Apparently, 15 years ago we had cracking problems with the weld joint in the field. Someone suggested using 312 to 316L filler, and put it on the engineering drawing, I assume for better ductility, and it has been there ever since. When stainless was used, the cracking problem went away.
I wonder about the welding parameters, and heat-treating specification from 15 years ago. Before there was a lot of attention focused on the cracking issue. Were they short-circuiting, did they omit the pre-heating? Did they reduce the hardness requirements, and/ or did the heat-treater temper the product back properly? Were the joints properly cleaned?...etc.
I don't have any historic data to work with, just testimony of the working products in the field. We start welding tomorrow, and I will document the welding procedure.
Currently, we are going to preheat the weld joint to 250 F, use 316 filler, GMAW, spray-arc transfer.
I will keep you up to date, if I stumble across any more data.
Cain
FYI,
THE FIRST THREE WELDS CRACKED, THROUGH THE CENTER OF THE FACE, EITHER DURING OR IMMEDIATELY AFTER WELDING.
WE ARE PUNTING AND RETURNING TO SMAW: E312-16.
I AM GOING TO TRY TO USE A 110 KSI METAL COR WIRE THIS AFTERNOON.
Can you try ER309 with the GMAW? The extra dilution you may have encountered while using GMAW instead of SMAW can change the chemistry enough that the weld deposit might not be austenitic any more. With enough dilution ER316 will become martensitic, which is more crack prone. With ER309, you will have higher alloy content, which is more tolerant of dilution and can remain austenitic. Is this a single pass weld? If so, you might want to check your bead depth-to-width ratio. A weld bead that is too deep for its width will aid in cracking as well. Especially in large single pass welds.
The 110 ksi metal core worked well.
I have new information. It seems the cracking problem with this joint has always been a fabrication issue, not performance. The problem is with post weld cracking. Apparently, the only filler that prevented this type of cracking 15 years ago was E312-16, SMAW.
I asked the question, did you ever try E11018? The answer; once, when we ran out of E312-16 and it worked well. That's why I went after the ER110 metal cor.
The drawing is also wrong; it calls for 312-316 stainless steel not E312-16.
We are getting there! :)
Very interesting about the differences between 309 & 316. Do you know why 312 SS is marketed as the filler for joining dissimilar metals? The high chrome?
This is a multiple pass weld due to SMAW: 7/8" fillet, 6" long on 6" centers. I put a 3/8" fillet down using the ER316L and got the cracking. I don’t know the exact depth of the weld, I was spray arcing around 27 volts and 300 amps.
You wrote: A weld bead that is too deep for its width will aid in cracking as well. Can you explain in more detail?
Thank you for your time!
My thought is that cracking at the weld [bead] centerline is a solidification cracking isssue. Dilution of the 4140 or [generic] A36 material into the sst weld metal can easilly be a cause for this (high C, in possible combination with elevated P, S, etc.).
I would not expect use of 309 to solve this sort of cracking problem since it is austenitic also. The point about using 309 to control steel dilution to prevent formation of martensite in the weld metal is well taken. Independent of a solidification cracking isssue, I would also support the preferred usage of 309 in this context.
What the historical nature of the cracking problem is has not been characterized. Cracking at the toe of the weld in the parent metal is a different cause than cracking in the weld metal. Which parent metal may have cracked is significant too. I would intuit the high carbon 4140 material to be susceptible to martensite formation int he HAZ, irrespective of reasonable preheats. I interpret the section thickness of the membes is "thick", and as such the concentration of residual stress from welding can be significant. I note that PWHT appears to be not indicated for your procedure. I expect there is martensite in the CGHAZ of the 4140 material, and [depending on chemistry] possible the A36 material also. Hydrogen asssisted cracking is a real possibilty in this context, cast materials may be more tolerant in this respect than wrought materials.
One therapeutic effect of using an austenitic stainless filler here would be for the weld metal to act as a sink for any process related hydrogen [SMAW typically has a higher H potential than GMAW]. Hydrogen held in solution by the austenitic weld metal is prevented from diffusing to the HAZ. I wonder that the specification of an L grade intends to compensate for diluted C from the parent material.
Clean absolutely all joint surfaces and filler materials. Chemical cleaning with a light solvent immediately prior to welding and after any interim mechanical cleaning is reccommended.
I suggest you conduct at least and audit MT or LP of the 4140 side weld toe no earlier than 48 hours after material reaches ambient temperature after welding.