hillpc

... as far as I can tell. The fuel pump current was completely normal the entire time, even after the injector and ignition were already shut down, till the fuel pump was turned off by the ECU a whole second after the engine had coasted to a stop..

Interestingly, the current measurement is so sensitive, and the scope so fast, that one can see the action of the individual commutator segments in the current trace. Maybe this could be useful in predicting a looming fuel pump failure. See https://www.picoauto.com/library/aut...sts/fuel-pump/

By the way, I made my own fuse extension for under $5 (plus a few parts on hand) to get the fuse out away from the fuse box so I could put a current clamp around one of the wires without having to run wires all the way to the trunk. Picoscope will sell you a kit to fit 4 different sizes of fuses for $89. (!)

But there may be something to learn here from the fact that the last 2 injector pulses before they stopped were about 5x the normal length. The first of the two happened in the engine revolution before the engine even started slowing down. This makes me wonder if the long squirts killed the engine by flooding it. I'm going to look at other injectors as well, and maybe look at their current (instead of voltage).

jfrahm

Really interesting info.
You could try unplugging the MAF, maybe it's acting up and as you observe running rich and flooding out.

Alternatively if you run a vacuum line out to the driver's window you can puff in starting fluid or carb cleaner, if you have time between when it's starting to die and dies you could try supplemental fuel.

hillpc

I wish I had reflexes fast enough to puff in some starting fluid as it's stalling. Most of the time the engine is stopped dead (zero rpm) in less than a half second after it first starts to slow down.

jfrahm

A skewing temp or MAF signal could cause that change in injector signal I suppose. Maybe you'd get an "implausible signal" code though, unless the signal is not implausible, just really wrong.