Water Cooled Stanless Welding?

Date 04-09-2009 12:58 Edited 04-09-2009 16:28

While asking for PQR advice on welding stainless in another forum, Aevald
mentioned an idea that sounded brilliant, but I have some metalurgical concerns
about it. Here is the background:

I am developing a WPS for joining 304 O-lets to 316L schedule 10 pipe. The O-lets
require a significant amount of weld especialy along the side cheeks. I know my
total heat input needs to be monitored for sensitization concerns as well as distortion
issues (which I expect will become a huge production issue). Aevald suggested filling
the pipe with water to help cooling while we weld the O-lets. Intuitively this sounds
like a great idea but my question for this forum is: Does too much cooling on stainless
steel weldments present quench type issues that may alter or derade the microstructure?

Has anyone done this type of water bath welding on stainless?

EDIT: I just came across this exciting discussion from the archives and answers many questions.
Instead of deleting my question, maybe I'll just post this link for others to more easily find.
http://www.aws.org/cgi-bin/mwf/topic_show.pl?pid=72823;hl=cooling%20stainless%20welds

DOUBLE EDIT: I'm not entirely sure of the conclusion of the above thread or if a consensus was arrived at.
But my interpretation is that many folks do water cool their stainless welds successfully, and Stephen's
dissertation seems to show theoretical acceptance of the practice as long as sulfur content is low.

Other comments welcomed.
Thanks.

By ravi theCobra (**)

Date 04-09-2009 19:13

Take a look in the Welding Journal back around 1970. Many nuclear plant main steam lines were welded full of water,
HOWEVER the pipe was typically 2 " wall thickness . Are you dealing with something that heavy ?

By js55

Date 04-09-2009 20:40

If the primary solification mode of the SS is ferrite, or mixed, then the speed with which the water quench cools the part will inhibit transformation. This will result in driving up your delta ferrite volume pecent. If the primary solidification is austenitic there is essentially no transformation to worry about since there is no ferrite to transform and so your ferrite is nill, except I believe there is always a little formed as the segregated Cr achieves a delta ferrite equilibrium content. If memory serves somewhere around 1.5 or less Creq/Nieq ratio the primary solidification mode is austenitic, ~1.5 to ~2 is mixed and >2 is ferritic. But this is rough, the numbers have eveolved, and its been awhile and I don't remember what alloys were used to establish this.
With weld metals you will have a mixed solidification thats why they call it delta ferrite. Delta ferrite isn't formed, so to speak, its left over from primary solidification. The HAZ will be different.
But it all depends on how much cooling you impose and the exact material chemistry.
My opinion is that if your worried that much about sensitization use a stabilized grade, L grades and contact the O-let maufacturer to determine just how little you can get away with welding on the fittings you have. Its actually surprising how little weld metal those things actually need.
Give some consideration to water impurities as well. If you're that concerned about sensitization you might also need to be concerned about water impurities.
Also, I don't believe cooling it faster is gonna help distortion.
Other than that I can't think of any other concern offhand for rapid cooling of the SS alloys you listed.

By js55

Date 04-09-2009 20:44

That other thread is a good one. I had forgotten about it.