I've been trying to diagnose this problem for awhile now and it has me stumped as to what exactly it is. I hate throwing parts at vehicles. Anyway, the problem is intermittent, and not always present.
Intermittent problems are usually electrical in nature, but can be something like bad gas or a stuck valve upon occasion.
What is the proper way to test the EGR and isnt the EGR used mainly for a cold engine? I can remove the vacuum hose from the top of the EGR at a warmed up idle and nothing happens.
No, the EGR is for a warmed up engine under load at speed. It opens at that time to admit a bit of exhaust gas into the
intake to reduce NOx emissions. It closes at idle and is partly open under light load conditions. If you remove the hose at idle, it should open and the idle should get lumpy, with a slight improvement when you plug the open vacuum line. No vacuum on the line indicates a leaky vacuum line or failure of the EGR control solenoid, no movement of the EGR pintle indicates leaky EGR diaphragm. If you can see the EGR valve moving when you connect/disconnect the line then you probably have plugged EGR passages. This won't cause what you are describing though. What could be an issue is if the EGR valve is sticking open sometimes. Mind you, that will affect idle far worse than part-throttle operation.
The Cat's seem fine with no glowing or excessive heat to suggest a plugged condition. Is there a for sure way to check for a plugged cat?
Put a vacuum gauge on the car. Start the engine and run at a fast but constant idle. Note the vacuum reading. Watch it for several minutes. If the reading slowly drops, your exhaust is plugged up -- generally the cat. But again, this is not an intermittent problem. If it's plugged, it's plugged, period.
Same with the fuel filter -- it doesn't mysteriously unplug itself either. Fuel filters generally slowly start plugging up so that the engine won't develop full power at high throttle settings -- pulling up a hill for example -- and gradually get worse and worse. Electric fuel pump issues can be intermittent though, as can fuel pressure regulator issues. And contaminated fuel is always a very real possibility -- especially with today's ethanol blends (ethanol loves to attract water). Never fill your tank within an hour of a fuel delivery at the gas station -- it stirs up all the water and other crap that settles to the bottom of the underground tank.
If you can catch it on a day when it is acting up, it would be worthwhile to perform a fuel pressure test by connecting a gauge to the rail on the engine. Unfortunately, the RM uses hard plastic fuel lines so you can't test the regulator by pinching off a rubber hose. You have to install a temporary shut-off valve in-line. You can get an idea of what's going on by watching the short- and long-term fuel trim values with a
scan tool though. Ideally, the trim should be around 127 +/- 20. If you see trim values way off to one end or the other (eg near 0 or 256) that indicates the computer is trying desperately to compensate for something. Either an o2 sensor is way out of whack, or there is a real fuel delivery problem.
Given your symptoms, I'd be looking very closely at the ignition cap, wires,plugs etc. and all the associated connectors; the PCM connectors -- especially the ground; body and engine ground cables and the battery terminals for loose or corroded connections. While you are looking, check the
intake system to make sure it hasn't ingested a loose plastic bag or rag.
If that all checks out, then I'd look at the throttle position sensor (TPS) (if it's worn out, the computer doesn't know you have pressed the accelerator and won't bump up the injector pulse width). The resistance should change steadily and evenly as the sensor is moved through it's range. Any jumps or pauses indicates the sensor is faulty.
Similarly, the mass airflow sensor (MAS) sensor affects what the computer knows about how much air is entering the engine. The system uses a special heated wire that changes it's resistance as it is cooled by the airflow so you can't easily measure it's performance but it is very sensitive to poor connections (high resistance) and dirt or oil on the wire itself (insulates the wire from the airflow)
The knock sensor is another possibility -- if the PCM thinks the engine is knocking, it retards the timing. IIRC, by 94 GM had done away with the old ESC module that controlled the retard and the KS is connected directly to the PCM so I
think any problem with that should throw a code.
Evap cannister purge system. This is an electrically controlled vacuum leak. Electrical faults will be detected by the PCM and throw a code, but the valve could still physically stick and leak vacuum.
