Hello we've had some issues with 304L cracking on thin gauage (14-20) stainless autogenous GTAW welds. I'm assuming the issue is due to segragation induced cracking and perhaps partially due to the nature of autogenous welds and the lack of dilution or cleaning of the weld puddle.
I reccomended the use of 308L filler rods and to stop fusing the weld surface, but the engineers and welders are baulking about the increase in production time associated with non fused welds.
I dunno if pieces keep failing tests it seems it's cheaper to add filler metal than repair weld.
Does this seem like an adequate solution? I've always been under the bias that fuse welds generally have more quality issues than normal welds especially for structural applications, but maybe that is incorrect thinking?
I am guessing you are correct, the centerline cracking is "hot cracking" or solidification cracking associated with segragation of low melting temperature something or other, probably sulfur. But in addition to the susceptable material chemistry, I suspect you have other factors that tend to produce the cracking, like weld puddle geometry and weldment stress. For instance, maybe a large, wide, concave puddle made at high speed, and involving a constrained circular joint, is more likely to crack than a small, slower speed, non constrained weld joint. So, maybe an adjustment to welding parameters can help minimize cracking without having to add filler.
Something must be going wrong here with your process because it is not like 304L to crack when fused properly. What kind of joint are you fusing? What is the part being subjected to when in use? Are you going for full penetration with a purge? Are you going with Partial joint penetration? Is the weld going concave on you and sugaring on the backside? Even then it is still hard to get it to crack. Sounds like a PJP joint that is being overloaded and then cracking or is this just on your test pieces?
Some thoughts.
Because the chemistry of stainless steel base metals is not the same as filler metals you will not get as much ferrite to help prevent hot (centerline) cracking. Base metals undergo a different manufacturing regime with differing priorities, ferrite content, to eliminate hot cracking, is not one of them.
304L BM Cr = max 20
308L FM Cr = max 21
304L BM Ni = max 12
308L FM Ni = max 12
Higher Cr equiv. for the filler.
In actuality its usually more pronounce than this.
Centerline hot cracking is generally an issue of 'tramp' elements being pushed along the solidificaiton front until there is no place to go, the last solidification point, i.e., the centerline. With all this stuff, S, P, and others, the microstructure is not able to hold itself together under the final cooling stresses.
It has to be treated as if you are welding a fully austenitic alloy. And it is. Or at least very close depending upon cooling rate. If your going to continue autogenous welding you're gonna need to reduce your heat input. This will tend to increase ferrite as well.
Segregation will be more a corrosion issue than a mechanical one. Besides, your not gonna see much segregation in single pass welds.
And if I may make a minor correction, there isn't a lack of dilution, an autogenous weld is essentially 100% dilution.
if you go to a filler then 308L is the choice.
Is this a machine application? In which case it should be little trouble to adjust your travel speed to assist in controlling heat input, since travel speed is the biggest factor.
In addition to the other excellent questions/ideas already presented, how are you cleaning your parts? Is there still oil from the mill or machining? If you do have some contaminants on the surface, that contian items such as P or S, that could contribute to cracking. Also check the MTRs for the base metal to make sure they have low P, S, and other residual element content.
Thank you all for the input that helps a lot, My post was a little vague I suppose the addition of 308L filler metal was not my recommendation, for brevities sake I'll attach the recommendation I made to the head engineer. I'm a welding engineering intern in a company with no welding engineer on staff, all weldments are designed by mechanical or engineers in other specialities with a very limited knowledge of welding design theory.
To clarify this is not code work, but I've been pushing hard to make parts of AWS D code shop practice, such as the section 5 which deals with fabrication standards. I suspected weld cleaning was an issue as there isn't tight housekeeping policies in place. The matarial is laser cut then preformed on hydraulic brakes and presses. I've seen evidence of pre cleaning with alcohol, there is no different grinders for carbon steel or Stainless but besides alcohol I've seen no evidence of widespread edge dressing.
The welds are PJP all position GTAW outside corner, inside corner (fillet) T, lap, and square butt. I would say the centerline cracking issue is most prevalent on fillet welds especially those on inside corners and lap joints. I spoke briefly about parts in restraint and recommended weld order that would help reduce that.
The stainless forms are welded to mild steel framing and as such are in restraint.
The chambers in test are held to I believe a few psi and then checked for leaks over a 24hr period. There is no inspection regimine for the welds themselves in terms of pass/fail metrics.
The service conditions are any where from room temp to 1300deg F and up to +- 7psi. Thermal cycling has been an issue but I don't have any information on service failure of welds which I believe is low.
attached is the an exact copy of the reccomendations I made (before I wrote this post) along with a picture of a typical joint detail along the mating edge of bent. I just took a great pic of a joint I thought was particularily bad, but I left my usb cord at home :(
You are not adding filler metal to inside corner fillet welds? In that case, it seems like there is a good chance that you are just geting undersized welds for the restraint conditions imposed on the welds.
Autogenous welds in 304 stainless often results in a low ferrite number making the weld susceptible to hot cracks due to the presence of very low quantities of melting point constituents.
Since I'm sitting in a hotel room in Georgia, I don't have all my reference materials with me. However, since you mentioned D1.6 in your write up, you might want to look in the back of D1.6 in the annex. There is a WRC diagram where you can plot the nickel and chrome equivalencies and predict the ferrite number. There are some recommendations for the ferrite number range you should target.
The use of 308 filler metal will put you in a much more favorable FN range than the autogenous weld will.
Best regards - Al
Of course you can get rid of the cracking with the 308L, but lets get to the bottom of why it's cracking in the first place because it can be done successfully without filler. Have they always been using this technique without any problems in the past? If you can, lets see some picturs of these welds that are cracking and some that are not.
Kix is on the right path I think. Even if the autogenous weld metal is fully austenitic, fully austenitic stainless steel welds are done by the thousands every day. Some even with added 'crackers' such as Cb, or Cu, or both, like Alloy 20. Given the fully austenitic nature of the weld (or at least very low ferrite), and the possible restraint, I think your acceptable parameters might be quite narrow, but I'm not convinced that the chosen method need be abandoned. If your not using a machine process however, it may be tougher. You just can't control your travel speed that well.
I'll get on the pictures if I can. The cracking seems to be an issue over the years where parts will fail to hold pressure, or fail during the first thermal cycle. No one can seem to give me numbers on the frequency, but I was guessing part of the problem was excessive restraint or undersized "seal" fillet welds in combination with poor metal prep. and when the stars aligned, that combined with the occasional low melt point inclusion would equal a crack.
thanks for all this information I'll notify the test department to send me a photo the next time it occurs
By sid mehta
Date 08-07-2008 19:24
Edited 08-07-2008 19:27
In autogenous weld gap between SS plates in butt joint should not be more than 10 % of the plate thickness. for sheilding gas use argon 98 % with hydrogen 2 % . This will improve fluidity of weld metal .
I would look at the crack is cross-section. If the crack is straight down the middle it is likely to be due to excessive restraint. Solution would be look at the design or weld sequence.
If the crack is zig zagging or you multiple small cracks that point towards the weld centerline, you have an issue with lack of ferrite and some contaminants; solution would be to add some 308.
Girish
http://www.welding-consultant.com
thanks for that, the next time I pull a cracked unit I'll macro it
