subaru 6-7 help

[VEMS]

works good when cold, misfires when hot is usually a symptom of coil(pack) failure.

[Kamuto]

Never had ecu disconnect from my laptop on this car

[MWfire]

Look how rpm falls/rise fast in warm log. That speed is impossible and that indicate you have packet lost of data.

[Kamuto]

Look how rpm falls/rise fast in warm log. That speed is impossible and that indicate you have packet lost of data.

now I understand your point, it will be coils, tomorow getting new ones   anyway that IBM is gold, 0 connection problems when I had changed my laptop and put ssd in it

[Kamuto]

so, today I have replaced ignition coils, put the ones from 2.5l subaru with internal ignitors, no more packet loss, but still having problems on hot engine


I even replaced both trigger sensor wires from sensor direct to ecu avoiding loom adapter, no change at all. What more can it be?
made new ground point avoiding stock ground wires, no change, and I believe tomorow on cold engine everything will be perfect... 

[fphil]

Hello Kamuto,
Since It may be related to your concern,  I tell in your post that I have reported to Vems for what seems to me an out of sync coil type issue. My engine is indeed oddfire, coil type, maserati biturbo (hence firing are at 90°, 150°, 90° etc), fw is 1.2.30.
Here is the case.
 I was steadily driving at 60km/h entering a tunnel. When  I swiched on the lights (0.:3.88) the engine shakened badly, made big noise, big smoke behind the car and seemed to have no more power.The engine stoppped when I clutched off car still moving. The engine restarted ok clutching on car moving (during busy trafic about Paris, warning lights etc   )
I was able to reproduce this disfunction at my parking slot by switching on and off the lights. Here is the triggerlog.


You see the double firing. After that the engine nicely stopped.

Philippe

[Kamuto]

so, sorting it again, made a new engine wiring for it, because of ghost noise issue shielded crank cables, good grounding, and so on, nothing has changed, then i scoped lm1815 input and output to the processor

primary trigger





I thought my scope just dont have enough resolution lm1815 output, but then i scoped secondary trigger



secondary whatever is the rpm always has output from lm1815


so what is the problem? I have 2 subarus with that problem and cant solve it.
do I have to change some caps on output or convert primary circuit to secondary?

what should I do? it would be nice to get a proper answer in few days because I have wasted really much time on it and still cant get it working properly.

[MWfire]

i what mode is your primary lm(open, 5v or gnd)?

[Kamuto]

Open, +5 does not helps, what r30 should be here?

[fphil]

I noticed the peak level of the VR sensor signal is fluctuating somewhat. Since LM1815 has an adaptive threshold this can be a concern.
Have you tried to record the signal as a wav file along with VT recording triggerlog in order to catch the default?
In any case with coil trig setup the fw should detect the missing or extra pulse, stop firing and resync

[Kamuto]

Trigger log is ok, I just thaught maybe just scope resolution is too small to catch lm output?

[MWfire]

If you want to make longer impulse from lm, just add 10nF between gnd and pin 14(rtc). Because trigger edge is rising you will not have a problem. Problems could happen on tw with many teeth(more than 60).

[Kamuto]

anyway what else possible reasons of it can be? like rpm spikes?

[fphil]

Why not try a .wav recording of the VR signal through the micro input of your computer? You should probably have to divide the signal voltage.

In my case, the primtrig lost which makes the engine to run out of sync may have come from some spike on the voltage power. I need to confirm.

[Kamuto]

added one file, named problem again

http://vems.hu/vemstune/sharingcenter/reports.php?cmd=view&key=RzNmwr

1.2.23 fw with old config, this one throws error but does not start firing on other cylinder.
it was tried with 1.1.96 1.1.94 same stuff happens, board is 3.8
1.2.30 with config from configlet gets trigger error and syncs on wrong cylinder engine dies then