At my wits end... ANY IDEAS? LS3 with 24X trigger.

[DLRacing]

The machine shop finally got the parts done.
Hopefully I get a chance to mount it and try it out this weekend.

[DLRacing]

Finally finished up my generator install and got back to work on the car today.
I got the reluctor wheel drilled and mounted and fabbed a bracket for the sensor.
Also made an extension harness.

[nathaninwa]

I'm excited to see this come together

[DLRacing]

Worked on it a little tonight.
I setup the trigger parameters, and scoped the CKP and CMP inputs while I turned the motor over by hand to determine when to poll the CMP sensor.
CMP goes low at about 1 deg BTDC on #1 compression stroke, and goes high at 4 deg BTDC on exhaust.
Given that I have 360 degrees of crank rotation to do so, is there a preferred point to poll the sensor, or just use tooth 1 (40deg BTDC) and go?

Thanks,
David

[nathaninwa]

I might be wrong here. But I think you set it up with your number 1 trigger angle and poll level going low to check for cam sync

[DLRacing]

Thanks Nathan, That's what I set it to for now.
I plan to check cranking timing before I try to start it up.

David

[DLRacing]

OK, I checked cranking timing. Set it to 0 and read about 4 deg ATDC, so I changed tooth 1 angle from 40 to 36 deg BTDC, which brought actual cranking adv to 0.

I then enabled the fuel pump and tried to start it.
It coughed and sputtered a few times, then backfired on a couple attempts, both through the intake and exhaust.
That's all I had time to try this afternoon. I plan to dive into it for a while tomorrow.

Logs attached. Still seeing multiple sync losses.
Anyone have any ideas? I'd love to hear them...

[billr]

Set it for "wasted COP" until you can get it starting/running peacefully.

Oh, and if you get no joy with that, I would add three more temporary timing marks on the crank so you can check timing on all cylinders. Nothing accurate, just pieces of tape on the pulley or trigger-wheel to be absolutely sure you ignition firing order is correct.

[gjestico]

I hate to say it but that wheel sensor mount is kinda flimsy. Also you are using a GM 58x sensor on a 36-1 wheel ?. I am not sure of any of that matters.

Also, How exactly have you got that gray 58x sensor wired ? You are probably aware its wired different from a 24x sensor.

[DLRacing]

I hate to say it but that wheel sensor mount is kinda flimsy. Also you are using a GM 58x sensor on a 36-1 wheel ?. I am not sure of any of that matters.

Also, How exactly have you got that gray 58x sensor wired ? You are probably aware its wired different from a 24x sensor.

Thanks for the input, but the sensor mount is 7075-T6 alloy. I can pretty much hang on it without it bending. The sensor gap is about .030".

The 58x sensor is designed to read the single row 60-2 GM reluctor. I see no reason why it shouldn't work on any standard wheel. GM's 24x sensor is the one with 2 elements and is specifically for the double row 24x wheel.
I scoped the CKP signal and it looks fine.
Also, the pinout on the 24x and 58x crank sensors is the same. It's the cam sensor pinout that differs between the two.

Thanks,
David

[Rick Finsta]

How is your ignition system grounded on the secondary side of the coils?

Try scoping the outputs as you attempt to spin it with fuel disabled and ignition enabled and see if you're having noise issues. That happened to me and I got all sorts of problems with sync loss, resets, and even wiped firmware!

Just a few other things to check.

[DLRacing]

How is your ignition system grounded on the secondary side of the coils?

Try scoping the outputs as you attempt to spin it with fuel disabled and ignition enabled and see if you're having noise issues. That happened to me and I got all sorts of problems with sync loss, resets, and even wiped firmware!

Just a few other things to check.

The coils are grounded through the MS3 harness directly to the cylinder heads.

Do you mean scope the coil outputs at the MS3Pro? I'll try that, although I'd expect to see a bit of noise feedback from the coils there, considering the energy they produce. I'd think those outputs would be sufficiently protected to protect the ECU from damage.

Thanks.

[jsmcortina]

The coils are grounded through the MS3 harness directly to the cylinder heads.

Could you explain exactly what you mean by this?

i.e. the ground wires from the coils must go directly to ground, not via the ECU.

James

[DLRacing]

Set it for "wasted COP" until you can get it starting/running peacefully.

Oh, and if you get no joy with that, I would add three more temporary timing marks on the crank so you can check timing on all cylinders. Nothing accurate, just pieces of tape on the pulley or trigger-wheel to be absolutely sure you ignition firing order is correct.

I'll try wasted COP again and see if that gets it started.
It was running before, and no changes were made to the coil wiring so the firing order is correct.

Thanks.

[DLRacing]

The coils are grounded through the MS3 harness directly to the cylinder heads.

Could you explain exactly what you mean by this?

i.e. the ground wires from the coils must go directly to ground, not via the ECU.

James

The coils plug into the coil subharness, which plugs into the Megasquirt harness, which has coil ground lugs which are bolted directly to the cylinder heads. They are grounded exactly as the guys at Megasquirt designed them to be, and pretty much the same as they were in the GM harness.

[jsmcortina]

OK, that sounds fine. I forgot this was a drop-on harness.

James

[DLRacing]

The noise filter was what was preventing it from starting. It was still setup for 24 teeth, and the timing was too aggressive.
I shut it off and changed to semi-seq and wasted COP and it started.
Set it to 20* fixed timing and tweaked tooth 1 angle to correct timing.
Timing does have some jitter, about 1* or so when idling at 1000-1200rpm.
Sync losses again appeared within a minute or two of running.
I recalculated the tooth durations and enabled the noise filter with new values, but it again seemed to interfere.
I reduced those times significantly to see if I could get it to be effective, but it still doesn't help.

With the 36-1 wheel, this should now be identical to hundreds, if not thousands of other installs out there, but it still won't maintain sync.

I still believe there's a hardware problem with this ECU. I wish there was someone else in the central VA area with an MS3Pro that I could swap ECU's with to find out for sure.

[nathaninwa]

I'd be willing to test your Pro in both my cars. 58x LS and 12 tooth vr sensor 2jz. Both projects run great

If the weather holds out, the next 3 weeks are test and tunes for my LS car

[DLRacing]

I'd be willing to test your Pro in both my cars. 58x LS and 12 tooth vr sensor 2jz. Both projects run great

If the weather holds out, the next 3 weeks are test and tunes for my LS car

Wow, that's a long way from VA to WA.

If I can't locate someone closer, I may ship it out to you and let you have a go at it.
I want to give Matt at DIYAutotune a call first, and see if he'll swap it out and test it himself.

Thanks,
David

P.S. Anyone else in this area, please respond if you can help in any way.

[billr]

There is one member here who continuously notes he "is in central VA and willing to help anybody", have you not contacted him?

Edit: that is "CRStune", and is in VA/DC area, not "central VA". Close enough? Maybe not an MSPro, though...