Still struggling with ignition issues [mostly solved]

[SQLGUY]

I think I went the opposite direction from what you had in mind. With the new tooth position, the trigger offset is now -280 degrees. I seem to be getting stronger spark, but so far I have not been able to start the engine (it does flood really well, though).

I'm going to let it sit (and charge) overnight and give it another try tomorrow.

Cheers,
Paul

[grippo]

I would pull the connectors off the injectors and crank and measure timing to make sure it is where you want it. If it is, then put the connectors back and try to start.

[SQLGUY]

First of all there's the problem that, while I can see a decent, blue, spark on the plugs, the spark isn't strong enough to always trigger the timing light - sometimes it works, sometimes it doesn't. On my car it works great.

When the timing light is firing I could see that I had my cranking pulse occurring at TDC with my -280 trigger offset and custom rotor. Regardless, after cleaning the plugs several times, letting the battery get good and charged, whatever, I couldn't get the bike to do anything but occasionally backfire.

So, I switched back to the stock TCI module and rotor. Still couldn't get it to start. Cleaned the plugs (AGAIN), disconnected fuel, and cranked a lot with WOT. Finally got a bit of a cough. Sprayed in a bit of starter fluid and got a good rev... great. Reconnected fuel... it started... and died, kind of like when trying to run on empty. After about a few repeats of this, I put another gallon of gas in the tank, just to make sure the fuel pump pickup wasn't sucking air. Still the same behavior: starts, revs, even to several thousand RPM for a second or so, then dies. It's like spark or fuel is cutting out entirely after a couple of seconds. The fuel pump is continuing to run up through the point that the engine dies.

The last time this evening I tried to crank it I got one really loud backfire. Surprised my neighbors haven't called the cops!

This is really getting discouraging. I don't understand where this new problem came from.

There's one success story on the site for a guy who used MS-II on a 1982 Yamaha XJ1100. That bike has a lot in common with mine. Same basic engine, same electrical and ignition systems, same final drive, etc. For his project he installed EDIS from a Chevy Cavalier. I didn't want to replace my coils, as they worked fine before. I am still not sure I want to replace them, as that's introducing one more big variable to this equation. I am wondering, though, whether I might be fighting with some shortcomings of the stock ignition system. What do you think?

Thanks,
Paul

[grippo]

Is it possible you have too low impedance (1 ohm) injectors - see the problem Blue had on the Fuel and Air thread. I doubt it because the timing light doesn't work right so spark is more likely the problem.

[SQLGUY]

IIRC, they were something like 12 Ohms each.

What do you think of the idea of upgrading the ignition system regardless? Would it be worthwhile to install the GM EDIS setup?

Thanks,
Paul

[grippo]

I don't like the dual tach system because it gives poor timing accuracy, especially on a light motor than can rev very fast from low speed. An upgrade would be to put in a toothed wheel, something with 24-1 or 36-1 teeth would be good. But I wouldn't do that until I solved this problem, because it might not work any better.

Tell me this, does the engine still work with TCI ? I read what you said about restarting and it ran well. But when you restarted it again and it didn't work, were you using TCI ?

When it fails to run well with microsquirt, with the -280 deg trigger offset, is the dwell still very short of 8 ms ? If it is, then I would move it way further in, like -180 or more. The reson I say this is that at low rpm an 8 ms dwell only lasts 30 deg or so, but at your 9500 rpm redline, it will probably be close to 300 deg.

[SQLGUY]

The starting and dying thing was running on TCI. This evening I went in and reduced the priming pulse and increased the VE map to see if I could get it to run better and see why it was dying. It flooded basically immediately and backfired really powerfully again after I had cranked it for a while at WOT/flood clear. I wonder whether the fuel vapor is being ignited by the HEGO heater. Does this type of thing tend to destroy O2 sensors?

Anyway, I'm kind of sick of this apparent weak spark issue. I'm going to go to the junk yard either tomorrow or Saturday and pull the coils and EDIS module off of a Cavalier. I can't help feeling that having spark that's usually not strong enough to trigger the timing light must be part of the problem here. I bought a spool of Accel 7mm copper core wire some years ago when I was building a new wire set for my car, so at least I've got some good wire from which to make a new set for the bike.

For the record, I'd like to suggest a "spark test mode", similar to the injector test mode, as an enhancement request for the firmware.

Cheers,
Paul

[SQLGUY]

I pulled a coil pack with DIS plate from a 95 Saturn something or other. Next question I have is: is there any benefit to using the DIS module, or would I be just as well off letting MicroSquirt directly drive these coils?

Having uS drive the coils would certainly be simpler for me from a wiring/mechanical perspective. It would also work with my current crank sensor setup rather than requiring me to fabricate a new sensor wheel.

Thanks,
Paul

[grippo]

I would definitely try the coils alone rather than change the whole setup.

[SQLGUY]

I measured the two coils at about 5mH and .25 Ohms, so 3.4msec max dwell per the calculator. They look like they will just barely squeeze into the space where the original coils sat if I bolt them up base-to-base. I had to cut into them a bit in order to solder pigtails for the 12V and switched connections; I am waiting for the silicone from that to cure.



The new wire set is also coming along nicely (a real hybridization, though: Belden on one end, Accel in the middle, and NGK at the other end). One big advantage of switching out coils is that I can now replace wires - the stock wires are permanently attached to the coils. A second advantage, of course, is the higher voltage of these more modern coils. Ironically, the stock coils and these GM Saturn ones were both Hitachi products.

Should be able to try ignition at least later tonight. Will let you all know how it goes.

Cheers,
Paul