Welp, after doing some more testing and thinking, it's looking like my brand new Airtex pump assembly is the culprit after only a few hundred miles of use. The PATS error with the PCM seems to have been a fluke, I can't replicate it. Using Forscan the PCM easily passes its key on engine off, and engine running tests without any errors (other than a P0411 for secondary air injection performance, but I never see that in actual practice so its not worth worrying about). I found a 200mv loss on the grounding wire from the PCM to the engine block, so I will address that (and I swapped around the PCM power relay in case it's getting sticky). But otherwise the main issue seems to be random loss of fuel pressure.
It's funny too, because in my haste to replace the fuel pressure sensor, I didn't really properly comprehend what the data was telling me beforehand. The new sensor is actually defective, as once the engine warms up to operating temperature and you shut it off, the pressure value immediately goes to 90PSI and stays there, which is either extremely high resistance or an open circuit. I plugged in the old one real quick and it read 0 as it's supposed to (since it's not connected to the vehicle), so that means that P0193 can basically be ignored, the new sensor is junk. I never got that code before swapping it either, so further evidence that the circuit is fine.
With the old sensor, sometimes with the key on, engine off I would see 80PSI, which always seemed way too high to me, but I ignored it because I couldn't find any real data on the flex fuel system specs. I should have known that would end up being a problem, since with the OEM pump it never crested 60PSI key on engine off. It worked once the engine was running, but clearly this pump assembly had trouble managing its deadhead pressure level. The relief valve in the pump shouldn't have let it get that high. Now it seems the problem is the opposite, I never see deadhead pressure that high anymore, but instead I'm getting a rapid 0PSI either right after the engine starts, or when I initially prime it. I always hear the pump making electrical noises when I turn the key to on (and it makes a different noise when it actually builds pressure), so its working electrically, but the fuel doesn't seem to be getting where it needs to go. I checked the filter for restrictions and nothing came out, just clean fuel. Once the pump actually does come back to life, it produces proper running pressure of 40PSI and stays on indefinitely until I shut the car off, and the car drives fine. My guess is that pressure relief is now sticking open sometimes, and once it does finally close and let the car run, the pressure stays low enough to keep it from having to open again. With priming, the car is asking for 75% duty cycle which seems to be overwhelming this relief. Just a theory though, no way to know without pulling it out. I have ordered a new OEM pump. In the mean time I'm going to run some tests on the driver module, but so far I've seen no evidence it's not doing what it should, nor are any codes being popped that would point to it as a culprit. The fact the car stays running indefinitely once you can actually get it started and that the pump duty cycle (20% at idle) seems appropriate is also a sign of good health of the overall fuel management circuit.
This car has definitely been the most challenging of the Taurii I've owned, and I've also never had this bad of luck with aftermarket parts before. But you definitely get what you pay for I suppose (despite those Carquest pumps being $350 lmao). Luckily for me I need to replace the filler neck anyway, so dropping the tank a second time is only a little bit of extra work. Plus every connection has been greased or antiseized from the first time, so it should fight me far less
I know that was a lot of text, but if anyone has any suggestions or comments, let me know.
