93limited

Hi folks, I posted a week or so ago about a no start I have on a '93 GC 4.0L-- see below. I did all the tests I could do from the shop manual and dug in to change the timing chain- it was high miles anyway. Well, the timing and the chain were fine. I thought I got lucky when I saw the timing gear marks 180 degrees away on the compression TDC, but turns out that's factory. My distributor was indexed correctly for that.

So, out of ideas, I thought since it had that flooded start sound, I'd disconnect the injector harnesses and see. Sure enough it fired off and would run pretty well just spraying starter fluid in the intake! Again, I've tested the injector resistance and they are a little high 14 ohms vs. 12.5 spec, but they are all the same and figure couldn't all just go bad. I've got 38 psi at the rails with the ignition on, so figured the fuel pressure regulator is OK?

So my only thinking is the ECU is firing the injectors too often or just constantly. No error codes, ECU sensor voltage and check engine light checks fine. Thoughts? ECU? pressure regulator? something else? Thanks!


from previous post:

93 GC 4.0L. 300K+ miles. Was running great, no symptoms, just passed emissions, then a week laterwouldn't start. Cranks over fine and tries to start (gets a few combustions-- like you get when flooded). OK compression, 38 psi at rail, confirmed hot spark, new and tested CPS, tested both ASD and fuel pump relay, tested cam position sensor (so know rotor is moving), coil, plugs, inspected rotor and cap, no wobble on distributor shaft (aftermarket distributor installed 50K miles ago). TPS and NSS switch test OK and security system seems OK. PCM is alive and seems fine. Only error codes are usual 12 and 55 (but I disconnected the battery as part of the testing early on). I've let it sit overnight as thought was flooded, but no help and the plugs look dry. If I pump the pedal while trying to start I can smell gas so it's getting gas. The only suspicious thing is when checking spark, it seems that I get a spark only every second or so, which seems way too slow. The starter is kicking it over at normal rpm. I even sprayed some starter fluid in the throttle body with no effect. I'm out of ideas other than just start replacing things.

93limited

Been rechecking all the sensors that might lead to way over rich at start condition. All input voltages check good at 5V. TPS tests OK and I tried to start with disconnected, coolant and air inlet temp tests OK, didn't check O2 as changed a few years ago and as i understand can't make too rich at start in any failure mode. Didn't mess with IAC since don't think it will cause a no start and replaced a few years ago too. The only thing a bit off was MAP sensor- reading 3.9V instead of 3V.

Thinking about trying an OBDII. But since the engine won't run and ECU has been cleared since the problem, can it pick up enough just on trying to start?

Thanks

93limited

So finally fixed the problem. Changed brain box, timing chain, coil you name it. Tested every sensor involved and all were good. Just knew it had to be a sensor. Turned out to be the coolant temp sensor-- that's it. Went from no start to running great. Temp sensor tested out fine with voltmeter. Read somewhere that if you ground it rather than just disconnect it you can test it- sure enough.

kjc

Good job trackin' that down. Could you elaborate on the testing part? I understand about the sensor testing ok with the meter, but still defective...
so, if you disconnect the sensor- no start, but ground the sensor(short it) and it will start..? Indicating the defective sensor?

93limited

Right. There's a resistance test - OK is 17.9K to 10.8K ohms cold and then 5V coming from harness with ignition on. Mine was 20K with the temp in the fifties. I figured that was close enough- guess not. The resistance goes down as the temp increases and the computer starts leaning out the mixture and visa versa. When you just disconnect the sensor it reads infinite ohms or says the motor is really cold. I'd throw a 22 code when I did that. So to simulate a hot engine and lean out the mixture, you disconnect the harness from the sensor and ground the harness connection out. Paper clip into the two female connectors and grounded to battery. Mine started right up then. I replaced the sensor and it has been running fine. I can only guess that the old sensor was telling the computer it wanted to be just a bit more rich that the old, low compression engine could start on. Like I said, it wasn't super rich, the plugs weren't wet, but it was too rich to start. Either that or I've got some problem with the connector and messing with it did the trick for now. My thinking anyway- I'm not a mechanic.