MicroSquirt.com read only forum (Jan.15,2018)

Hi all,

I thought about this after an issue with Ignition timing. I have a 12-1 trigger wheel, on a single cylinder engine, which mathematically would mean skip teeth should be 24. After posting this, Matt told me that skip teeth cannot be greater than trigger wheel teeth, so I should change it to a 2 cylinder and use 12 skip teeth. So I did this, and set up my injector setting for 2 cyl, 2 injector simultaneous, 1 squirt per engine cycle, port injection. This should give me a single squirt during intake. I did not recalculate the req fuel, because it will change, so the only difference will be a wasted spark during the exhaust stroke. I read about the injector timing delay, and basically understood that it is the percentage of degrees between tach signals to fire the injector. Now with the "fake 2 cylinder" setup, I will be getting two tach signals per engine cycle, 1 every 360 deg. So, how will Microsquirt take this into account? Will it want to fire x degrees before both tach signals, even though I told it 1 squirt? Or, will it do one squirt, but during the second tach signal, into the exhaust stroke? How will it know the difference between tach signals? I could try 2 squirts/alternating I suppose, but still, I'm going to have a disconnected injector and I don't that one to be the one firing on the intake stroke. Should I just make a different trigger wheel, or is there something I'm missing? Thanks.

I don't know if I am wrong here but you treat it as a two cylinder, get one tach event per crank revolution, so you can use your 12-1 no problem. This would mean that ign1 (or ign2 but one output is not needed) would be a wasted spark signal essentially. If you selected alternating injection events (not sure if you can for 2 cylinders., haven't got access to the manual or PC at the moment) then wouldn't one of the injection events be x after ign1 too? So you are creating an inline twin and not connecting the spare outputs.

Yes, but how does it know which injector to fire at a given time. With my setup, it will be getting a trigger every 360 degress, and that's all it knows. It doesn't know if its on compression stroke of 1 or 2, and doesn't know which cylinder is connected to which injector. So, I can have injector 1 and ignition 1 hooked up, and leave inj 2 and ign 2 disconnected, but who's to say it wouldn't be firing ign 1 and inj 1 at compression of the non-existent second cylinder. I wish I could set skip teeth > actual teeth.

I don't remember if it was true with 2.890 code, but with the 3.430 beta code you don't need to fake the numbers to get a 1 cylinder sequential or a 2 cylinder wasted to produce the correct ignition and fuel outputs and the correct rpm. And you can use 24 skip teeth for 1 cylinder sequential. Read the setups at www.microsoft.info, left sidebar, Dual Spark link. They talk specifically about 1 and 2 cylinder, wasted and sequential. Then go to the link at the very bottom and get more info on the definitions of the terms used and more setups. Also go to the beta tester subforum here on microsquirt and read the release post and pertinent parts describing the changes for 3.430 and the new inputs for it, including the tie_2_cylinders input. While a lot of this is for the sequencer, it is also applicable to the micro, because that is the 1 and 2 cylinder equivalent of the sequencer, which covers 3-8 cylinders.

So lately, I made an external VR conditioning circuit, since I determined the one internal to microsquirt was dead (Tooth analyzer did not show any input, but did show a signal when ran through opto) Anyway, I set up my parameters ideally using the 2.891 code (12-1, 24 skip teeth, 6 delay). It showed speed as double, about 730 RPM cranking, and I had to cut delay in half to get spark to line up at the timing mark, then all I did was change cylinder and injector count to 2, and this seemed to fix things (I left delay teeth at 3, and skip teeth at 24) It seems strange that this setup would work, but everything seems to be firing on time.I will try the beta code the next couple days to see if I can get things to work with a more logical setup. I also need to go buy a new injector o-ring. It seems I smashed it after checking the flow rate.

Let me get this straight. You're saying this configuration will work, but only using dual spark, crank wheel trigger, no cam wheel. I thought you were talking about M-N with no dual spark. I tried the dual spark configuration, and I'm able to get the right speed and use the correct amount of delay teeth, getting relatively close to a spark where it should be, and trigger wizarded it the rest of the way. Right now, it seems like the injector delay isn't doing what it is supposed to. It seems too close to spark, even at 80%. Guess I'll keep checking things out.

EDIT:
Well, I figured things were close enough to try to start it up. I set up my ignition as dual spark with 12-1 crank wheel, and ran both ignition outputs to the coil. This seems to be the key to being able to set my advanced ignition options respective of what they actually are. I timed the engine up so that it fires at the advance mark on the flywheel when cranking (I'm not sure if it uses the advance table when cranking, but my timing light died so I'll have to check later.) I'm not sure about the injector timing setup. From what I read, it is the percentage of degrees of the tach pulses (so 720 degrees for 1 cyl) to fire the injector after TDC. I have this at 60%, which I believe is about halfway through the intake stroke. Anyway, I got it to start and continue running 2 times, by cranking at about 20% throttle and bringing up the throttle when it catches. It won't stay running if I close the throttle all the way though. Most of the time it doesn't start. Sometimes, it sputters, usually after I crank it for a little, then let it sit a few minutes. Nearly every time I crank it though, it will have continuous small backfires through the intake. Seems like a timing issue.

Here is what I would do now. Disconnect the injector wires, set table advance to 0 in the cranking region, then crank the engine and check the timing with a timing light. If it isn't where you want it, change trigger offset until it is. If you are sure about the mark on the flywheel is correct and that is where you want spark when cranking, then you are good. But make sure no other advance was used. You can do this by checking that you set trigger rise which causes spark as soon as you get a tach pulse. Or you can set "calculated" instead, and be sure the table advance is 0 in the load, rpm region where you are cranking.

Success there says you have timing and ignition correct. Next you need to verify you have a good tach input signal. Hook the injectors back up, start a datalog, and start up the engine while looking at the rpm gauge. If it keeps jumping to zero or to a very high rpm, then you have a noisy or weak VR signal. If it just dies out but the tach is solid (the trigger +/- column doesn't keep counting up or down), then you want to change the amount of fuel or maybe the advance - whatever you feel is the most likely.

Well, good news. I finally got the motor to start repeatidly and quickly . After checking the ignition timing and injector timing using a white LED, and finding everything in the right place, figured I needed to recheck my configuration. I originally had a 12-1, 24 skip tooth running a dual spark setup, but reread the manual and found this setup was only used with a cam sync as well. So, I changed to a 12-1 with 12 skip teeth, changed injector characteristic to "alternating" to keep the firing every 720 deg, and it fired right up. I wasn't sure how this setup would work initially, since I'm still not sure how it can tell TDC at compression and exhaust apart, but I'm guessing that now that it sparks at both locations (every 360 deg) it cleared up the issues I was having, and the injection timing is just 'close enough'. Also, when I tried to set alternating injection with 1 cyl, Megatune gave a warning, but apparently did it anyway. Right now, in addition to the great starting, it even runs pretty good through the limited VE map area I tested. The only issue was I had to give it a little throttle to start up and hold an idle, but that's probably due to a combination of the cold coolant temp, rich mixture and temporarily closed IAC. Hopefully I can get this tuned up in the next couple weeks.

To those curious, this setup is configuration 5a, illustration 3 in the dual spark section of the manual.