Annealing 321 Stainless

[goucho]

I wasn't sure whether to post this in the steel forum or not. I would like to anneal 321 at 1900F and what to know how it will react. Here's my case:

We've been thermal forming a piece of 6-4 Ti between an inner ring of 321 and an outer ring of 410. The 6-4 piece is about 9 ft in diameter (in two halves). We're using the differing CTE's to cause an interference between the rings to form the 6-4 at 1350F. After an initial 10 cycles, the rings size are set and they stay pretty much the same diameter for another 50 cycles. After that they start to creep. The inner gets smaller and the outer gets bigger, thus reducing the interference which causes the 6-4 to be partially formed. We are now at ~120 runs and the part is out of contour.

In an effort to bring the interference back, we are going to be cutting the inner ring in half and basically shimming it outwards. Before we do that I wanted to try annealing the inner ring to 1900F for about 7 hours (it's 9" thick). I don't really care about fixing carbide precipitation since it is not in a corrosive environment. I'm thinking though that this should relieve the stress that's been built up from the repeated compression of the ring, hopefully causing the ring to expand back towards the nominal diameter. This way the initial interference will return and the part can then be fully formed.

However I'm afraid that since the part will soften up, it may cause undesirable effects when it goes in for the first interference cycle after the anneal. What should I be concerned with or what should I take into account? Should I try annealing the outer ring as well?

Thank you for taking the time to read this!
-Benjamin

 

[weldstan]

Creep caused distortion will not be recovered by annealing.

 

[FennLane]

If the deformation is caused by a creep mechanism will it also cause a residual stress or does any residual stress cause the creep?

 

[goucho]

Thanks for the reply! If this is true, is there any benefit to annealing this ring?

It's also interesting because we had tried this a couple years ago with an older Outer ring and apparently it shrunk in diameter closer to nominal. We never ended up running it any more though and it's been scrapped. Hmm... I'll have to look into this some more.

I'm also curious about FennLane's question.

 

[MagBen]

whatever caused the distortion, if it was permanent, it is unlkely to have the material back to nominal by annealing.
Replace material with high creep properties. A286 seems a good candidate, whose austenitic structure may make the CTE close to 321.

 

[EdStainless]

I answered in the other forum.
Double posting is considered bad form.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[goucho]

I apologize for the double post. I know that some people only look at one forum so I figured I might as well get as much visibility as possible. I would delete the other post, but there doesn't seem to be an option to.

Anyways, I am enjoying this discussion and I appreciate the responses. This discussion that the 321 will not expand through an anneal has me thinking about why the 410 shrank back towards nominal after we annealed it. Is there a density difference from any phase changes? It makes me wonder about the Inconel 718 that we process. When we age it at 1325F for 8 hours, cool to 1150F, then hold for the remaining time to get 18 hours total, the part shrinks by 0.0007" per inch. Then when we solution treat the part (if we need to rework it) we heat up to 1750F for at least 15 minutes, the part grows back to nominal (pre-aged) size. What is causing the change in size with the furnace loads?

EdStainless:
"annealing the 321 will only soften it.
If you are worried about residual stresses then try a stress relief, however your 1350F should be a very effective stress relief temp. so I doubt that residual stress is involved.
I don't see how any heat treatment will help your situation. Just make the mechanical alterations."

 

[weldstan]

410 SS will go through phase (crystallographic) changes (ferrite or martensite to austenite back to ferrite or martensite) which accounts for some of what you saw. Plus martensite exposed to high temperature will somewhat change its body centered tetragonal structure to more body centered cubic. When the austenite is transformed back to martensite it should get back to its original size excluding any plastic (creep) deformation. The 410 appeared to have experienced very little creep deformation.