Heat Treatment of 1144

[mrbebu]

Hello All,

I have a situation where a batch of parts (highly re-used specialized M12 to M14 sized hex nuts) are undergoing brittle failure in initial tightening cycles (fail in the 2nd to 5th tightening cycles).

My spec calls for 34-37 HRC and the parts came in at 41-42 HRC, although the heat treatment certification claims 37 HRC for the batch. In addition to my own hardness tests which have confirmed the high hardness, I've also sent failed samples and known good (used) samples to an accredited Metallurgical/Materials testing facility for complete analysis. Their conclusion was the failures are due to excessive brittleness from higher than recommended hardness for 1144.

My question is, is it possible to safely conduct a "partial annealing" at X temperature for Y time in a tempering furnace to bring the parts back into my spec range of 34-37 HRC, or do they need to be fully annealed and re-treated?

The caveat here is that the parts are already plated (some yellow cadmium, some black zinc) so I'm not sure how those plating chemistries will behave during a partial annealing/softening process, or even a full "re-do" cycle. In the case of the latter, I suppose they'll need to be stripped first.

Thank you in advance for your responses.

 

[EdStainless]

You could temper them at a slightly higher temp that was used initially.
This may take some experimenting to get right.
Look at the documentation for this alloy.
Did you also check for surface damage (decarb)?

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

 

[TugboatEng]

Have you considered hydrogen embrittlement as a cause of the failure? That would explain the higher rate of failure at the higher hardness. Electroplating can expose the material to conditions that can cause hydrogen embrittlement.

 

[EdStainless]

TBE, for that matter even poorly controlled acid cleaning can result in hydrogen embrittlement.
They should be baked out after plating or cleaning.

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

 

[mrbebu]

My initial thought when I first found out of the failures was hydrogen embrittlement, which is why I sent them to a lab for analysis immediately. But, they ruled that out as they found no clear evidence of it. I was then a bit shocked to see what they measured for hardness, so I confirmed that on my own [old] hardness tester.

H.E. is another reason why I have an upper limit of 37 HRC so I don't need to bake out after electro-plating. So, if the root cause was actually HE, it would still be a byproduct of the Heat Treater taking them to a hardness level that absolutely requires addressing that?

Either way, I'm worried about all the re-work introducing defects/unknowns or exposing them to more hydrogen. Or, maybe I'm just being too paranoid? The heat treat facility has already offered to take them back to address the issue, but I'm really not sure I want to go that route.

As for documentation regarding alloy/s, can you recommend a reputable source for this?

 

[EdStainless]

I wouldn't rely on HRC37 as being 'safe' from HE.
A retemper shouldn't create any problems.
If they clean afterwards you should be asking about their procedures.

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

 

[mrbebu]

Perhaps anecdotal, but in my case there are thousands of these plated nuts from different batches out in service at HRC34-37 without a single failure over the last 5-6 years. That said, I will most certainly reconsider HE baking though.

Ed, any recommendation for documentation for this alloy (1144)?

 

[EdStainless]

Google until you find info.
The temper data should be similar to any 44C steel, but the starting hardness will be different.


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

 

[mrfailure]

It is often difficult for lab analysis to find clear evidence of hydrogen embrittlement. Here, I suspect it highly likely because the nuts are harder than specified, and are in a range more prone to HE. As noted, even your specified hardness range is still at risk for HE. Proper bakeout and cleaning procedures should thus be specified to avoid the possibility of hydrogen embrittlement.

 

[MagBen]

curious to know how the metallurgical analysis ruled out HE? did they get a general H content of the steel? if it was about 1 ppm, i doubt it was HE, at least not the main cause. lots of debates on HE since H is too slight, not detectable in many cases, so hard to find direct evidence. but a rule of thumb seems to be, if less than 39HRC or <1200Mpa strength, HE is not so likely. high hardness itself is not a good thing for cracking. i think it necesary to limit your hardness to a max of 37RC.

what was your temper temperature? if it was at 600F range, the temper itself can serves as baking out process. it maybe worth running a LECO test (if not done yet) to see if H content is high.

 

[Tmoose]

How are these being tightened?
If torqued, what is the maximum torquage?

got pictures of the failed nuts?

Any kool failure analysis pictures from the accredited Metallurgical/Materials testing facility ?

I think of 1144 as StressProof® , although that is just one available flavor mostly used as-is at around 220 BHN.

 

[mrbebu]

curious to know how the metallurgical analysis ruled out HE?...

In their SEM/EDS analysis it is mentioned that it "...disclosed no evidence of fatigue or any clear evidence of hydrogen
embrittlement..."

As for the limit of 37RC, that is already my max limit called out on all prints. I may lower this to 36RC in future revision, or I may also consider just moving to ETD-150 (HRC32 minimum, but usually comes in 34-36 range). Will have to see how the manufacturer would respond to machining the ETD material though.

I'm unsure of the temper temperature. Perhaps worth noting that we call for vacuum tempering to minimize scale and build-up for higher luster plated finishes.

Thanks.

 

[EdStainless]

Vac temper may not be a good idea.
Are they doing this to a special Vac HT specification? (AMS or similar)
At the lower temperatures for tempering getting good uniformity or temperature and soak time in Vac is very difficult.
I would rather see this done in Ar.

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

 

[mrbebu]

How are these being tightened?

Torque tightened. I'd rather not get into the torque value discussion at this moment.


As for pictures, not much else detail for obvious reasons, but here you go:

 

[mrfailure]

Note that analysis for hydrogen content is not really useful because the hydrogen that caused HE will have dissipated through the matrix by the time the component is analyzed.

 

[Tmoose]

Do the pictures show a fracture surface?

 

[mrbebu]

Tmoose, yes those are all pictures of the fracture surface.

 

[Jughandle]

To the question, yes, ‘re-annealing’ should not be detrimental. Time/temp prob too low for grain growth, or geometric distortions (unless they are Class 3B threads - maybe). Coating will need to be reapplied, e.g. losing water-of-hydration leaves Zn phosphates as powder. Then H embrittlement relief, jus’ to be safe. Nice pics; too bad you were stuck with 1144.

 

[Tmoose]

So the nut bodies just suddenly split in half vertically ( assuming the screw axis is vertical ) during 2nd to 5th tightening cycles ?
Or does one side open up, and the nut spreads open a bit?

Did you say how close to yield the tightening is estimated to be?
Are all the tightening cycles the same, or does the torque increase at each cycle?

Are the nut faces making decent contact with the face of the bolted component ? Or is the bolt or the component angled, like the flange on an I-beam?

https://cdn-haoob.nitrocdn.com/DSNu...wp-content/uploads/2014/07/BC-MC_C-flange.png

 

[MagBen]

H can be baked out under inert atmosphere, may not as effective as in vacuum though

@mrfailure: H will dissipate when HE occurs, especially when cracks are opened up. But a testing before HE can give you a general idea how much H is entrapped, as a whole not on a specific localized region.