Civiclink 92-95 cranking dwell time

[Gavin Magee]

Hi all...

I have a quick query about dwell time on cranking.

I currently have a Honda CRX VT (91 model obd0) on the dyno. The car was running perfectly as I was adjusting idle/fuelling etc before getting into tuning. Suddenly the engine stopped (unfortunately I wasn't logging at the time). This was as if someone just switched it off... no spluttering, no hesitation or anything, just stopped immediately.

After some basic checks I noticed that the dwell time on the tuning parameter list was showing more than 50ms which is of course ridiculously high.

I then tried cranking the engine to see what dwell was doing and it was showing between 20ms and 50ms, yet the highest figure in the dwell table is 6ms.

Needless to say, the engine is still not firing and I got one or two loud backfires when cranking.

I am still seeing a spark during cranking although it doesn't look particularly strong but not the worst I've seen.

The question is... why am I seeing these huge dwell times in the logs during cranking? Is there something going on in the background of the software during cranking that results in massive dwell times? Surely the dwell table should already account for the lower battery voltages when cranking?

Just to add, this is the civiclink92-95 G4x ECU with an adapter harness to suit the OBD0 vehicle. The trigger setup also required changing to work with the OBD0 distributor. Only mentioned for reference.

The calibration and a couple of logs of cranking are attached...

Many thanks

Gav.

Niall CRX VT.pclx CRX cranking 1.llgx CRX cranking 2.llgx

[Vaughan]

5 hours ago, Gavin Magee said:

After some basic checks I noticed that the dwell time on the tuning parameter list was showing more than 50ms which is of course ridiculously high.

I then tried cranking the engine to see what dwell was doing and it was showing between 20ms and 50ms, yet the highest figure in the dwell table is 6ms.

The dwell runtime displays the measured dwell not the requested dwell, the difference comes from the dwell time being converted into engine angles and the engine speed changing between the start of dwell and the end of dwell.

 

Log one shows a big spike in manifold pressure right at the end of cranking which implies a big backfire and the long dwell at that point would be because of it calculated the end angle on a sudden spike of engine speed.

Log 2 shows the exact same sequence of events.

 

Can you take a trigger scope while cranking please and maybe double check your ignition timing with a timing light.

[Gavin Magee]

Thanks Vaughan,

That makes at least some sense with regards the dwell time (runtime) values so I'll rule that out as a possible issue for the moment.

You were indeed right on the base timing. I checked again and I was roughly around 10deg out (retarded from target). I can only assume I must have referenced the wrong mark on the crank pulley by mistake.  That has now been corrected and verified with the engine running.

Again today, I was able to fire the car from cold and continue to make some adjustments during warmup. Although this time I let the logging run. Again once the engine got above 90deg or so, I noticed one or two slight hesitations. I restarted logging and managed to capture the issue. It seems I am loosing trig 1 signal (only when the engine is hot). You can see the log below 'CRX rpm drop trig err'.

I also tried cranking when hot and I'm consistently getting a loss with the trig 1 signal. I've tried reducing the voltage threshold around cranking and idle with no joy. I also tried increasing the trig filter and neither helped at all. Possibly just an old and worn sensor (this car being a 91 model)?

 

Logs are attached along with the current calibration and the scope file during cranking.

CRX cranking trig V thresh reduced filter 3.llgx CRX cranking trig V thresh reduced.llgx CRX VT trig scope cranking.llgx CRX rpm drop trig err.llgx Niall CRX VT.pclx

[Adamw]

Yeah there is definitely some issue with that trigger, every 4th tooth on trigger 1 is lower voltage and distorted waveform compared to the others.  I've never personally seen one of these OBD0 distributors, do they have the 4 tooth wheel right in the bottom under the 16tooth?  If it does it seems that is possibly interfering somehow - whether it is via a shared ground or some magnetic effect im not sure.  Assuming it does have the 4 tooth wheel in there, the first thing I would try is snipping the wires off that bottom sensor.