2012 Range Rover No Start

Vehicle began as no start, codes in ECM for crash signal B10A2-31 missing message and lost communication with RCM (causes no start in this vehicle.) Eventually found missing ignition power at RCM under center console. Traced fault to badly corroded wiring under passenger side floor trim between A/B pillars. Repaired the wiring, had the vehicle running, but now the vehicle will intermittently get the crash signal code in ECM when cycling ignition, and the ECM will code for lost communication with RCM when this happens. Also getting intermittent code U0402 in ECM bad signal from TCM. When this happens, many other modules will complain of no communication with TCM U0101.

The wiring corrosion was very bad, and looked like there was some burning involved. Power, ground, and data all look at the ECM, RCM and TCM. I haven't scoped data yet, but voltages look legitimate, and consistent at each module and the DLC.

What's really confusing to me about this one is that the ECM is only complaining about a bad signal from TCM, while the other modules are complaining of no communication at all. My scanner can communicate with the TCM all the time, and the TCM never codes for anything. Live data in TCM is showing ~9v for sensor voltage output (not sure what this is supposed to be) and engine RPM remains at 0 when cranking. ECM live data shows correct engine RPM when cranking. Another thing that I'm confused about is that the CEM is not coding B10A2 when this is happening. My understanding is that the ECM, CEM, and RCM work together on validating the crash status. Tried probing CAN directly from TCM to ECM and the situation is the same.

I read on Identifix that a failing sensor in the valve body/TCM can cause no communication, I guess I'm just uncertain because I can always communicate by scanner (even with K-line disconnected, which just runs directly from transmission to DLC - what is this line even for if the modules and the scanner are communicating on CAN?) and the ECM is suggesting that it can still talk to the TCM, while the other modules are coding for no communication at all. Could it be that the ECM is meant to process the TCM signals for the other modules, and because it can't interpret the TCM signals, the other modules assume the TCM is offline? There is no selected gear shown on the cluster (which is one of the modules complaining of no communication) when the problem is present, but I can read the selected gear in live data.

Sometimes doing a battery sweep will bring things back online, but not always. Vehicle starts and runs fine with no codes when communication comes back, but usually only for one ignition cycle, and then the RCM and TCM will do their thing again. I'm not certain if the RCM and TCM always have this problem in unison. I want to suggest TCM replacement, but that doesn't really explain the RCM issue. And if one of these modules were messing up the network for the other, wouldn't there be a great deal more communication codes? Usually a network being scrambled will affect the whole thing, right?

I'm not really sure what to test next. Maybe scoping data for inconsistencies with RCM or TCM unplugged?