Anyway, I would like to know your opinion on the distinction in 3 XH types proposed in (1)
I understand your attempt to divvy up the curve but I would perhaps do it based on a little different reasoning. Similar breakpoints but for different reasons. The HAD region is clear, no current forms the breakpoint. The division between LXH and HXH I would do based on whether or not any of the cathode is exposed to the gas phase. As I have indicated previously, when the metal electrode gets exposed, it starts to unload. There you have a different situation as the H2 (or D2) coming out of the electrode can now react at the electrode surface, just like it would at a recombination catalyst. So there should definitely be some consideration of recombination plus the electrochemistry going on in the other part of the electrode. It gets a bit more complicated, as has been discussed previously.
During 'normal operations', ie with the electrodes fully covered by electrolyte, the excess heat results seem to fit the curve shape of the P/(P*-P) term. IOW, the apparent excess heat would seem to explainable fully as an artifact of the F&P model, in particular the enthalpy term for the electrolysis gases. See the attached plot to observe this. I used the vapor pressure of water from
and converted to atmosphes by dividing by 760 to get the solid line. The dots are the excess heat signals from the figure you pointed to (your (1)) plotted vs. the T I read off the graph. The T data are a little crude since I just eyeballed it.
However, if you want to talk about LXH and HXH, I can adapt.