I've built an MS2 Extra parallel install for my '97 1.8 (currently stock), using Joe Perez's inverted ignition circuit, a JBPerf 4-channel board for sequential injection and a MAF input. Huge thanks to all who contributed to the wealth of information on this forum that helped me through that process.

All looked good on the JimStim, however when I plugged in the MS2 and started it from hot, I got 5s of VERY rich (AFR 8:1) running then it spluttered and died. When I reduced req_fuel by 30% and tried again it coughed then refused to start. I assumed fouled plugs but even after cleaning them it won't start even on the stock ECU and there's no spark. I'll work on getting the spark back (probably ignitors/coils I guess), but my main question is related to how I managed to kill the ignition system with my MS2.

As I said, I included Joe's inverted spark output mod in the original build, however when testing on the JimStim, if I ran with ignition outputs inverted in TunerStudio, the ignition LED's were mostly on, pulsing off, whereas if I ran them not inverted, I got the correct response on the JimStim (off, pulsing on). This clearly goes against Joe's instructions, but this is how I (stupidly?) ran it given that JimStim is seeing what the ignitors would see. I've double checked my wiring and it looks OK. D14 and D16 on the MS2 are constantly lit no matter whether I select normal or inverted spark outputs (seems correct looking at the circuit).

So, my questions are:

1) Should the JimStim ign LED's be nominally off and pulsing on when all is working as it should?
2) Is 'inverted' still the correct setting in TS when running the spark output mod?
3) If the answer to 1 and 2 is yes, any thoughts on why the ignition system got fried when my JimStim outputs looked good?

I've attached my .msq if it helps in any way. Many thanks in advance for any assistance.

 

with inverted you wont see them pulse on the LEDs, but it's still correct using Joe's output circuit.


now your coilpack is dead?

 

I assume you mean the MS LEDs, but do you know what I should see on the JimStim? The JimStim LED's seem contradictory to my 'invert' setting.

Not sure on the coil pack yet, I guess this would help diagnose the MS2 issue, but I was kind of hoping to do this the other way around! I checked the resistance across each pair of coil outputs and it looked sensible (~10kOhms), but I'm sure this isn't a definitive indication of a working coil pack.

 

if the coilpack is fine, how did it "kill the ignition system"

are you syncing when installed on the car?

 

I haven't diagnosed why I've got no spark yet, my first guess was the coil pack. Annoyingly it'll be a couple of day before I can make more progress in checking this out.

Yep, got good sync. I ran the car for a few days with the MS connected but ign and inj supplied from the stock ECU, so the MS was just watching. All sensors reading fine, good sync throughout the rev range. Being able to do that was a nice advantage of the parallel install, although clearly it doesn't test the output side of the MS.

 

OK, so as you've probably already realized, I'm an idiot. Here's how I worked it out...

I unplugged the harness from the coil pack to check the ignitors were getting 12V and to check continuity on the signal line from the ECU. All looked good so I connected a voltmeter between signal and ground and cranked the engine with the stock ECU installed. As expected on my cheap meter, the signal bounced around between 0 and 5V when cranking and was around 0V just with ignition on. Then I plugged the MegaSquirt in instead and did the same test. As soon as I switched the ignition on I got a solid 5V on the meter, giving me a small hint that I'd fried my ignitors/coils.

Sure enough, when I set my spark output to 'inverted' in MS, as stated by every single thread that refers to Joes spark output mod, I got the same response as with the stock ECU. My new coil pack will arrive later this week so I can confirm it then.

So, pending confirmation, it looks like if you wire your spark outputs with the extra transistor as described by Joe, the correct JimStim response is for the Ign LEDs to be fully on, with a short blip off when a spark is demanded (unless I've built my JimStim wrong!).

 

To button this one up, it WAS the coil pack and the car is now running again, wahoo!

Corrected the spark output to inverted and is now running nicely with megasquirt in control!