A REAL CHALLENGE for you Experts out there

I bought a 1992 Lumina 3.1/Auto trans recently. I got the car cheap because the owner played with everything under the hood and it wouldn't run. It has 90K

To get it running, I corrected the plug wires, (they were in the wrong order) and it would start and run very rich and get richer until it would finally die. It would re-start again but very rich. The Error codes from the ECM were:15, 21, 23, 35, 43, 53. I thought this was a bit unbelievable so I reset it and they repeated. I got a known-good ECM at the junk yard and swapped the Firmware and installed it. Now the engine will NOT start on its own unless I give it a little shot of starting fluid. Then it will start and run pretty good. Fuel Pressure is solid at idle and high RPM. Vacuum is around 22 while running but the idle is unusually high. The ECM error is 21. I have replaced the Throttle position sensor and at idle it is putting out 550 mv. The high idle is obviously from air entering the intake plenum. I removed it an replaced every hose and checked every connection. I re-installed it with new seals. I also replaced the EGR Valve. With the vacuum gauge attached, I shut off the ignition. The vacuum rapidly falls towards zero. IF I try to start it before the vacuum reaches zero, it will start successfully. If it falls to zero, it cannot restart without starting fluid.

How much vacuum should this engine show while starting? The Vacuum tank under the fender is not leaking and neither is the brake booster. Should vacuum drop so rapidly to zero at shutdown? I have sprayed starting fluid around the usual spots while running to see if there was any reaction--none.


Desparate Paul

where is your volt gauge reading? Code 53 is system over voltage, and since all sensor work off a reference signal, if the system voltage is too high, it will trip a bunch of ghost codes. Get your overcharging problem taken care of first, then decipher the driveability issues, as you will never get accurate sensor output with the system voltage being out of range. The vacuum drop sounds like it's a leaky fuel pressure regulator diaphragm, that's why you're not getting any change in idle when you spray around the engine. A pinhole in the regulator diaphragm will allow the vacuum to bleed off, and will cause hard starting or a no start until you prime it.

where is your volt gauge reading? Code 53 is system over voltage, and since all sensor work off a reference signal, if the system voltage is too high, it will trip a bunch of ghost codes. Get your overcharging problem taken care of first, then decipher the driveability issues, as you will never get accurate sensor output with the system voltage being out of range. The vacuum drop sounds like it's a leaky fuel pressure regulator diaphragm, that's why you're not getting any change in idle when you spray around the engine. A pinhole in the regulator diaphragm will allow the vacuum to bleed off, and will cause hard starting or a no start until you prime it.

Thanks for your reply. I must admit, I did not think of the fuel pressure regulator diaphram leaking. I have a pump-up vacuum tester an will test this since I currently have the intake plenum off again. As far as you diagnosis for the the voltage/ghost errors, you are right. I probably should have mentioned in my profile that I am an electrical engineer and that I design engine/Antilock brake/airbag control circuits for Chrysler. They all work in a similar fashion. There is one addition I will make in your statement about the system voltage versus the sensor measurements. In general, all of the control modules run from 9.8-15 Volts DC and have numerous, highly regulated temperature compensated DC supplies that "force" voltage to the outboard sensors. This is how a system like this can be so predictable in a wide range of temperatures. The sensor voltage is generally 5.0 Vdc. On the faulty ECM, the voltage sources were correct but many measurements "appeared" wrong because the ECM Digital voltmeter was bad. I actually located a faulty op-amp on the board to set my mind at ease after I replaced it.

I will test the fuel diaphram and let you know how I make out. There is also a large vacuum actuator on the transmission. Could this leak?

Thanks for your reply. I must admit, I did not think of the fuel pressure regulator diaphram leaking. I have a pump-up vacuum tester an will test this since I currently have the intake plenum off again. As far as you diagnosis for the the voltage/ghost errors, you are right. I probably should have mentioned in my profile that I am an electrical engineer and that I design engine/Antilock brake/airbag control circuits for Chrysler. They all work in a similar fashion. There is one addition I will make in your statement about the system voltage versus the sensor measurements. In general, all of the control modules run from 9.8-15 Volts DC and have numerous, highly regulated temperature compensated DC supplies that "force" voltage to the outboard sensors. This is how a system like this can be so predictable in a wide range of temperatures. The sensor voltage is generally 5.0 Vdc. On the faulty ECM, the voltage sources were correct but many measurements "appeared" wrong because the ECM Digital voltmeter was bad. I actually located a faulty op-amp on the board to set my mind at ease after I replaced it.

I will test the fuel diaphram and let you know how I make out. There is also a large vacuum actuator on the transmission. Could this leak?


I tested the fuel regulator diapram and it does not leak. I also tested the fuel pressure at start and it is 38 psi before the starter is engaged. It maintains that +/- 2 lbs no matter how much fuel I draw off the bleeder. I'm assuming that's OK. I still have the same problem; it will not start on it's own unless it gets a shot of ether in the intake air. If turned off, and started within a few seconds, it starts fine. If I wait until the system vacuum drops to 0, no start. What should the enginge vacuum rise to when it is cranking over? I have a new battery also. The error on the ECM is still Throttle position sensor voltage high.

Questions: Could I have a wrong version ECM module? I took it from a nearly identical car but ???

Could the throttle bypass valve be causing this? I have not checked it yet and the engine still idles really high.

start by correcting the high voltage issue, get a voltmeter, and check the batt without the car running, it should be around 12.6, if the batt is good, and not discharged, then start the car, I believe the ac delco alternators put out between 13.8, and 14.4 volts, if there is any more than say 15 volts going to your battery when the car is running, I would start by checking and/or replacing the alternator.