Diagnosing problems with NemiSys scan tool

I've been trying to diagnose an intermittent drivability problem in a '95 Buick Park Ave Ultra (supercharged 3.8L V6) for some time. Briefly described, there will be a sudden jerk or shudder while driving. It occurs most often at freeway speeds, which sometimes causes the TCC to pop in and out which is VERY noticeable. But also sometimes happens when accelerating away from a stop when the TCC is not involved. I'm pretty much convinced it's an engine problem, and I've replaced a bunch of stuff... all the usual suspects, so to speak.

The last week or two I've had the use of a very nice scanner, the SPX NemiSys, that allows me to capture a frame of the OBD-II datastream at the time of an incident. After peering at a lot of frames I think I may be on to something. It seems that when it happens there is a sudden drop in injector pulse width. The following URL is an example

formatting link
At the top you can see I'm cruising at about 63 mph when there are 2 blips in vehicle speed, the second of which is followed by a dip in speed. The throttle angle is constant through all of this. What I'm now focusing on is the sudden drop in injector pulse width that corresponds to the speed glitches. The time steps here are about 1/4 second (I think), so the pulse width changes take place over 1/2 to 1 second, dropping from around

7.5 ms to about 0.9 ms. Not shown are the coolant and air temperatures, but they are constant. The O2 sensor output is shown. It is jumping around a lot, which I believe is normal, but it does seem to go lower at the time of the pulse width drops. My guess is that means the O2 sensor is just reporting the lean condition the results from the pulse with drops.

So, is there anyone here that can tell me where to go from here? What could be causing the pulse width to fall off like that? FYI, the coolant temperature sensor and MAF sensor, fuel pressure regulator and filter are new, as are the plugs, wires, ignition module, throttle position sensor. The O2 sensor has not been changed. BTW, I get no trouble codes of Check Engine lights. Finally, let me say I'm doing all this because the dealer and my long-time independent shop has had their go at it several times.

TIA

Ed

Reply to
Ed
Loading thread data ...

Interesting problem. I'm not sure there is enough information here to formulate an accurate diagnosis, but I can venture a couple of guesses. Perhaps a glitch in the speed sensor signal. One that might not show up on the scanner, but might if you scoped the sensor signal wave form. I would also like to see a pattern of intake pressure synced with ignition during this episode. Maybe you are just having a plain old induction back fire due to a mechanical or ignition miss. I don't suppose you have a portable ignition scope to watch the secondary events during the problem. That sure would shed some light on what is happening. Another thing in your graph puzzles me. Why is the speed and engine RPM holding while throttle position is decreasing? Where you going down hill? There again, a TPS or MAF signal glitch that might not show on a scanner. You said some parts are new, but how about the wiring and connectors. Many times, intermittent problems are due to connection problems. Is this problem predictable or repeatable under certain circumstances? In other words, does it occur only once warmed up or mostly at certain load conditions etc. There in may lie the clues to finding the solution.

I guess I could think of some other possibilities, like maybe a sudden rise in fuel pressure, or even some alternator frequency induced PCM malfunction (that's kind of exotic), but the lack of a DTC makes me think I would be looking for something more basic like Ignition system or valve train. Maybe even fuel contamination (water?).

Good luck,

Reply to
Kevin

============= =============

Research the system you're working on and understand the input/output of the computer functions before you do anything when it comes to diagnosing a symptom that's not readily identifiable using "quick" scan and performance checks.

with that said............

scrutinize the cam/crank signals.........with a lab scope. synchronize the signals with the injector signal and you may find that the sync signal is glitching at the moment of the injector glitch.

Injector timing is output using engine sync. signals.

~:~ MarshMonster ~:~

Reply to
Marsh Monster

I know. I feel the same way :-( Somehow I get the feeling this shouled have been carried out in a GM research lab!

I've been trying to figure out if the car id fitted with a MAP sensor. I don't think so, since a search in Alldata for it brings up nothing. GMPartsDirect.com brings up a part of that description, but when I sent them a question about it I was told "not in stock," so i think it doesn't exist. It does have a MAP sensor, which I've always understood to be an alternative to the pressure sensor. the MAP sensor was replaced, a $300 experiment by my trusted garage. No help whatever.

I don't ever hear anything like a backfire.

I have little Hitachi scope that I use in the garage, but don't think it would work very well plugged into the cigar lighter.

Could have been a slight grade. The scanner will record up to 5 episodes, and I have trouble remembering what was going on at the time episode #3 occurred, if you know what I mean.

I am also thinking along those lines, since there has been a lot of work over the years that may have given techs the opportunity to tug and twist. Hard to know where to begin. I've spent some time looking over ground path schematice in Alldate, but maybe I should do some more.

I've been looking for things like that. Here is all I'v ecome up with. If I drive down the freeway about 20 miles nothing happens on the way. If I stop for a latte (drinking it there... 20 minutes or so)and then come back it will act out on the way back. Also, sometimes it will stumble pulling away from a stop on the first drive of the day...

2-5 minutes away from the house. Trouble is, these two scenarios are diametrically opposed.

Some people at alt.autos.gm jumped on fuel pump right away. That's a $400 experiment, not carried out. I did replace hte regulator and filter. No help. Also put a gauge on it, in the garage. In spec. It really irritates me that OBD-II does not give a feul pressure signal. Must be a safty issue.

I was thinking alternator problems for a while, for 2 reasons. One, the red light comes on occassionally at start up and it never used to do that. Second, somewhere I found a GM bulletin about a S-10 having a TCC problem due to a wierd frequency interaction with the alternator. The fix was a special part that I suspect was a filter of some kind on the alternator output. However, people on some forum told me the scans I was getting are normal.

Never though about fuel contamination. We liv ein a moderate climate (So. Cal) and buy fuel only from major stations. How woudl I follow up on that?

Thanks. Your interest is appreciated.

Ed

Reply to
Ed

You could check for fuel contamination by emptying the contents of the fuel filter into a clear container and letting it sit. Any contaminants (other than alcohol) should separate and become visible (including water). You can check the alcohol content by adding a small measured amount of water to a measured amount of fuel. After it sits and separates the volume of water will have increased by the amount of alcohol present in the fuel. Use a calibrated beaker so you can see the difference. OTC also sells a kit to measure vapor pressure in order to determine fuel volatility, but It might just be easier to switch fueling stations for a while to see if there is an affect on the symptom.

Reply to
Kevin

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.