U-bend tubes heat treatment 1

[isoca]

Hello everyone,
A customer has asked us to anneal the bends of SS 304 U-bend tubes. As is often the case, he has not specified the treatment conditions.
The treatment will be carried out using electrical resistance equipment (joule effect)
For now we know that it must at least reach 1040 degrees Celsius, however I am not very clear about the time.
As indicated by ASME section VIII, the minimum time is 10 minutes, which seems excessive to me. I have carried out some tests and have observed that 20 seconds (or less) is enough to obtain a structure free of deformation and with an acceptable oxide thickness without the need to remove it.
Anyone know any standard or specification that specifies parameters or give some guidance for this type of treatment?
Thank you very much

 

[EdStainless]

Did you make the tubes?
If not then I would suggest that you not do this.
You have the min temp correct, but there are a lot of details.
The Cu contact shoes should be Ni plated and water cooled.
You have to inspect each ubend for any signs of arc damage and reject those.
The ID of the tubes needs to be inert gas purged.
The tubes need to be supported so that the bends don't sag.
We used 10 sec at temp, and we used three optical pyrometers to assure uniformity. You can heat fairly rapidly to about 800C, but then you need to taper the power off because the inside of the bends will heat faster and you don't want to over shoot.
You need to gently cool at first, light fan circulation. Otherwise the residual stress formed by rapid cooling will be greater than the ones from bending.
You need to dimensional inspect each bend to assure that it is in tolerance (see A688).
And then each ubend gets Hydro tested and the legs cut to final length (the ends get damaged in the process).
Are you putting these straight into a bundle or will they be packaged for transport? The boxed must be built carefully to protect the bends and not flex them.
None of this is obvious or easy.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[mfgenggear]

holy cow not a typical annealing process.
Ed is this by experience or is there a specification for requirements?

 

[SnTMan]

None of this is obvious or easy.

This is true. But, fact is, it is commonly done. One illustration of the technical sophistication present in an apparently mundane industry.

Regards,

Mike

The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[EdStainless]

A/SA688 would usually be the governing spec, but it takes a lot of reading between the lines to get the details all correct.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[isoca]

Dear,
First of all thank you very much for your responses.
I have some experience in this type of treatment, mainly in nickel alloys. I also know that many times it is requested without being necessary (generally is requested just in case...)
The problem I have now is that the client requests it, but he questions the HT time and asks me for a reference that supports the proposed treatment conditions.
The A688 does not indicate anything about time and the only reference I found is ASME VIII which indicates a time that seems to me not adequate (in fact if I follow the ASME guide it should not be performed). Hence my query arises.
Thank you again

 

I can endorse Ed'S advice.
In making these sausages there's much more in the way of not just technology but know-how than meets the eye. Which is why for example, tubes for nuclear steam generators come from an extremely short list of suppliers.

"Everyone is entitled to their own opinions, but they are not entitled to their own facts."

 

[EdStainless]

The argument in ASTM is that different processes may require different times. Since this is just a stress relief operation and it is being done at solution anneal temps the response in very rapid. We could never do too little, as long as the entire region reaches 1900F (under temp is just as bad as over temp) the results are the same. We picked 10sec because the procedure needed to be repeatable.
I left a step out earlier, before you anneal wipe the contact and ubend off with solvent, otherwise you will burn in finger prints and such.

Yes, I agree that this is often not actually needed. But I have never been able to talk a customer out of it. Though we would often flat out refuse to do it on true ferritic grades and most duplex grades. The local degradation of properties is worse than leaving the strains. For some services we would do it (ones where the surface condition didn't matter). In many chemical process applications where there is a risk of external corrosion the oxide from the ubend anneal is a detriment to service. If they are worried about pitting and SCC then they chose the wrong alloy.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[0707]

isoca

You almost answer your own question but you must tunned it on the correspondent methalics Handbook.

luis