3.6 board intermittently unresponsive & trigger disappears

[Machismo]

I have a early 'diy' 3.6 board which has been sitting on the shelf for some years now. I recently wired it in the car and updated the firmware to the latest version, which went well and I also installed a crank vr sensor on 60-2 wheel which gives a good trigger when cranking. After that I installed two FQPF20N06L fets for the two 1 ohm tbi injectors + 6.8 ohm resistor in series so that the impedance would not be too low. Ignition is at the moment handled externally by HEI distributor and the timing is set to 10 degrees advance so I won't have to deal with that yet and can concentrate making the injection work properly first.

Everytime I tried to start the engine, it did, but the injectors run erratic and when the engine shut down after few seconds(from too much or too little fuel), one or both injectors usually stayed on and would have flooded the engine if I had not turned the ignition in off position which turns the Vems board off completely, of course.
This is my first question. If engine does not turn anymore and vems has no trigger, why does it still dump fuel and why does the fuel pump also keeps running endlessly? It seems the board is somehow in a 'tilt'. I've read that the modern firmwares include anti flooding changes so I guess it should not do that.

When looking at the trigger log, at first while cranking signal is flawless with no disturbance or missing parts. Then if I understand it correctly comes the first injector pulse, which makes the trigger disappear completely for roughly half crankshaft turn. After that trigger returns and seems to work nicely right until the next injector pulse. That repeats forever.

Does this sound familiar to anyone? When I get home I will attach a trigger log for you to see.

[Machismo]

Actually, now that I look at the trigger log, the pulse is labeled: IGN0_FALLING So is it the ignition pulse after all?



I have no ignition drivers installed at the board, but I had to select some number of ignition drivers(1-8) from vemstune so I selected 1.

I'm sure this is something very simple, but I have no idea what's going on..

[Machismo]

Today I experimented with three different grounding schemes, of which one was the official recommendation. They acted all the same. Then took off the injector drivers completely and the problem remained the same. After that I changed the number of ignition outputs in Vemstune from 1 to 4 and suddenly the problem was gone! Also tried 2 ig. outputs and that worked also... So now injectors don't flood the engine and fuel pump shuts down properly every time. I still don't understand why exactly it didn't work with one output setting?

[Machismo]

Now I have a new trigger problem which might be more typical? However I have not been able to solve it yet. My trigger wheel is about a 130cm diameter 60-2 trigger wheel from Alfa Romeo and Alfa Vr sensor which is for that model. Triggering works perfectly under 1000 rpm but over that errors start to appear. And over 2000 rpm signal is lost completely.

Here's a typical situation where rpm goes over 1000:

[Machismo]

Today I tried to change the sensor distance from wheel and also turned it in different angle, just in case. Then I used different resistors to pull the signal up and down. None of these had any effect.

Any suggestions?

[Kamuto]

change connection speed... this is used only for diagnostic purposes, only on engine cranking, not on the engine running...

[Machismo]

change connection speed... this is used only for diagnostic purposes, only on engine cranking, not on the engine running...

You were right, I increased the baudrate from 19200 to 115200 and the trigger "errors" were gone. You just earned 20€ or 100mBTC, because I've been trying to figure out this for couple of days now! 

Of course I still had the problem of max 2000 rpm, but looking at it now at different perspective, I noticed that it totally cuts the fuel over that.  So I investigated and found two settings from Base Setup - Rev limit / Overrun - Overrun Fuelcut:
Fuelcut Below was 25 by default(I still have many default settings), that was way too much pressure so every time I let go off the pedal after pressing the throttle it stalled. I changed it to 10 for now, no more stalling.
Fuelcut Rpm was 2000 for some reason, which was the hard limit I was against. Changed it to 6000 and now it revs with no problem.

Just for the record, here is the the proper trigger log now when I increase the rpm, it neatly just cuts the view when it's out of the diagnostic range.



And now fuel injection also works with 1 ignition output selected, I don't know what that was about, it does not happen anymore.