiCrMo-3 would be as well. John Grubb at Allegheny Ludlum might be able to clarify. I have always found him to be very helpful and extremely knowledgeable. This could very well be a case where metallurgical logic does not translate into service failures. But it makes for interesting chat room conversations.

By js55

Date 02-02-2007 15:06

Sorry, let me clarify. The question is whether Mo seg has proven problematic in service when considering the differerence between the Mo content of AL6XN and NiCrMo-3 (625). I would guess there have been corrosion failures related to Mo seg. But thiswould be based perhaps more on an understimation of the corrosion problem than a difference between 625 and 6XN.

By chuck meadows (***)

Date 02-02-2007 15:22

I agree about all the "metallurgical possibilities" and Mo segregation and things like that, but we could pick apart just about any base metal and filler metal combination there is. AL-6XN was actually formulated to compete with the 254SMO, and the filler metal for that is recommended to be a P12, or NiCrMo-3. Again, any combination will work with many base metals, so this is a topic for discussion in a forum like this. But, the bottom line is recorded failures, and I haven't heard of any failures using NiCrMo-3. Or NiCrMo-10 either. I think it is safe to go by manufacturers recommendations, who have done extensive testing on base metal/filler metal combinations.

By js55

Date 02-02-2007 18:09

Chuck,
I remember a similar discussion going on about Alloy 20 (320). There was concern about residuals involved in hot cracking. So they came out with the ELR version (and actually there have been some improvements on 625 over the years-what we have now is not your grandfather's 625). But If memory serves there were no reports of failures due to this mertallurgical concern over 320 residuals.
Bottom line though. Having these discussions and being aware of these issues, even if a failure hasn't occured is certainly preferrable to ignorance. And you are right on, I don't know that there is a perfect filler metal base metal combo when it comes to corrosive applications. NiCrMo-3 and 10 have their issues. The problem always has been that as soon as some problem is somewhat 'solved', the services get tougher (higher temp-more corrosive) and the application of the alloy is extended, the limitations exposed, a new resolution ifsformed and the cycle begins again.

By chuck meadows (***)

Date 02-02-2007 18:56

I hope I didn't give even the slighest impresion that these discussions were anything but good. I agree about the 320 and 320LR. I certainly enjoy discussing, and reading views from others. I cannot remember when the 625 chemistry is anything but what it is today. When did it change?

By js55

Date 02-02-2007 21:11

You didn't. That was actually a criticism of myself, since I have a tendency to 'remind' a discussion that there can be vast differences between metallurgical knowledge and service results. And, LR thank you, think I partially slipped into Ti ELI.
I think what they did was just refine their target alloying ranges, minimal but important. Refine them beyond what AWS/ASME allows in the specs and cleaned em up. We can control these things much better these days, as you well know, and I believe we do. Sure would like someone to chime in that knows a lot more than I. The history of alloy evolution is a history of problem solving and is actually fascinating.

By chuck meadows (***)

Date 02-02-2007 22:05

Sorry..My brain goes into neutral every Friday about this time. I gather most of my insight from our extremely huge R&D Dept. in Sweden, which is done strictly on stainless steel. Technology is changing almost daily and "old" formulas, such as 625 just to name one, are being used in "new" applications all the time. I guess nothing stays the same forever. I think that all these metallurgical testing is done largely as a precautionary measure than a "can't do" scenario. We know what will and what won't affect many alloys, but knowing these precautions do not always guarantee failures, which we both know. Stainless, being largely a niche application, needs the high degree of technology. Well, my brain just went fully in neutral. Have a great week-end.