Temper Embrittlement 1

[swall]

Since the subject of temper embrittlement came up in the 4140 thread, I thought it might be useful to discuss temper embrittlement. I have worked in automotive and aerospace low alloy steel heat treatment and failure analysis for over 30 years and have never encountered a failure from temper embrittlement that I could document. If you look at the references like The ASM Handbook (Failure Analysis), Key to Steel and ASM "Principles of Heat Treatment", you will find conflicting statements concerning at what temperatures temper embrittlement can occur and at what temperatures it can be mitigated by rapid cooling. I posed the question to a mill metallurgist with over 30 years experience and he, like myself, had never seen a documented case. What do you guys think?

 

[metengr]

Swall;
I have seen more thermal embrittlement in the fossil power industry in a number of Cr-Mo materials. We have over the years seen 4140 stud material used in elevated temperature service for bolting on large steam turbines suffer a significant loss in toughness (studs simply snap on re-torque) as a result intergranular fracture along prior austenitic grain boundaries. This was not hydrogen embrittlement.

There are also instances of 1-2%Cr, 0.5-1% Mo alloy components (headers and link piping) at fossil generating stations that upon exposure to 800-1000 deg F have degraded toughness.

I am a firm believer in thermal embrittlement of certain alloys containing tramp elements.

 

[davefitz]

I had heard in the early 1980's that the US electric power iindustry had experienced several temper embrittlement failures of low alloy ( P11,P22) tubes in the 1950's and early 1960's. I had also heard of some tht occurred in spain in the 1970'sor early 1980's.

I have not read of the failure analyses or explanation- I was under the impression it was caused by "poisons" entering the alloy in the tube mill-levels of sodium, sulphur, potassium etc. that might be acceptable for carbon steel were contaminating the low alloy . Once discovered, it was quickly addressed as a QC issue, and the practice of documenting sources of batches of alloy tubing was implemented to allow quick searches for other contaminated tubes once a failure was found.

I suspect the failure of the P91 china main steam line of 2007 might be caused by a similar temper embrittlement QC issue.

 

[swall]

Lets not muddy the waters here with embrittlement due to long time service at elevated temperatures. I am specifically interested in hearing about temper embrittlement from quenching and tempering low alloy steels and having the embrittlement occur in ambient temperature service.

 

[CoryPad]

Interesting subject, especially the nuances between Tempered Martensite Embrittlement and Temper Embrittlement. Krauss covers this well in Steels: Processing, Structure and Performance, chapter 19. You can use Google Books to preview the sections on these phenomena (pages 396 through 404).

Temper Embrittlement is more likely in large parts like shafts and rotors for power generating equipment. My experience in automotive is similar to yours, swall - nothing I would definitively assign to Temper Embrittlement.

 

[arunmrao]

swall,
I had made some mill liner castings in 4140 . They were oil quenched and tempered,however, they cracked very soon.I had used flash tempering too before the castings were completely tempere,hardness desired was 400BHN. Changing the alloy to 4340 reduced the frequency of cracking. There were no issues. It was a rush job for replacement and I had no time to think.
Does this experience qualify as a case of temper embrittlement?


If you think education is expensive, try Ignorance.
- Andy McIntyre


_____________________________________

 

[redpicker]

Personally, I don't like to see temper embrittlement referred to as a failure mechanism. In every case that I have seen blamed on temper embrittlement, the true cause of failure was low toughness. Now, temper embrittlement may (or may not) have been the cause of the low toughness, but my point is that, in most (if not all) of these cases, toughness (in the terms of CVN results) was never a specified property, so was CVN tests were never performed.

Specifically, with 4140, if you are in a quenched and tempered condition and your tempering temperature is below 1000 F, you are going to have low toughness. If you are dealing with a large shafts and rotors for power generation, for example, you are going to need a minimum of 30 Ft-Lbs CVN (or higher). If you don't have this, small defects, service cracks, etc..., are going to initiate a catastrophic faliure. If a faliure occurs in such a part whose material does not have this minimum impact strength, it really is irrelevant as to whether the impact strength is low due to temper embrittlement, poor transformation during quenching, or to low hot working reduction ratio.

If such a part was put into service without being qualified thorugh CVN testing, I would consider the faliure to be more attirbutable to a design flaw (that is, lack of CVN requirements for the material, a design requirement) than to temper embrittlement.

rp