Siemens deka 630cc flyback - SOLVED -

[Ignitec]

Do I need to use 30v TS diode for flyback? I tried common flyback wire, but it seems ECU cannot control injectors very well that way.

Thanks

[rob@vems.co.uk]

The TS flyback is the only one I'd ever use.

[Ignitec]

Thank you, I'll try... Can this cause cossie turbo to have very weird acceleration (like misfire, but not misfire) every time at 3800RPM and 150Kpa and above? Lambda goes crazy all the time, and engine won't idle with lambda more than aproximately 0.75 - 0.85... Tried to change everithing except injectors (including vems controller, a few coilpacks, spark plugs, ignition cables...)

[rob@vems.co.uk]

This wont help your fuelling, but I think something else may be a problem with the symptoms you describe there.
Get the flyback sorted out and then we'll work out what else could be wrong.

[Ignitec]

I haven't 30V TS at the moment, so I made ddflyback (ten dy399 and two resistors) and got great improvement. Cossi now have nice idle, but still wants to run a little richer, and goes up to 5k RPM and 250kPa smoothly. TS is on its way, I'll inform you of situation as soon as I try it, but I think it will be okay

Thank you very much!

[Ignitec]

Now I have really strange situation: with 30V TS engine idles very nice, runs perfectly smooth sometime up to 200kPa, sometime up to 170kPa, sometime 180... Then begins to misfire. In MegaLog Viewer RPM signal goes a little bit crazy at a time with strange variations (impossible?) of 100-150RPM in 1/100 sec . I've changed VR sensor on crank wheel, but didn't drove car after that. Anybody had something similar? Here's log with those RPM variations.

http://rapidshare.com/files/280996959/escort.xls.html

[gunni]

Please state the firmware your running as numerous things between firmwares can have various changes.

[Ignitec]

I've tried 1.0.73 and 1.0.78. This log is with 1.0.78.

I've converted cossi trigger to 60-2 on crank, used V3 harness, no cam sync, wasted spark ignition, wb02, egt, CLT i MAT from vems.hu, v3.5 board. Didn't had new diode and coil on power supply, so used ones from a friends used controller. It's first standalone i work with, so it's built with maximum of patience and care, but it seems something went wrong... Coilpack was really close to harness, so I moved it further and grounded well, put ground to head, block, starter and alternator to avoid noise, tried everything I might see as a cause of problem. And still no success... Here's MSQ used last time.

http://rapidshare.com/files/281091036/escort.msq.html

[z0tya]

Please read my topic:
http://195.159.109.134/vemsuk/forum/index.php/topic,962.0.html

A bit similar situation.
Did you check Vr polarity?

[Ignitec]

hmmm... I read topic with your problem, but it seems your rpm signal was more noisy than mine. I have "saw like" signal only when misfiring, usually just before it starts, and when I release accel pedal, continues for a little bit and then goes normal...

I haven't tried different polarity of VR, but i surely will. How I found "correct" polarity? Put led diode on two pins that are not shield ground, moved screwdriver to the center of the sensor, and from center of sensor, and when diode blinks when I move screwdriver TO center (if not, I change polarity of led), I assume thats correct polarity. Is this correct?

[z0tya]

Good polarity: if you measure positive voltage (DVM, DC 200mV mode, relative to GND) when approaching a ferromagnetic object (like a wrench) to the sensor. (and negative voltage when moving away).
http://www.vems.hu/wiki/index.php?page=GenBoard%2FManual%2FVrSensor%2FPolarity

One more tip: use the 2nd lcd screen to see trigger error (but i have not seen it on reverse polarity)
http://www.vems.hu/wiki/index.php?page=InputTrigger%2FTriggerLog

[Ignitec]

Thats exactly what I did, screwdriver to sensor generates enough current for led to blink. led didn't blink in that position when I moved screwdriver from sensor, so led + is sensor + when it blinks when moving feromagnetic object TO sensor. Right?

[Ignitec]

Yesterday I found what's broken... Vems, harness, sensors and all other was okay. There was construction error in fuel tank, delivery pump was making mix of air and fuel, and pressure pump passed it to injectors. There wasn't any pressure variations, so I didn't look at pump and tank at all. Thanks to all people who wanted to help me!