Trevor,
You are absolutely right, the process is called nitriding. That was simply a translation mistake, sorry for causing confusion. In German there are various names for the process, obviously "nitrogen hardening" is not one of them used in English?
I also agree to the diffusion of nitrogen into the ferritic metal matrix, where it prevents dislocations and creates residual compressive stress, thus improving the fatigue strength and the "toughness" in terms of crack deflection.
But: This does apply for the diffusion zone only, while on the surface of the part a very thin link layer of (epsilon and gamma-) iron nitrides, iron-carbon-nitrides, nitrides of the alloying elements and cementite is formed (10-30 micrometers thick). This is what gives the part a very high surface hardness and what increases the wear resistance as well as the tribological properties (the iron nitrides form a more or less porous surface at the very outside which can be used to carry lubricants).
With alloyed QT or nitriding steels hardness values of 65HRC and above are reached.
For the reason you mentioned I always use the nitriding process and not carbo-nitriding. Why you mentioned the ammonia furnace treatment in that context is not clear to me. Where would the carbon come from in ammonia atmosphere (=NH3)?
Compared to the nitriding process, carbo-nitriding is not true to size, so can not be done without subsequent grinding. The nitrided journals will only need a bit of polishing and are ready to fit.
bonny,
If the process parameters (especially the temperature) are set correctly, there is no danger of softening pre-hardened parts. It is common practice in automobile industry to have parts like camshafts hardened (martensitic hardening here), then ground, and finally nitrided. The nitriding temperature has to be lower than the tempering temperature of the steel. Plasma nitriding is done between 380-570°C, while the QT steel 42CrMo4 (common steel for cranks) has a tempering temp. of 540-680°C, for example. Usually the nitriding temp. is chosen 30-50°C below the tempering temp.
Regrinding is not a problem, a case hardened crank will have about the same or even higher surface hardness, while the hardened layer is even thicker. It just will need to be nitrided again after a regrind.
The problem however is that we usually don't know what heat treatment or hardening had been done before or originally. In case of the BSA cranks I performed hardness tests on them (micro indetation), showing values of 35-38HRC, which could be significantly improved by the nitriding process. I assume the cranks had been quenched and tempered (probably directly after forging), and that was about it.
Many repro parts (cams, followers) I have seen for the A's seem to be nitrided as well, I recently got a set of followers for my Matchy that definitely had been.
Cheers, Markus