With the extremely helpful XeMODeX flow chart, I learned that the voltage between pins 5 & 6 sometimes did not match battery voltage and would instead sit around 1.3 V. They also sometimes did match battery voltage. The voltage between the ground pin (5) and the positive battery post by the fuse box was always correct. Swapping the ETM relay with other relays doesn't change the behavior.
Further investigation led me to also find this odd 1.3V value at the ETM fuse, so I think the wiring from the fuse through the plug to the ETM works.
When the voltage at the fuse & plug is correct (battery voltage), the car starts as expected.
When it is this odd 1.3V value, the car cranks a few seconds before starting, as I would expect from an improper throttle setting. I also see the VIDA codes ECM-902A (communication with throttle unit - signal missing) and/or CEM-1A5D (Communication with the electronic throttle module (ETM) - Signal missing). I assume this is related to the ETM not receiving the proper 12 V across pins 5 & 6.
Once the car starts, it seems to run and drive fine.
What I would like to know, if anyone may help, is:
- Why would the power through the ETM relay to the ETM fuse be somewhere between 0 V and battery voltage? Bad ground?
- What logic tells the ETM relay to close, and are the steps visible to VIDA?
- If the ETM doesn't have battery voltage at engine start, and the car throws an ETM signal missing error, what signal & logic opens the throttle when I press on the accelerator?
- Last month, the car received a replacement used low-mileage transmission, which has not shown any issues. It replaced a transmission that was failing all on its own.
- I replaced the ETM with a used unit from another 2.9 S80 from eBay seller SwedenAuto within the past year. The previous ETM was original to the car and exhibited severe failure symptoms, such as delayed throttle response, surging on the highway, etc. The new ETM has not had any issues until (maybe) now.
- I have been trying to fix intermittent P0422 & P0432 (catalytic converter below efficiency threshold) errors for a few years, despite the cats running at the proper temperature and the car always passing California smog tests. Fixing vaccum/evap leaks and replacing the PCV canister has not fixed this. If I clear the codes, they reappear in about 1 out of every 3 drives, with no discernable pattern for length or temperature. I could see how this may relate to an intermittent throttle body communication issue.
- Rarely, the map lights/sunroof/garage opener will lose power. Restarting the car restores power. I assume it's a loose wire or bad solder joint, though I have not opened up the panel to inspect.