1999 Park Avenue timing goes off

[procharlie]

I am helping a lady with her 1999 Park Avenue. She has already spend all her money to have Idle Control sensor, ignition control module, spark plugs/wires, and coil packs replaced and still has a problem. The car will sometimes run OK, but then runs very rough. I drove it with my error code reader connected and looked at data, and found the ignition timing went wild when it ran bad. The timing was sometimes normal, around 20 deg advance usually, but then when it started missing the timing fluctuated rapidly ranging from small positive to negative values. It generated an error code for random misfires, which is not surprising. I don't know what could cause this except for a bad cam or crank sensor.
I read about the cam pickup magnet coming loose from the cam gear, but wouldn't that stop it altogether? I am thinking of just replacing the cam sensor first since that is easiest, then maybe crank sensor if necessary. If anyone knows of any other causes for this problem I would appreciate the help.
Thanks

 

[ryenie]

I would run the car a few seperate times and try and mimic the issue and hopefully produce a more helpful code than random misfire. Other than checking the magnet, you need an oscilloscope to properly test the cam and crank sensors themselves, otherwise it's just a visual inspection to the sensors to look for obvious signs of damage, you can check for a voltage spike using a voltmeter while cranking but it won't tell you what you really need to know. The engine can run without the cam sensor although it will run rough. If it doesn't receive a signal from the cam sensor the pcm will fake one based on the crank sensor reading. So it's definitely worth taking a peak at the pickup magnet and the cam sensor itself, if it's loose it can cause issues. It might also be a good idea to check the ground coming off of the icm. It should hook to one of the bolts underneath the bracket it sits on. If it's corroded or rusted it could cause issues. As well as the icm not having good clean contact with the coils and the mounting bracket. The crank pulley and crank sensor would be the next thing to inspect. I'm pretty sure the crank sensor is where you get the timing reading from. If it get's full of oil, dirt rust rocks, etc. it will cause issues as well, or if the metal ring on the backside of the pulley is damaged it won't read properly. So there's still quite a few possible causes, but hopefully this can help you.

 

[procharlie]

thanks for the good suggestions. I will look into them and let you know when I learn more.

 

[procharlie]

Update:
I replaced the cam position sensor with a known good one, and it didn't solve the problem. I checked the ICM ground, looks good. One thing I have noticed is that when driving it now, the timing goes bad much more when I rapidly depress the throttle pedal. If I slowly depress it, it seems to run better, but not perfect. I looked at the TPS reading while depressing the throttle pedal and it looked very good, steady, from 0% to about 80%, as far as I let it go, so I don't think that sensor is bad.
If anyone knows why the timing would go bad more when giving it gas, I would appreciate any feedback. I will hook up my fuel pressure gauge and monitor fuel pressure to see if that is OK.

 

[ryenie]

Did you make sure the magnet was still on the cam gear when you replaced the sensor? The sensor won't work if the magnet is loose or fallen out, otherwise i would think there is an issue with the crank sensor. The knock sensor can also affect the timing. If it detects a knock or ping it will retard the timing to reduce the knock. It's simple to diagnose. If you take a socket extension and tap it on the metal so that it bounces back off of the engine it should shudder slightly, might be more noticeable on the live data, but you should be able to notice it. If it doesn't shudder i would say it's bad. But i would think the crank sensor is the more likely culprit.

 

[HotZ28]

If the engine has a misfire, the timing will fluctuate! The cam sensor magnet dislodging/breaking was more common in the early 90's, but not very common in 1999. You need to look at individual cylinder misfire (counts) on the scanner to see where the problem is. Random misfire could be one/two cylinders, but that can be determined with a misfire capable scanner.