DIYPNP - Relays cycle rapidly, no start
 

Background: NA8 with DIYPNP, ran well for 1.5 years. At an autocross event the ECU came loose from it's mounting and suffered what I presume was a decent impact during a transition. The car died and would not restart.

Symptoms: With key on the following components cycle rapidly; injectors, fuel pump relay and IACV. The tachometer sits at 1k RPM as well. Key off it goes to 0 RPM, however, after unplugging the cycling components it would remain at 0 RPM with key on.
Tested for spark visually and there was none.
Battery voltage checked out.
PC will not connect to the MS while it's plugged into the car. Powered it up outside the car and it connected fine. Reflashed the firmware and all symptoms are the same.
Also tried replacing one of the injector drivers on the microsquirt board for the heck of it to no avail. It burnt out previously so thought it may have been related to the issue.
Runs fine on stock ECU

Any help would be greatly appreciated

Does the ecu voltage read 5v with the MS in the car?

VREF is 4.8v. Tried checking it off the car using a power supply and was the same

Check your Inductor isn't broken (L1 iirc)
and your Crystal Clock.

An impact could break the clock and cause it to not work properly at all.
IIRC even just vibrations damage them hence why they are typically hot glued in place to dampen vibrations

Just a thought though.

Would this inductor be located on the daughterboard? I wasn't able to locate L1 or inductor in the diagrams.

Tried replacing the microsquirt board to no avail. Seems to talk to the laptop normally now though. With key on the relays still go loco and the 'lost sync counter' keeps going up while the tacho in TunerStudio jumps from 0 to ~150 rpm sporadically.

Attached a log of cranking since the MS seems to talk to my laptop now. Hard to tell in the log exactly what happened because of the garbage RPM signal but I started the log- cranked a few seconds, waited a few more seconds then cranked again.

Also attaching the tune for reference.

I replaced the microsquirt module and there was no change in symptoms.

To add to the confusion, the fuel pump relay failed (or so I thought) and I drove the car (on stock ECU) by jumping the F/P and GND pins at the diagnostic connector. Finally got around to replacing the relay today and it is the same. The fuel pump doesn't power up unless F/P (light green wire from fuel pump relay) is jumped to ground. I will bench test the relay but I do believe it is in working condition.

I have checked resistance on this light green wire from the F/P relay to the ECU connector and it is less than 1 ohm. This leads me to believe the factory ECU is not pulling this wire to ground under the normal conditions that trigger the F/P on. Could this be caused a faulty ECU ground? I have visually verified the condition of the grounds next to the throttle body, under the booster, behind the intake manifold and at the PPF.

Hoping the MS issue is somehow connected to this problem but that may just be wishful thinking.

You say it runs fine with the stock ECU in an earlier post. Does this mean you are still running the stock AFM?

Yeah, I run the factory AFM with the stock ECU. It does still run fine that way

When using the MS2 I remove it though. Using a gutted AFM and the usual GM IAT for that

Any chance you made a change to the AFM between the last time you had it running on the MS2 and now?

No, it has been the same unaltered stock AFM since the beginning.

Made a jumper going from the extra adapter board ground pins (4A, 4B) to battery negative to verify this wasn't a ground issue. This did not change the symptoms. Although after attempting to crank again the tacho would sit at ~5250 RPM with key on engine off :confused:

Going to verify critical components on the mainboard are all good next. At least what I can with a DMM

Edit: Found capacitor C3 has no continuity across it's pins. Have ordered a replacement along with another U3. Not entirely sure of C3's function or whether it would have an effect on normal ecu operation but heck it :ugh2:

2 composite logs of cranking from today. Seemed when I first moved the key to the on position there was no crazy relay cycling like before. Cranked a bit then stopped and the tach snapped up to ~5200 rpm and the buzzing of relays ensued once more.

I've swapped a few components in vain and removed any extra components so it matches the TruboKitty.com recommended setup on the main board. Got in touch with DIY Autotune tech support (Matt) as well. We got to checking voltage at VR2 on the main board while spinning the CAS by hand. This led to some interesting results. Here's a video of the DMM while I'm spinning the CAS by hand (+ on VR2, - at battery ground). You can hear the relays/injectors/other things cycling as soon as the wheel crosses each tooth (5v). The signal itself is questionable- staying at .348v instead of returning to 0v between teeth.

I'm no expert but maybe the crank angle sensing portion of the sensor is failing and only shows symptoms on the MS while the stock ECU has filters (or things such as these) to deal with a janky CPS signal.

Haven't ruled out the possibility of a component failure. Acquired replacements for U3, U4, C1, C2, C3 and C4 and may try those soon.