If that 16 year old “hunk of crap” ran decently until just recently, I submit that it isn’t really “crap”. And, I think that the Service Bulletin posted by weekend warrior is worth considering as a possible solution to the problem.
When you did the fluid change, did you just do a drain and fill or did you take it to a shop and have it put on a flush machine.
If you DIY, did you do a reverse slam. This is a common procedure for Saturns and all it is is setting the parking brake and put it in reverse for 10 min to half hour. A couple of those circuits gunk up from lack of use and the reverse slam cleans them out.
Is your throttle fly-by-wire? If cable operated, did you try to set the idle with the throttle stop screw? If so, that could be your problem. The throttle stop screw is not an idle adjustment, it is only there to keep the butterfly valve from jamming in the throttle bore.
Do you have a scan tool that will display both the accelerator pedal position and the throttle valve position? It seems like if you could watch those two numbers while you moved the throttle pedal, you’d get to the bottom of it pretty quickly. Position sensors often use a variable resistor to sense the angle of a shaft, both the shaft that rotates when you press on the gas pedal, and the shaft that rotates to move the throttle valve. With time & use, that sort of variable resistance sensor often tend a to develop dead spots, so at certain angles the resistance is out of whack. Have you ever had an older radio with a volume control knob that didn’t vary the volume smoothly as you turned the knob? Same thing. I expect you need to test all of those sensors to determine if their resistance varies smoothly and consistently as the shaft rotates.
On modern cars like yours, the replace this, replace that and see what happens usually isn’t very effective. Easy to run out of money before running out of ideas what to replace next. The scan tool method is better, b/c it not only tests the sensor, but the electronics that reads the sensor. Do you have access to the vehicle’s pertinent service data? Even an aftermarket manual like Haynes or Chiltons might have a diy’er test for those positions sensors. From what I can tell the sensors are powered with a 5 volt reference signal, and may not be bench-testable unless they are powered up, best done using a scan tool to read their outputs during a “sweep test”. Best bet probably is to either secure the needed scan tool, or hire a shop that has it already. Once you know what the problem is you can always replace the part(s) yourself if you want.
Note that the problem could be something entirely different, like the throttle valve is sticking inside the throttle body.
I would look hard into the Accelerator Pedal Position (APP) assembly (gas pedal), it has it’s own TPS and an off idle sensor IIRC, if it acts up, it can enter limp home mode… Also since the transmission no longer has a mechanical link for the PR valve, it gets it’s info from the throttle pedal (at least some of it) which can cause shifting issues, or possible limp home mode to keep from burning the transmission up…
The P2135 is a a/b voltage Correlation issue between the throttle body and the gas pedal…
I think the ob2 will arrive today before I try & order part #2. I’ll see if I can post the live data after.
But not 100% sure what to look for besides the numbers need to correlate. I assume if accelerator position is wack, it’ll be the APP replacement. If both are messed up and goes wonky I assume that’ll point to electrical problem.
And gonna assume tps/throttle itself is fine since that was replaced. Tho not impossible.
No experience with your car or its diagnostics but I’d guess the ATP (first in the list, which I’d guess is “A”) and its “B” version are two separate sensors measuring the same thing (the throttle butterfly valve inside the throttle body), so they should more or less be the same as you move the accel pedal through its range. They seem pretty close, but no idea if they are as close as they are spec’d for. It appears the accel pedal also has two sensors, D & E, which should probably match each other and the throttle positions through the entire range. E appears to be a problem, isn’t matching either D or the two throttle valve positions. It’s possible however that E is correct and the other 3 are the problem. An argument could be made for that theory b/c the commanded throttle position is close too E. Maybe somebody here with this sort of experience will chime in. In any event it appears you are getting close to a diagnosis.