Surface cracking after liquid nitriding

[prairiefire]

We've recently run into a problem with surface cracks occuring in nitrided parts both during the nitriding process, and in during proof loading of the parts. Of the 8 parts which we sent out for liquid nitriding (the ARCOR process), 1 came back with cracks on the surface and another 4 deceloped similar surface cracks during their first proof loading (well below yield strength). The cracks travel circumferentially around the part at some of the locations of the highest stress concentrations. I've run an FEA case to simulate the proof loading, and no plastic strain is generated at this location. The part is made from a single piece of 9" 4330-V-mod round stock.

During the nitriding process, we've been assured that our parts are sufficiently clean, the process is run as designed and no thermal shocks are encountered. Neither the owner of the ARCOR process, or our nitriding outfit can provide us with any information regarding the cracking, and I can find no resources which can explain why cracking occurs or how to prevent the cracking.

I don't yet have the results, but we're sending the parts out to get a micrograph of the cracks to determine if the cracks occur before/during nitriding, or after removal from the liquid bath.

Any insight into the cause of the cracks would be useful. Would gas nitriding produce more predictable and desirable results (single or double stage)?

Thanks!

 

[swall]

How did you come to the conclusion that that parts were cracked after nitriding?

 

[prairiefire]

All of the mandrels were proof loaded before they were sent out for nitriding, and no issues were encountered. We have hundreds of these mandrels out in service right now without nitriding and this type of failure has not occurred on any of them. I suppose we aren't 100% sure that the cracks occurred because of nitriding, but all of the evidence seems to point that direction.

 

[israelkk]

What was the hardness before nitriding? Has the part gone any surface cleaning before the nitriding? Is there a chance that a chemical cleaning was done that may cause hydrogen embrittlement?

 

[prairiefire]

-Hardness before nitriding was around 38 Rockwell C. After nitriding, it was about 58 Rockwell C.
-The nitriding outfit insists sufficient surface cleaning was done.
-The solvents they used shouldn't have caused any harm to the material.

 

[TVP]

One possibility may be related to the cooling rate after nitriding. Do you know how the parts are being cooled immediately after they are removed from the nitriding salt bath? I have had some problems with brittle parts when they were cooled too quickly. Allowing the parts to remain at a temperature of ~ 400 C for 60 minutes prior to further cooling to ambient may eliminate nitrogen that is dissolved in solution, which contributes to brittle behavior under tensile stress.

 

[prairiefire]

I received some more information on the nitriding process we used:
-Nitriding bath @ 570°C for 1.5-2.5 hours
-Oxidizing bath immediately after nitriding (350-400°C) for 20 minutes
-Water cool (near ambient) before bead blasting
-Oxidizing bath and water cool again
Our nitriding outfit says this procedure was followed, but I doubt there is any way to know for certain if the procedure was followed exactly. I don't think the parts were cooled too quickly, but the possibility of remaining at a high temperature for too long sounds interesting.

 

[swall]

Two questions after re-reading the entire thread. 1--What was the purpose of the proof load test prior to nitriding? 2--What was the tempering temperature and time from the original heat treat prior to nitriding?

 

[prairiefire]

The proof loading is just part of our QC procedure after manufacture. All mandrels are proof loaded, and then we decided to send this batch out for nitriding.

We spec our tempering temperature to be above 593°C. We don't spec a tempering temperature and time however, we specify a set of material properties and a tolerance and tell the heat treater to obtain them. Looking back at some MTR's from different manufacturers, tempering temperature has been between 600 and 625°C and tempering time between 5 and 8.5 hours. The stock being tempered is approx 8" diameter.

 

[coatingengineer]

Hi. For surface fatigue, the problem may be in the diffusion layer. Alloyed steels can result in thicker carbides or nitrides in grain boundaries, which may initiate cracking. I wish I could be more helpful. But, right now, I wouldn't rule out: process, material, or geometry.

William Gunnar

http://www.IndustrialCoatingsWorld.com