Nitriding of Crankshafts

[RRoller123]

My latest area of interest / education, involves the hardening of engine crankshafts, among many other items, gears, sprockets, etc. Good video, very involved. We may be faced with the decision as to whether to re-nitride harden the crankshafts, based upon inspection after disassembly and machining, if needed.

I have plenty of questions around this process, including:

1.) Which nitriding process did the factory use? There are 3 main ones available; Salt Immersion, Gas or Plasma. I think we can rule out plasma, as it was not available back then afaik. And the SI process involves cyanide, so that is unlikely. Most likely it was Gas, which is NH3 Ammonia gas, molecule broken into N and H3 by heat and pressure, and injected into the pressure vessel and absorbed by the steel over a period of time.

2.) How do they get rid of the H gas during processing? Being smaller and lighter than , it could easily disperse into the steel, where it is not wanted. Being lighter, maybe collected at the top of the pressure vessel and burned off? Yet the NH3 was constantly swirled to keep chemical density uniform. Curious how they do this.

3.) Thickness (or depth, actually) of treatment? This modern video, below indicates thathe process runs up to a maximum of about 6mm, or ~0.023" maximum. Yet it takes many hours to achieve this, approximately 48 hours or more, possibly 72 to get a really large depth. So it is likely that normal depths are far lower, like 0.010" to maybe 0.015".

Now, since process soak time is a human variable, and we know all too well what human nature is, it is not with high confidence that one would expect anything over, say 0.020" on a consistent basis. The only way to truly tell is either to do a cross section (destructive) or to do a Rockwell C hardness measurement after machining.

4.) Note the clear Eutectic point on the N-Fe phase diagram, like for solder.

University technical lecture on the process:

https://www.youtube.com/watch?v=LSjHmZLfWco

5.) There is another, quite serious potential issue. There is a video of an Viper engine, ruined by the nitriding process, DUE TO HUMAN ERROR OF THE CUSTOMER, who did not measure the journal diameters again AFTER nitriding. The process expanded the journals a couple of thousandths and closed the gap. The engine was assembled and run, it froze, and threw a rod through the side of the block.

6.) Which of the 3 processes was used? I have not heard of this before. It implies that, if it is determined that machining of a crank is needed, that the nitriding be done before, assuming that a thick enough depth can be attained (longer process soak time) to allow for the removed material. Otherwise, the machining needs to take into account expansion, which can lead to "certain uncertainty".

7.) Is it true that the nitriding process can and does expand the material? Steel in this case. The answer has obvious implications to the process for machining any materials, crankshafts certainly, and adds the new variable of how MUCH machining is going to be needed, to determine whether to re-nitride, and when (before or after).

Viper experience video below.

https://www.youtube.com/watch?v=kiEEsJkNMVA

I find this a fascinating study, and may it be of use to all of us doing extensive engine work. Or just an interesting intellectual pastime. That is ok too.

Pete

[18Fiatsandcounting]

Pete, I have to admit that I know virtually nothing of this topic, but as I have some background in materials chemistry, my first question would be, "Can you re-nitride a crankshaft that has already been through the process during its original manufacture?" Or could you introduce issues such as microcracking, phase separations, embrittlement, etc.?

Apologies if I am only showing my ignorance with this question.

-Bryan

[RRoller123]

You know as much as I do! I am relying upon one undergraduate Materials Science course, albeit an excellent one. I have heard from the suppliers that they DO re-nitride, if enough was removed in machining that the Nitride depth has been compromised. So I am assuming that this is advisable.

It makes sense, as it is an absorption process, and done at temperatures that do not permanently transition the metal. The N atoms diffuse into the BCC and FCC structures (Body-Centered and Face-Centered-Cubic crystalline structures of Fe), as per the Lecture. Assuming that the existing N atoms don't block the subsequent soak absorption, it would be reasonable to assume that new atoms move in, and that the elevated temperature and pressure redistribute the existing N atoms, forcing them deeper into the crystal structure. I.E. the opposite of painting on a new layer. More like rewashing clothes, new absorption.(?)

The experience of the fellow with the Viper engine, ruined because of dimensional changes, I find really interesting. This bears some more investigation! Which process did he use?? It may be that this dimensional swell always happens, and a couple of thou are ignored or overlooked? (not likely, I think) Or that this is accounted for by subsequently under machining? I don't know a machinist who does these operations, who I could ask. Maybe someone here does and could get some input on this.

An interesting Engineering study, with actual and clear consequences for us!


p.s. That Materials Science course was taught by Dr. Ebner at Boston University, a wonderful fellow who had a profoundly positive impact on so many of us there. He retired, but kept an office and never left! Went to the campus every day and advised, assisted, whatever. They gave a dinner in his honor a few years ago, and as he got up to the lectern to speak, he had a heart attack, fell over and passed away. Living and leaving as he loved his school and his life. Unbelievable.

[RRoller123]

So here are some more references, a really interesting study.

It is surprising how little discussion there is on the You Tube or the Literature about expansion of the object that is being treated, But it does come up, and as far as I can find, the data seems to be that the object being treated will expand about 0.001" per 2.0" of diameter. This obviously is dependent upon the material, soak time, a lot of things, but this is the only real data that I can find. Interesting that there is so little on this important aspect of the process.

This correlates to the experience of the guy who blew up his Viper engine by not remeasuring when his crank returned from nitriding; Most of the references state that the machinists are expected to compensate for the expansion IF THEY ARE TOLD ahead of time of the conditions.

What a critical path risk item! Good to be aware of this.

Her are some more references that I have been through, that are pretty good, if one is interested.

https://www.machinedesign.com/materials ... arburizing

https://www.lycoming.com/parts/crankshafts

https://www.highpowermedia.com/Archive/ ... rankshafts

https://bryantracing.com/tech-talk/

https://www.dsmtuners.com/threads/the-c ... ks.309281/

https://www.highpowermedia.com/Archive/ ... rankshafts

[RRoller123]

Ok, so your humble correspondent has been researching this for almost a week now, relatively full time (retirement has benefits), and after reviewing dozens of papers, podcasts, etc, this one is, imho, the best, clearest and most applicable to our situations:

https://www.youtube.com/watch?v=CVJHXQK7V0w

The other ones can be set aside, this one covers pretty much all of what is important to us.

Anyone involved in machining a FIAT crankshaft should be aware of this technology, its limitations and its critical variables.

Fast forward to 4:00 to get past the introductions and right to the heart of the matter. These people are experts in the field, no doubt at all. A three years ago podcast, so still quite relevant.

This podcast applies to extrusion dies, injection dummy blocks, etc, but it corresponds very well to the family of alloyed steels that are used in our crankshafts.

Takeaways:

1.) It seems to me, in reviewing all of these papers, etc, that almost no one Nitrides with a long enough process time to get more than maybe 0.010" of treatment depth. Haven't seen it anywhere in any one of the papers or video podcasts, although they state that it is possible to get down to a depth of ~0.030" if the saturation time is extended to a few DAYS. Not likely. The idea that there is anywhere near 0.020" or 0.030" Nitride thickness on our crankshafts is most likely a myth, and would ONLY be verified by a destructive cross section sample. This has clear implications for grinding and polishing FIAT crankshafts. Until I see evidence of that, I will operate under the assumption that the Nitride layer is maybe 0.005" or possibly 0.010" thick, no more.

1A.) The process at FIAT 40+ years ago is extremely suspect. Most likely was Salt Immersion process back then.

2.) Nitriding can be done repetitively with no problem. That is covered in the podcast. There is a good chart showing the effects of repetitive treatment. As I guessed in my last post, the free N atomic level atoms diffuse further into the steel with subsequent processing.

2A.) GAS Nitriding is the standard. Salt Bath Nitriding is environmentally hazardous, illegal in many nations, and has been phased out. Uses cyanide salts. Ion Plasma Nitriding leaves shadows, does not get inside hollows, etc. Gas process is preferred.

3.) The white surface layer that sometimes occurs is a SERIOUS DEFECT, caused by improper process control. The White Layer (white because the crystalline structure is reflective), is extremely hard, but also brittle, and pieces breaking off will pass into the oil, and potentially damage crank journals, etc.

4.) The process window is very narrow, and very small! Temperature must be keep below the Tempering temperature, to avoid softening the steel. Key control variables are atomic N saturation, furnace dispersion uniformity of the input NH3 ammonia, temperature and time. It is critical that only very new, properly computer-controlled equipment be used for this.

5.) The single most critical input variable, for the part being Nitrided, is starting cleanliness. The crankshaft must be absolutely spotless, literally surgically clean, for this to work. No contaminants, no fingerprints at all. Fortunately, we are only concerned with the journals, so this should be easy to achieve. If our engine rebuilders are not up to these cleanliness standards, then it is critical that the Nitriding service provider is, and that the provider pre-cleans the parts and handles them carefully.

6.) The Nitrider needs to have a full process load, so they must use dummy blocks if the chamber is not full. This is a key parameter to verify with the supplier. They also may be using dummy samples for cross sectioning, to verify both hardness achieved, and depth of process achieved.

So, bottom line for me is this: If a FIAT crankshaft needs to be ground, more than a couple of thousandths, it should be re-Nitrided, and only by a very competent, modern computer-controlled Gas Nitride process supplier. One cannot rely upon an assumed depth of any original nitriding. If the crank is only polished, I would not re-Nitride.

Questions, comments and additions always welcomed. Share the knowledge!

Pete