4140 Threaded Stud Failure 8

[Metalguy]

Had a somewhat strange failure recently that I had to do the analysis on. The stud is 1/2" dia A-193-B7. Operated for years at ~475 deg F. During removal from the valve (room temp)it snapped off. 100% brittle cleavage fracture. Did the usual checks-small grain size, normal chem, Rc28, only trace amounts of Sb, Sn. Had ~.12% Al, but N was high at 200 PPM. Microstructure looked fine-tempered martensite. Had the lab. make a small tensile bar from the stud-met all requirements!

Thought about temperature/temper embrittlement, but most of my tech. sources indicate IG fracture, not TG. Thread was cut, not rolled, and the root was pretty rough.

Any ideas?

 

[TVP]

I definitely think that the elevated levels of Al & N are contributing to the poor fracture toughness. N content should never be higher than ~ 0.012% by mass with today's steelmaking practices. ASTM F 2282 Standard Specification for Quality Assurance Requirements for Carbon and Alloy Steel Wire, Rods, and Bars for Mechanical Fasteners has a requirement for 0.009% N max, but a more typical level is < 0.007% for cold-heading quality steels. In any case, the N can only be assumed to be fully tied up when the ratio of Al:N is 3:1, so strain aging should be considered as extremely likely. Dynamic strain aging occurs in the temperature range of 100-300 C (210-570 F).

ASM Handbook Volume 1 has a section on Aluminum Nitride Embrittlement under the more general topic of Embrittlement of Steels. "The fractures occur at the primary austenite grain boundaries...Platelike aluminum nitride produces small, shiny fracture surface facets that are generally observed."

 

[Metalguy]

CoryPad,
The toughness has to be very low in order to get 100% cleavage fracture at the relatively low strain rate the mechanic applied. Twisting one end of the stuck stud with vise-grips is far slower straining rate than an impact tester, etc. Also, the embrittling elements "usually" cause IG fracture, which I don't have. That's why this case is so oddball.

 

[Metalguy]

TVP,
If we look at the Al:N ratio it's 4.5. Also, strain aging "usually" causes IG fracture.

I'm still baffled!

 

[Metalguy]

Carburize,
I tried to look at Brians forum but it's blocked--by the computer guru's here at work. I'll check it when I get home. But I'm WAY to slow for IPSC!

 

[metengr]

Metalguy;
I would suggest if you need additional information on this subject, the reference book by Sinha titled “Physical Metallurgy Handbook” has a rather detailed section under tempered martensite embrittlement (TME). The Al-N embrittlement is also discussed separately. I happened to review the TME section last night when I was looking for information on an unrelated topic.

To make a long story short, Sinha states in the TME section that two modes of failure can occur - transgranular or intergranular. The transgranular mode can be caused by precipitation of FeN compounds during tempering along interlath paths of tempered martensite resulting in brittle fracture. This is all I remember, I don’t have this reference book with me today.

 

[SMF1964]

Some random thoughts:

1. We've experienced quasi-cleavage/cleavage (mixed mode) failures in low alloy lifting lugs under tensile overload conditions at 50°F. Run a charpy series on the material and get a transition temperature of 75°F. Take a charpy sample and 3-point bend (tensile test loading rates, not impact) it at LN2 temperatures and get a cleavage failure. Could you be dealing with a shift in the DBTT during service?

2. Swall: If you are looking at embrittlement of leaded steels during moderate temperature exposure, are you talking about the lead melting and causing LME? We see that in shotgun deslagging related failures.

3. Strain ageing embrittlement: You are in the ballpark for mechanism (but that usually affects carbon steels) and for Blue Brittleness (which can affect low alloy steels).

 

[swall]

The ASM Handbook section mentioned that leaded steel embrittlement was observed below the melting point of lead, at temperatures as low as 400F. But, as I mentioned previously, I think it is a stretch that this could be the source of Metalguy's problem, even if the 4140 turns out to be the leaded version.I would have expected the stud to have failed in service, rather than upon disassembly.

 

[unclesyd]

I think that something is wrong with this particular stud. This is based on the fact that we have thousands upon thousands of B7 studs of every description operating in our polymer process without any signs of any type embrittlement what so ever. The studs operate from ambient to 300°C . Some have been in service since 1952 either as Crane Alloy or B7. Some of these studs have been cycled 100's of time to 900° F-1000° F then returned to service. There is not corrosion as the studs are Dagged prior to use. I have Magnafluxed, sectioned, and chemically analyzed a very good represenetive sample of these studs, again finding no problem. We didn’t analyze for N. As stated in previous posts I’ve checked several thousands of these studs for distortion using a thread rolling die and check gauges as none of these studs are torqued and are used for seating RTJ and Flexitallic gaskets. No problems. Proof testing always revealed the specified yield values.

The only problem I’ve seen with B7 or Allen SHCS is in an area where they was possible or positive exposure to HCL vapors in very low concentrations. These studs broke with a bang, literally. The culprit was H2 embrittlement to first degree. HCL has been the only acid problems.
When we have been forced on projects to use B7 with SS flanges in HNO3 service, no problems with H2 as the HNO3 corrodes the fastener too fast.

We have guest operation on site that did have some embrittlement problems with B7 handling NOX and NO. We were not privy to the investigation.

Here are a couple of miscellanous problems that I've encountered over the years.

Someone called out B7 galvanized and the supplier stripped the zinc with brick wash and was proud of it.

We had an incident with a sister company had a large quanity of B7 studs rust in storage and they gave out an achievment award for someone comming up with a claening procedure using toilet bowel cleaner (HCL). To compound the problem they comingled the acid treated studs with several thousand others.

 

[CoryPad]

Metalguy,

Now I understand why you say this would have to be low toughness steel. I am sure you would agree that this steel should not have low toughness. I think you may be placing too much emphasis expecting to see intergranular fracture with the common low toughness phenomena in steels.

For example, hydrogen assisted cracking usually is intergranular, but it can be transgranular. Trasgranular propagation is a common mechanism in steels used in sour service for oil/gas production. In "Hydrogen-induced intergranular fracture of steels", Engineering Fracture Mechanics 68 (2001) 773-788, the following is in the summary:

"There are two kinds of hydrogen-induced cracking found in heat-treated steels. One is induced by concentrated plastic flow, and the rate of this cracking is controlled by the rate of displacement of the loading points. This type of cracking is transgranular with respect to the prior austenite grains, and it occurs at high levels of stress or stress intensity."

So, maybe this problem is due to a less common mechanism, but you might need to look more at the issues mentioned here to solve this problem.

Regards,

Cory

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