Rough running GM3.8L, technical description and update

This problem first presented in October 2002. The history is copied below.

An update:

This engine is still running rough, the tach reading is jumping and erratic, and code 18 is setting nearly all the time. Occasionally code 17 also sets.

Against my belief that the cam sensor was bad, I replaced it anyway. No improvement.

Since the last round of troubleshooting, I have just completed the following:

Checked all wiring continuity between the cam sensor, crank sensor, ignition module, and the PCM. All were under 1 ohm and with no intermittents when the wiring was moved.

All pins on the PCM were checked for pushed pins and grip tension. No problems found.

The ignition module was replaced with a brand new unit. No change. Original swapped back.

PCM was removed and the MEMCAL removed and reseated. No change.

Removed both serpentine belts and ran the engine. Still ran rough but no mechanical engine noise. Replaced the belt driving the supercharger and the previously observed engine rattle resumed. This same belt drives the A/C and water pump. The A/C was not engaged and the water pump spins silently. I conclude that the supercharger has serious internal problems.

I also conclude that the noise from the supercharger is not causing the knock sensor to produce a 'false' signal which would cause the PCM to retard the timing. I was conserned that this was a cause of the poor running.

I further disconnected the knock sensor and ran the engine. The rough running was present and I set the code for a knock sensor circuit failure. Reconnected and reset the code.

I grounded the bypass signal output of the PCM that instructs the ignition module to run in the 'module' or fail-safe mode with a fixed

10 degree ignition advance. The engine still ran poorly, no change detected.

I have 148K miles on this engine. Original timing chain. I put a timing light on the engine and found that at idle the timing mark jumped back and forth at least 10-15 degrees.

QUESTIONS: Is this a clue that the timing chain is shot? Could a bad timing chain cause my poor running? Can it also set a code 18 (cam/crank sensor)?

Is is possible that a fuel problem (clogged filter, clogged/dirty injectors, other) could cause a code 18 to be set? There don't appear to be any codes that get set for fuel problems.

On the old cars (before computers) both fuel and ignition problems often presented similiar running characteristics.

I have an old scope but it is hard for me to see if there are any substantial changes in the waveforms since my last look. The scope I used previously was a much better unit and I am going to try to borrow that one back to take another look at all the waveforms.

Robert Hancock- any suggestions?

Anyone else have any ideas.

All ideas for troubleshooting and suggestions are greatly appreciated.

Rich

----------------------------------------------------------------------------------------------- Previous thread from last October:

I would suspect that the cam sensor is just bad, it shouldn't have those spikes.

-- Robert Hancock Saskatoon, SK, Canada To email, remove "nospam" from snipped-for-privacy@nospamshaw.ca Home Page:

formatting link

I have a supercharged 3.8 V6 in a 1992 Olds 98 Touring Sedan that is > running rough at all engine speeds and loads. It doesn't stall, but > falters repeatedly. > > Code 18 always sets within 10-20 seconds after starting with the SES > light on. This is the cam/crank error. Sometimes the SES light will > extinguish with the history code of 18 still in memory. > > On occasion, code 17 (Spark ref circuit) also sets and even less > frequently code 42 (EST circuit) sets. No other codes set. > > History: > About 15 months ago the dual crank sensor failed intermittently only > when hot. I put a scope on the signals and observed the 3x signal to > quit pulsing. The engine would quit soon after that. No restart for > about 5-10 minutes, then would run for another 10 minutes and repeat. > Replacement of the dual crank sensor fixed the problem. > > Currently: > I have scoped the signals and found the following: > > Cam sensor output: solid 0-5V pulses, sometimes with random spikes to > 0V > > 3X crank sensor output: 0-5V pulses, all correct widths > > 18X crank sensor output: 0-5V pulses > > FROM THE IGNITION MODULE: > > Cam signal output: 0-5V pulses, has some occational spikes to 0V > > Fuel Control output: 0-5V pulses > > Spark Ref output: 0-5V pulses > > PCM OUTPUTS: > > Bypass output to ignition module: 3V-5V pulses, occasionally a missed > pulse. > > EST output to ignition module: 0-5V pulses > > > MY QUESTIONS: > Should the Bypass output of the PCM be a solid 5V instead of the 3-5V > pulses that I observe? > > What am I missing with this problem? > > > Other info: My supercharger has a bad bearing and makes a lot of > noise. A possibility that I investigated is that the detonation > sensor was (intermittently) picking up the noisy bearing and causing > the PCM to command retarded spark timing and thus the engine would > falter. A scope on the detonation sensor output showed a 2V dc level > with a 0.3V p-p AC signal- this seems normal. However, there were > also narrow 5us spikes from 2V to 1V on the detonation sensor output. > > Does this detonation sensor signal sound OK? > > Bottom line: I don't know what to look at or do next. Ideas anyone? > > TIA, Rich >
Reply to
G. Richard Stidger
Loading thread data ...

See below.

Reply to
Robert Hancock

I'm not entirely certain, but I think they will generally start like that, but generally run rather poorly..

Reply to
Robert Hancock

Robert,

Thanks for your input and explainations. I am borowing a better scope to look at the signals. Will post results, hopefully tomorrow.

Rich

"Robert Hancock" wrote:

Reply to
G. Richard Stidger

---------------------

Reply to
kc8adu

Yeah, that looks like a fairly dirty transition, some glitches down to zero volts. It wouldn't surprise me if that caused some problems, the ECM might see those as extra reference pulses. (I'm assuming the scope probe was solidly connected to the wire, it wasn't loose or anything..)

Reply to
Robert Hancock

Can't see the scope picture but from what you're describing, it sounds like there may be mechanical interference between the 18x sensor and the shutter wheel. Did you use the appropriate spacing tool when you changed the crankshaft position sensor?

If you can, try synching the 18X waveform to a secondary ignition event to see if they relate, i.e., the downward spikes are actually the secondary ignition firing.

IIRC, in an earlier post, you described the by-pass circuit as toggling from 5 volts to zero volts, did you ever resolve

-why- that is?

#1 rule when working on any GM DIS ignition system, make damn sure that the ignition module mounting is bright, clean and shiney. This -is- the primary ignition ground path, and if it is dirty or corroded, the ground will try to go through the logic low circuit (black/red) and make a real mess of everything.

Reply to
Neil Nelson

MotorsForum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.