Still struggling with ignition issues [mostly solved]

[SQLGUY]

Good news is that the new coils and wires work and produce an obviously stronger spark than the originals did.

Also probably good is that the bike at least wanted to start a bit (haven't put the tank back on, so no fuel except that left in the injectors and a bit of starting fluid I gave it).

Bad news is that the new stronger spark still isn't enough to trigger this stupid timing light, at least during cranking. A couple of times when the engine caught for a second or two, the timing light did flash a bit, but during cranking there wasn't the slightest flicker. I went as far as to give it power from a separate 20A 12V regulated power supply, in case the lowered voltage during cranking was confusing it... no help.

Also bad news (I think) is that I still had a couple of strong misfires and one backfile (actual flames in the throttle body). Any ideas on this?

Anyway, tomorrow, I'll try things (carefully) with fuel and see how it goes. I'm also going to tear into this timing light (Innova 3551) to see if there's any internal trigger sensitivity adjustment, or somewhere I could tack in an alternate trigger. Having to get the engine running in order to verify initial timing is a bit too much of a catch 22 for me.

Cheers,
Paul

[SQLGUY]

That is, I can start the bike, and it will rev up a bit, for a few seconds, then die out like it has either no fuel or no spark. I've attached a data log and my current MSQ.

Please let me know if this gives you any ideas as to what's happening currently.

Thanks,
Paul

Oh, also, I found that I could trigger the timing light reliably by wrapping a piece of wire a few turns around the #1 ignition wire and running that to the trigger input of the light instead of using the stock pickup. My #1 spark is hitting right at TDC during cranking, so it looks like my trigger offset is correct.

[SQLGUY]

I went ahead and moved the ignition pickup finger to allow a -80 degree trigger offset. I did this because in my previous datalog I could clearly see dwell being eaten away above 1600 RPM.

Unfortunately, this didn't fix the rev-up-and-die-with-a-lot-of-backfiring issue.

To simplify things further, I changed my spark map to TDC at cranking and 7 degrees of advance everywhere else. It's with that map that I generated the attached datalog. This log isolates a bit better the behavior I'm seeing. Hopefully it will give someone an idea of what I need to change/fix.

One other thing I should mention is that, occasionally, rather than dying once the RPM's start to drop, the engine will recover a bit, rev up a bit, then lose power again. In some cases it will go through three, at most four, cycles like this before finally dying completly. The other potential symptom, I think, is that it's not hard to restart after it does die, though the behavior will be the same and will likely include some misfires and/or backfires.

Cheers,
Paul

[grippo]

During the 5 or so seconds when spark was steady at 7.5 deg advance, was it also steady with the same reading on the timing light ? Everything looked good on the datalog. You seem to have a solid tach, steady and reasonable dwell. If the spark is still too weak, can you jack up the dwell to 4 or 4.5 ms and see if it improves. But make sure the ignition drivers aren't hot. Second thing you could try is leaning or enriching by just temporarily raising Reqfuel say 10%. Another thing would be advancing to say 10 or 12 deg . If the car is still backfiring before it dies, it could be that the spark output is still too weak and a lot of misses are causing excess gas and backfire in the exhaust.

[SQLGUY]

and that the timing is likely OK.

One thing I don't understand in the datalog is that negative spike in CLT at 717223.

I also don't understand why my PW and RPM drop off like they do between 717220 and 717223, with the TP remaining steady and MAP actually rising.

What kind of behavior should I expect if my VE table is too low? It would seem that the 10msec cold priming pulses and the 5msec warm priming pulses provide plenty of fuel to run up the engine initially, which makes me surprised that the engine should be getting fuel starved even at the 4.3msec minimum pulse width at the low point of the datalog. My plugs are also getting a bit suity even from these short runs. Could that be a red herring as to richness, or is it just that the mixture and ignition is bouncing around too much to reach a reasonable steady state of combustion?

The spikes towards the end of the run are where it started backfiring and misfiring. Here it looks like fuel delivery should have increased a bit, but that didn't help things. Note that the throttle position still hadn't changed, so it's just that MAP had risen far enough to take me back towards the cranking corner.

The other thing I should mention is that I noticed that the 12V feeding uS and the coils (which runs through the ignition switch) is .5V lower than the 12V supplied by an accessory relay I added which feeds the HEGO and injectors. I am going to rewire things so that uS and the coils also get their 12V from the accessory relay (to ensure everything's seeing the same voltage) but I did try jumpering the two supplies, to eliminate the .5V drop, and it didn't change this behavior either.

[SQLGUY]

First of all I rewired the system so that the ECU, injectors, and coils were all running from a common switched source, so voltage measurements by uS would be consistent.

Then I tried to start the bike and again was getting backfires, kickback, and general bad behavior.

I checked the timing on cylinder 1; it was hitting right at TDC during cranking, just as it should, with -70 degrees trigger offset.

I then checked the timing on cylinder 2, and it was way off. I would have thought that my "Offset (advance) for Output #2" in advanced options would be 0, since my my two VR pickups were 180 degrees apart and their respective cylinders were also 180 degrees apart. I guess that's not the case.

Here's a picure of the crank trigger setup with the engine at cylinder 1/4 TDC. The bottom pickup is for cylinders 1/4, the top one is for 2/3. The crank rotates CCW.



One other complication factor here is that my #1 ignition channel is for cylinders 2/3 and the #2 channel is for 1/4.

I spent some time going back and forth between these: first setting cylinder 2 to fire at TDC by adjusting base trigger offset, then adjusting the cylinder 1 firing with the advanced offset. When I had 1 right, 2 would generally be wrong again or apparently firing multiple times per rev without a coherent trigger. This was all done while cranking with the injectors disconnected.

The last settings I had were -40 for base trigger offset and 20 for advanced offset. I know the docs say advanced offset needs to be negative as well, but negative values here had no impact on offset for #1. These numbers also make no sense to me, as the rotor "tooth" is physically 90 degrees before the VR pickup at TDC, and, the two pickups are opposed. With these settings the timing light showed #1 hitting at TDC and #2 apparently double-firing at TDC and BDC. The result was still a bunch of coughing and kicking when I added fuel back in.

FWIW, I've currently I've got my masks set to 5ms and 70%, and it's a good thing I keep my hair really short or I would have pulled it out by now!

So what's up with my #1 ignition channel here? Why do these two settings seem to be interfering with each other? I thought the two ignition channels were independent.

[grippo]

In dual tach mode the 2 channels are completely independent. You should not have to use the Advance offset for ouput #2. If you have a 2 channel scope you need to verify the 2 tach inputs are 180 out of phase. Then, with the advance offset = 0 deg, scope the 2 outputs and they should also be 180 out of phase.

[750essess]

I think your offset should be -290 with dual tach inputs. I had the same issues. Also zero on channel 2 offset.

[SQLGUY]

In dual tach mode the 2 channels are completely independent. You should not have to use the Advance offset for ouput #2. If you have a 2 channel scope you need to verify the 2 tach inputs are 180 out of phase. Then, with the advance offset = 0 deg, scope the 2 outputs and they should also be 180 out of phase.

I'll give that a try tonight or tomorrow... I suppose the timing light may be getting confused, or may be confusing me.

What about the mask settings? The description in the help says that percentage mask is the percentage of the tooth to use as mask before expecting the next tooth. Since I only have one tooth, 1/4" wide, and the circumference of my pickup area is about 9.8", I would expect to need a mask percentage of about 3500% or more. I found, though, that the maximum allowed is 200%, and when I set to this I was only getting spark once every two or three revs.

Thanks,
Paul

[SQLGUY]

I think your offset should be -290 with dual tach inputs. I had the same issues. Also zero on channel 2 offset.

How do you get to -290? Wouldn't that be if the trigger tooth was 70 degrees before TDC? Mine is about 90 degrees after TDC.