95 SL2 Idles bad, Drives Fine
 

Hi everyone, I have been working on diagnosing a car a got with no luck. The car idles very bad (low), knocks (and shakes violently) . But as soon as you give it gas or begin moving it runs fine. Here's what I've checked into or replaced.
Fuel pressure was tested and within spec.
MAP sensor was tested and within spec.
Idle air control valve was replaced (throttle body was cleaned)
Checked for vacuum leaks, there are none.
Gasket for intake manifold was replaced.
New Spark plugs and wires.
New PCV valve.
New EGR Valve.
It is also important to note that the car does not run with the egr valve plugged in, I have it unplugged because with it on the car sputters and dies immediately.
Do you guys think it could be a bad throttle body position sensor, a bad PCM, a defective torque converter, bad motor mounts, bad o2 sensor or I was also thinking it could be a broken valve spring perhaps. There was some scoring on the spark plugs but they had not been changed in many years before that.
I will take and I appreciate any advice. Thanks in advance everyone.

what codes did you pull (assuming SES light is on)?

I'm not sure just how violently the engine is shaking, but given the age of the car, a failed motor mount is a definite possibility. Next, I'd measure the action/operation of the throttle position sensor, to make sure it is in specification limits. You claim that the engine runs fine when the RPMs come up, so I wouldn't worry about a broken valve spring (for now). Hopefully, you used OEM style/type spark plugs, and not platinum plugs, because these little engines just don't run well on platinum ....

Did you replace the TOP mount with a NEW one from the DEALER? If you can put a number 2 pencil between the outer metal strap and the rubber, you need a top mount.

Thanks for the reply's everyone, there are no codes to pull i'm afraid, every time I shut the car off they reset. I will check out the operation of the throttle position sensor, as well as the motor mount, and get back to ya. Thanks guys.

is the ses light on at all?

I have found over the years of working on something that could be called a derelict car, that is one that is long of tooth, has been bought cheap due to previous owner not fixing problems due to lack money, interest or skill and the present owner not knowing how it is supposed to work jury rigs it into some kind of operation and then starts disconnecting smog equipment to make it run better. And then asking questions on how to make run even better yet by purchasing parts for it that one of the basic problems is the smog equipment has been mickey moused and messed with under the false impression it will now run better.
Ethe EGR valve is generally a prime offender in this senerio. People do not understand it or like it or know what it is supposed to do and many times have no clue how it supposed to be connected and don't know it is was connected right or not in the first place.
Sorry but this is what I have seen. No offense meant but part of your problem is signaled by the operation of the EGR valve if it was actually hooked up correctly in the first place.
What does the EGR valve do? It recycles exhaust gas back into the intake manifold under certain conditions. That has a specific function but that is not important, looked at it slightly different if out side gasses are recycled back into the intake manifold it is functionally acting like a vacuum leak. A controlled vacuum leak that recycle air back into the intake manifold.

Now how and when it does it is an interesting concept in that as the smog dogs created working algorithms as to when and how much this is supposed to take place and the computer gave more precise control as to when and how other parts of the mechanisms that run the engine compensated for this action to keep things running smoothly. Because what ever exhaust gasses due to the combustion chamber when they get there to control smog particles, it was still functionally acting like a vacuum leak. It is just that on a carbureted car it was controlled by ported vacuum and engine heat. These in many case were poor parameters but the only ones available and on many of these cars disconnecting things did indeed make things run better some time. This is also included into many carburetors were designed to control the engine under these conditions specifically and in many cases they were complex, wierd in operation and problem prone themselves and mechanics neither understood how they were supposed to work and as such generally managed to really screw up an engine performance by trying to adjust them or disconnect parts that were not understood. Ford had some Dandys but Ford was not unique in this.
On a computer controlled car that function in many cases can be controlled or altered by a computer generated signal under a wide and unpredictable set of standards and disconnecting it can cause the engine to quit.
Now I don't know how or why it works on a Saturn. I do know on it and my other computer controlled cars they run fine with it working and crappy with it not working. So i have quit disconnecting it.
So if you are disconnecting it because the car runs crappy and it runs better with it disconnected something else is wrong to and the EGR was either adding to or subcontracting from the problem.
As one of the functional equivalents of an EGR valve is to function as a controlled vacuum leak and by disconnecting it so there is no more control to it's vacuum leak function my guess is that you have a real vacuum leak someplace else. So until that is fixed or the reason found as to why disconnecting the EGR valve causes the engine to run better I would not worry about what else is wrong.
I would be worried about why disconnecting the EGR valve makes it run better.

WHY?
Because disconnecting an EGR valve on an engine with nothing wrong with it is not a good idea. My experience says on a computer controlled car disconnecting it makes it run worse. So disconnecting makes it part of a problem.
One of which in my opinion and in this case says this needs to be solved until the EGR and it's function is found out to be either o.k. and it is running as it is designed to do.