I'm running out of options...

[DaveSmed]

Its sounding like a hard cut thats intermittently restored. Odds are SOMETHING is up with the CPS signal, since it is unlikely something odd is going on in the DME (especially with a progressive degeneration) and you said you already eliminated the AFM.

To give you an idea, the issue that caused this exact same problem on a friend's car turned out to be a trigger wheel that had a bit of a wobble to it. It slowly started to work itself looser and get progressively worse. The quality of the signal was still good enough at low RPM, and even at high RPM (cutting out badly per noid light, using engine braking down a hill for a steady reading) the AC voltage still looked OK. A scope would have told the truth though, too much scatter.

Now, the trigger is a definite must check item, but that scatter can come from a lot of sources, a flaky diode in the alternator passing a bit of AC, aftermarket radio equipment, etc. There are a lot of variables. A scope would really help, but the temperature sensitivity you mentioned earlier tells me to check the AC voltage at the alt after you inspect (remove a bolt or two, look for wear) the trigger wheel.

[OrganicMechanic]

AJ--If you need more parts from my bins your welcome to them--I think you know where everything is...you did put it all there. Let me know if you need anything, it does sound like the cps signal degrading. I think I have a spare you can grab and thrown on there to confirm.

[OrganicMechanic]

also---check the injector sub-harness where it joins int the main--it is under the intake manifold and next to the block--I had a car doing strange stuff and it came from the sub harness pins being corroded (the green fuzz).

[Mossman]

Dave's loose trigger wheel sounds interesting, look into that. I'd also definitely hook a fuel pressure gauge up. T it into the feed line on firewall side of fuel rail. If your rail isn't seeing the proper pressure it can cause funny stuff to happen. Either failed FPR causing rail to see full pump pressure and motronic trying to deal with that (unlikely in your case cuz you're lean), but tired pump unable to keep up, or plugged fuel filter will cause low pressure, especially under load. In my IX it would run great for 3-5 minutes then as it was warming up run like absolute shit, unable to rev, couldn't even maintain idle speed in 1st with foot to the floor...etc. Ended up being a fuel filter with loose guts that magically shifted around and got plugged after a couple minutes of running. I changed EVERYTHING in the engine bay chasing it and it was a damn fuel filter. Moral of story, get a fuel pressure gauge.

[u3b3rg33k]

This is starting to sound like what happened to me when my CPS died. car got more and more pissed at life until it wouldn't run at all.

new cps -> happy days

Also, do what ^^^ says and check rail pressure. it's easy, and cheap.

[e30frosh]

This is starting to sound like what happened to me when my CPS died. car got more and more pissed at life until it wouldn't run at all.

new cps -> happy days

Also, do what ^^^ says and check rail pressure. it's easy, and cheap.

I ordered a new cps today. It seems from above posts that it would be my last option considering all the other parts have been replaced or swapped for known working ones. As for fuel pressure, it was checked on the dyno and was perfect until the rpm and ignition map went psycho.

Thanks for all the info guys.

[DaveSmed]

Hope it works for you! It seems strange that the CPS can work well enough at low RPM, one would think that its an all or nothing thing.

My guess is since the AC voltage generated increases in amplitude as well as frequency, it lies just under the threshold of the breakdown voltage for some part of the sensors construction until whatever RPM.

I need to buy an oscilloscope.

[u3b3rg33k]

Hope it works for you! It seems strange that the CPS can work well enough at low RPM, one would think that its an all or nothing thing.

My guess is since the AC voltage generated increases in amplitude as well as frequency, it lies just under the threshold of the breakdown voltage for some part of the sensors construction until whatever RPM.

I need to buy an oscilloscope.

Actually that's exactly what happened to me. the CPS "worked" fine at low engine speeds, but as rpms went up, quirkiness ensued. I think it has something to do with them being VR sensors, not hall effect.

My understanding is that as the rpm of the trigger wheel increases (be it crank or ABS), the voltage increases as well.

http://www.megamanual.com/ms2/pickups.htm says the VR sensor output voltage can vary from .6v to 50v, which technically qualifies as high voltage. While 50v isn't huge, it's still much hotter than 12v.

[DaveSmed]

50V in a sensor that has been soaked in engine oil for ~20 years oughta do the trick. Can't imagine that the dielectric resistance is too great there.

Ahh, old cars, whats not to love?

[e30frosh]

Bump up again...

Replaced the cps and still no difference. Sigh. Everything else checks out. I am now thinking it may be possible that my extra ecu is bad as well. I now need to locate a 173 in known working order to test. If anyone has one that they would be willing to let be borrow, I'll be more than happy to pay shipping both ways.

Cheers,

AJ

[Brian D]

Check the wiring from your afm, especially where it ties to the support rail on the side of the engine block (right behind the alternator). If those wires are chafed through at the support bracket, all kinds of weird shit will happen. Drove me nuts.

Good luck.

[OrganicMechanic]

check the injector sub harness as well---

[DaveSmed]

This sucks. It might be time to hunt down someone with a Vantage Pro or something comparable. I would really love to see your CPS output at the affected RPMs. I still think there is a major signal issue there. Sadly, I do know from my friends car who had the same problem, a DMM is insufficient to catch the problem. I was pulling my hair out until a particularly observant individual was watching the engine as it was running and asked "is it ok if the crank pulley wobbles a tiny bit?" Damned if I didnt feel dumb right then.

The beauty of the Motronic systems is their simplistic nature. Sadly, that can also be a curse as they tolerate NO malfunction of a few key sensors.

Really the entire fuel injection control comes down to the AFM and the CPS at its most basic, input wise.

Hard cuts usually point to the CPS, soft cuts the AFM. If those two work, and Motronic is processing data correctly, the outputs should work and the car should run as expected.