Mike's 87 Chevy Camaro 383ci engine (14th)

[malcom2073]

The car starts! It starts faster than the old decoder did even, woohoo!

Fairly certain the ignition module fixed the dying issue... time will tell on that one. Now I've got a new beast to tackle: timing!

Got the timing hardware put on my Jaguar A3, but when I plug it into the car, it looks like it something is seriously wrong with my settings, because every time I plug in the timing cable the car dies. Fred had a couple good points to try, increasing bypass cutin/cutout, reducing coil dwell time, and ensuring my decoder is triggering on the correct polarity (rising or falling side of the signal).

My other car has dropped out of comission due to the clutch taking a crap, so this is on the backburner for a couple days. Tons to try once I get back into it though!

[malcom2073]

The car is now running with FreeEMS timing control!!

So I lost my timing light and was doing this by feel (BAD idea). Turns out what I thought was 8 degrees, was actually 28 degrees. My car likes about 20-25 degrees of base timing (Timing with the ignition module doing it, rather than FreeEMS control).

I also learned that with my hardware at 20 degrees BTDC, my decoder offset is actually NEGATIVE 20 degrees. This explains why it did not work previously. That would be 700 degrees (of a 720 degree cycle). However, because my decoder syncs every 90 degrees and doesn't care which tooth it syncs off of, I could just do 90-20 to get 70 degrees of offset. Put that in (70 * 50 == 3500, 0x0DAC in hex) location id 0xC003, and presto engine started and ran! I don't even have the bypass in, so it cranks on FreeEMS control too!

Tried to do some steady state tuning, got it up to 4000rpm!! (3800, close). Engine is scary loud and mean sounding. Fred suggested I introduce an intentional air leak (basically static IAC), so I'm going to turn in my idle screw a bit, try to get it a bit more air. This should let me idle below 1000rpm, and run much better.

Overall a very successful day. Amazing how much the right tools for the job make all the difference.

All in all, it's seeming like I'll get this car running before I manage to fix the transmission in the other car, HAH!

[Fred]

The car is now running with FreeEMS timing control!!

My evil plan unfolds. Eventually all cars will run this way. Bumping into one in a carpark will transmit the disease and it will spread like wild fire with the quality of supermarket carpark drivers, in general. mu ha ha ha.

I also learned that with my hardware at 20 degrees BTDC, my decoder offset is actually NEGATIVE 20 degrees. This explains why it did not work previously. That would be 700 degrees (of a 720 degree cycle). However, because my decoder syncs every 90 degrees and doesn't care which tooth it syncs off of, I could just do 90-20 to get 70 degrees of offset. Put that in (70 * 50 == 3500, 0x0DAC in hex) location id 0xC003, and presto engine started and ran!

Say "Thanks for pointing that out, Fred" :-p :-) You're welcome.

I don't even have the bypass in, so it cranks on FreeEMS control too!

I definitely do NOT recommend this. Please set it up properly like Andy's car. Your cranking timing will be very inaccurate this way unless you have it setup a certain way, and even then it's not great.

Tried to do some steady state tuning, got it up to 4000rpm!! (3800, close). Engine is scary loud and mean sounding.

Pffft:

All in all, it's seeming like I'll get this car running before I manage to fix the transmission in the other car, HAH!

Classic! Good work, Mike! Great to see some progress!

Fred.

[malcom2073]

I lied, the car never ran under FreeEMS timing control. Turns out I forgot to put the bypass jumper back in, so it was running in bypass mode the whole time (ignition module control).

I hooked a Logic analyser up to the Jaguar while it was in the vehicle to try and verify some things, and I found out some interesting facts.

I did find out that FreeEMS control for timing while cranking is bad, hence the bypass. This is because of the significant difference in timing of the pulses of my decoder ring during cranking. Each tooth is 90 degrees apart, which is a HUGE amount of time for the engine to change velocity at such a low rpm.

I also figured out that I'm firing on the wrong side of the signal, rising edge. I should be triggering on the falling edge since that's when the ignition module fires (dwell 2.5ms before, fire on the mark). Slight decoder code change and I'll be back up and running again.

Only thing left to do is make that change, get the car running and idling with ignition module timing, then hook the LA up again and verify that the Jaguar is doing the proper thing before I plug the bypass in and let it take control! Hopefully today

[Fred]

The timing being bad while cranking should come as no surprise. I swear I told you that 10 times. GM did the bypass for that reason: It's not possible to do a good job of that, really.

Re wrong edge, not surprising either, and once again, I told you to check this stuff and verify it was correct. Good that you now have, you can now actually make progress rather than stabbing in the dark.

Looking forward to more progress. With 20btdc timing at idle, you can likely lean it out to near stoich and stop fowling your plugs.

Fred.

[malcom2073]

http://www.youtube.com/watch?v=JqbUuRh3JAU

[malcom2073]

So I made the change to the firmware to have the signal on the falling rather than rising edge, and got it running with FreeEMS timing control.

It seems to idle around 500-600rpm, I have to pull a vacuum hose the size of my index finger to get it to idle when cold, otherwise it just stalls. It lurches over a few times at 300-400rpm then dies otherwise.

I pulled the FreeEMS timing control, and it didn't seem to have any effect (FreeEMS timing control at 20 degrees, with -20 degree offset seems to be identical to ignition module timing, this is GOOD). Which likely means that the low RPM and running rough is due to me changing fuel timing to be on the falling edge.... which SHOULDN'T have that effect from what I've been told

Tomorrow I will flip back to the rising edge and test it, if it still runs like crap I'm going to be tearing apart the intake to see if something fell apart on the throttle body or inside the intake. It's running dangerously rough, and it's even made a sound like backfiring out of the intake twice now, but there's no MAP spike... so I'm not entirely sure what that nose is. It's starting to scare me :/


Edit: Something I thought of. When I was testing FreeEMS timing during ignition, it couldn't handle 200-300rpm, due to how much change there is in the signal between the teeth (90 degrees apart). when it's running at 500-600rpm, it's running really rough: timing signal, even from the ignition module is jumping around a LOT on the timing light. I wonder if it could be jumping around enough to misfire or even back fire into the intake.

[Fred]

All sorts of shit is possible. But: you said without FreeEMS timing it's still low and rough. You're running multiple shots of fuel, so they are evened out, fuel timing can't change that. What it can change is your MAP sampling point, which can change your MAP reading, a lot. This could explain the tune change, but not the roughness or slowness once you bumped VE.

Also keep in mind that you have a dizzy... if you have a timing range of say 15 - 60, then you're using HALF of the possible range where you're LIKELY to get a spark on the CORRECT cylinder. You then have 22.5 degrees each side of that where it should still work OK, but increasingly there becomes a chance it'll jump to a neighbouring cylinder instead as the rotor and cap are not perfect points and do not have perfectly even surface voltage requirements. IE, your dizzy must be physically timed reasonably. You can't just mess with it like you could a crank wheel etc.

"I wonder if it could be jumping around enough to misfire or even back fire into the intake." << No, but it could jump to the next cylinder if it's that bad.

(FreeEMS timing control at 20 degrees, with -20 degree offset seems to be identical to ignition module timing, this is GOOD) << yes it is. Don't mess with this circuit/code alignment.

Fred.

[malcom2073]

So things learned today: (When I say "edge", I mean edge of my crank position pulse)

1: When running on rising edge (wrong edge), I can run with about 25 degrees of timing on the OEM ignition module, get about 1.0 lambda and idle at around 500-600rpm.
2: When running on rising edge (wrong edge), I can run with about 25 degrees of timing on FreeEMS, get about 1.0 lambda and idle at around 500-600rpm.
3: When running on falling edge (correct edge), I can run with about 30-35 degrees of timing on the OEM ignition module, idle at around 400-500rpm, but has to run super rich (0.8 lambda)
3.1: I moved my MAP ADC reading to the rising edge, in an attempt to rule out the map being read at a weird spot, it had zero effect.
4: When running on falling edge (correct edge). I can run with about 58 degrees of timing on FreeEMS, get about 1.0 lambda and idle at around 900-1000rpm.
5: Hot engines are hot. Don't try to pull the distributor out of a hot engine, you'll burn yourself. Multiple times.

All these things (1-4) were verified with a timing light and logic analyzer. All FreeEMS runs were ran with the proper decoder offset for the particular distributor offset.
1 and 2 would be semi-acceptable, except for it being such a low idle, and the fact that it's syncing off the wrong edge of the signal.
3 and 4 are frustrating, because I'm supposed to be on that edge.

So there IS a configuration on which the engine runs correct and well... but it's the wrong configuration.

Also of note, when running off the rising edge, my WBO2 is fairly smooth. When running off falling edge, it's spiky and very rough looking in the log.

I don't know, I'm at a loss where to go at this point.

[Fred]

LOL @ 5 :-)

Where is 6 "put it on the correct edge, added some air flow to raise min idle, drove it and got it rough tuned" ? :-)