Wire EDM cutting and Hydrogen Embrittlement 1

[israelkk]

We have Custom 465 parts heat treated to H-950. Then wire EDM cut in few places. When loaded the part break into many pieces, in a very brittle fracture. We suspect hydrogen embrittlement although there whs no checmical cleaning, pickling, or passivation. After baking another part in an oven for 24 hours the part was tested and found to be OK. Is it possible that the EDM procces produced hydrogen that penetrated the base metal.

Any article or reference will be appreciated.

 

[unclesyd]

israelkk,
Can you come back with the thermal history of the part in question?

What type equipment was used to age the part?

How are the parts loaded?

Though we haven't processed very much 465 we have used EDM on thousands of parts made from 455 and 17/4 without any problems. I have used ECM on PH Steels without any problem.

I have never ran across any reference to EDM causing or contributing to H2 embrittlement.

 

[israelkk]

unclesyd

Thank you for your post. I do not have answers to your specific questions yet. However, the H950 heat treatment was done in a very respected aerospase company. The parts were analysed using SEM by a metallurgy lab with no conclusive answer what caused this problem. The part is a machined helical spring where the coils were cut by wire EDM after final machining and heat treatment.

 

[unclesyd]

I would try to get the thermal history and in the mean time I would contact the supplier of the material, I assume Carpenter, and appraise them of the situation. I would have a few broken parts for whoever you present with problem.

I also wouldn't depend on a bakeout procedure until I got to the reasons for the problems.

An unknown embrittlement mechanism in a high material is not good.
There are some known issues with PH steel's embrittlement but these are usually the effect of enviroment or temperature and time related.

Keep us informed as many are probably considering using the material.

 

[tbuelna]

israelkk-

At the company I work for, the M&P procedures for EDM require that all remelt must be removed after the EDM process. The remelt layer with wire cutting is very small, but it is still present.

Regards,
Terry

 

[swall]

Plus the remelt layer contains micro cracks.

 

[rorschach]

Was the conductivity of the waterin the EDM machine checked? I wonder if maybe some chlorides had gotten into the water.

 

[unclesyd]

What bothers me about this problem is that a bakeout helps the problem. If it was remelt or microcracking this procedure should have not effect on the subsequent part failure.
I haven't seen any temperature below 550°F have any effect on PH steels unless there is fairly rapid cycling over a long time before the any effect to manifest itself.

What was the bake out time and temperature?

 

[israelkk]

rorschach

Thank you for your response I will be attending a meeting tomorrow with the metallurgist, the designer and others and will ask this question and others. I happens that the heat treater did not follow the heat treatment with tensile test samples as it should be and there is no proof (such as oven temperature and duration chart) that can verify the heat treatment temperature was indeed 950F and not 900F. So we will have to update the part drawing and specifications to assure that next time it will be possible to trace every step since this is a critical part in the system.

 

[MintJulep]

heat treatment was done in a very respected aerospase company

and then

the heat treater did not follow the heat treatment with tensile test samples as it should be and there is no proof (such as oven temperature and duration chart) that can verify the heat treatment temperature was indeed 950F and not 900F

Never trust a reputation.