Examination for Ti Embrittlement 1

[moltenmetal]

We're working with a small used Ti gr7 pressure vessel. The surface shows extensive, blue-coloured heat stain from the use of external band heaters. The concern is nitrogen/oxygen embrittlement from long-term use with titanium temperatures beyond the ASME code limit of 315 C for the material.

Hardness testing has been suggested. We'd compare interior and exterior hardness I guess. But what criteria would there be for acceptance/rejection?

Any other tests that might tell us something useful in relation to the safe re-use of this vessel?

 

[CoryPad]

Penetrant testing would be advisable.

Regards,

Cory

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[metengr]

I would suggest using portable, microhardness testing and compare regions of the pressure vessel in and away from the heat tint regions. This could be used for comparative purposes.

 

[metengr]

moltenmetal;
You might want to obtain a copy of the design and fabrication handbook for Ti from the web site below for additional information;

http://www.timet.com/pdfs/ti-handbook.pdf

 

[moltenmetal]

metengr: we have both Timet docs already- the fabrication and the corrosion resistance references. Both are EXCELLENT references and I recommend them to anybody who is working with this material.

One of the notes in the Timet fab document is that the metal can be exposed to comparatively short anneal periods at quite high temperatures in an air atmosphere- ~ 650 C for up to 4 hours with an air cool- without resulting in damage to the material (i.e. O2/N2 pick-up leading to embrittlement). Is this accurate?

I'm not familiar with the microhardness testing methodology. Is it purely an indication of surface hardness? Is it a method which impresses a probe into the metal, or does it use another apparatus? Depending on the geometry of the test apparatus, we may be able to manage to test both interior and exterior surfaces- that should tell us enough because only the exterior was heated while in contact with air. The vessel is small enough that we can transport it easily to a testing facility. It's also small enough that destructive metal removal/sampling is not worth considering as an option.

CoryPad: LP to examine for macroscopic surface cracking is also a good idea. But what if the cracking just hasn't initiated yet?!

 

[metengr]

Is it purely an indication of surface hardness? Is it a method which impresses a probe into the metal, or does it use another apparatus?

Yes. Portable microhardness testing will result in a very small indentation. If you have surface embrittlement, like an alpha case, microhardness should be able to detect this surface effect.

I believe that if the surface temperature of the Ti vessel was below 1000 deg F, you are probably fine with the visual discoloration. I believe the PT is something to do to confirm there was no in situ cracking, as a result of harmful thermal stresses or gradients.

 

[unclesyd]

We have several Ti, grade 2, vessels that operate at 285C almost continuously for many years with no problems with embrittlement cracking.

We also have other Ti, grade 2, equipment that operates a temperatures of 120C-150C that have blue bands due to problems with heating elements. There has also have been no problems with the integrity of these vessels.

We also have a large Ti, grade 2, column that unbeknownst to us numerous welds we cleaned, blue bands, prior to delivery. The column has been operating for 20 years with no problems.

 

[moltenmetal]

unclesyd: glad to hear that. We have similar experience, but not with equipment in continuous service. We baby the material to avoid surface temperatures beyond the ASME stress value limit of 315 C, which is often taken as the MAWT for the vessel. Perhaps we're just being overly cautious, but surface temperatures under heater bands can get very high unless you're measuring and controlling them.

This vessel was used by people who heated the heck out of it until the internals got hot enough, without regard to exterior surface temperature- hence our concern.

The welding guides I've read, including the Timet reference, all say that during TIG welding it's essential to keep the shield gas on the weldment and HAZ until all parts are below 400 C. Not doing so actually does result in welds that fail bend tests due to embrittlement. Yet Timet says that you can do hot work at 650 C and an anneal at 788 C in air with air cooling followed by surface de-scaling. Something doesn't add up here!

 

[TVP]

The point to consider is surface descaling after exposure to elevated temperatures. Hot working or annealing above 650 C will result in diffusion of O or N into the Ti, and possibly the formation of alpha case. This can be removed by subsequent descaling however.

 

[moltenmetal]

TVP: on this vessel we're talking about a heavy heat tint rather than what I'd call "scaling". But as to the diffusion of N and O into the metal, showing up on the surface as heat tint but extending deeper into the metal- that's what I'm worried about.

You'd need to go a long way in increasing the amount of O and N in the base metal to cause it to embrittle, considering the permissible O and N content of Grade 4. But if you do have a harder-than-desirable case, can this result in crack propagation into the underlying parent metal resulting in a nasty ka-boom?

I'm still a bit unsure whether this is a real concern or whether I'm worrying over nothing! Have embrittlement-related failures caused by excessive longer-term surface temperatures been observed on units in service, as opposed to failures resulting from improper welding etc? I don't even know where to look to determine this! This vessel was built by a competent titanium specialist, so the welds were fine for certain. The question is- what's happened to the metallurgy since then?!

 

[metengr]

moltenmetal;
Can you have field metallography/replications of the pressure vessel surface performed by competent metallurgical/testing laboratory? I would do several regions for comparative purposes (heat tint regions). This information would be beneficial in answering your question about serviceability.

 

[moltenmetal]

metengr: pardon my ignorance of metallography- are you referring to a polish and etch followed by micrographic exam, or something else? Tough to accomplish given the (presumed) need for a flat surface to do this on. It's a very small vessel, and the only flat surfaces on it at present are not affected by the heat tint.

 

[metengr]

moltenmetal;
Yes. The surface is polished (as in metallographic sample preparation in the lab)using portable equipment and etched in the field. The size of the polished region can be as small as 3/8" in diameter. After the vessel surface is locally polished and etched, it is examined either using a field microscope (designed for field metallography) or a replica is extracted, and is viewed in the lab. This is a very useful tool for in situ examination of vessel microstructure.