I have tested two good TPS and four bad TPS. Heres the kicker. They all have different ranges. No two alike. Some are close while others are so far off. This is a serious problem. The signal voltage that is being sent to the ECU is the important figure here. Each TPS signal that is being sent causes a different reaction from the ECU. The bad TPS have been disassembled and the killer defect is on the negative side of the resistor strip. I'm sure this is related to the materials used in construction. The strip is a flexable circuit board material with the resistor being carbon coating over a split super thin copper circuit. I'm thinking the resistance is too much for the material. I have searched other TPS and found some good construction, but the range is different and can not be modified. Maybe I will start makeing these, who knows.

 

I just replaced my tps with the Airtex p#5S5313.
Made no difference with my erratic idle.

 

Not only have I been checking out the TPS but the throttle body as well. If no change with the TPS switch than you may have something going on with the throttle valve assembly. Still looking into problem areas. There are two seals on each side of the throttle valve shaft. perhaps the problem is there. My spare throttle body makes my engine run like crap. I have disassembled it many times and found that the throttle valve has a gap between the bore. Maybe the problem is there. It may even be a defect in the internal air paths. Any seal leaks will cause disruption.

 

Are you referring to the shaft in the throttle body the butterfly and linkage is attached to? I guess it could be worn out and have excessive play(creating a leak). I would never suspect it since I only have 87k miles on the thing. But stranger things have happened.

 

The seals on the TPS side and the linkage side are rubber. These engines get hot, hot, hot and the rubber dry-rots. I would check there for air leaks. Also I expect something strange inside the air chambers for the IACV. Have you sampled your voltage on your TPS? Use a multimeter. Positive to the solid red wire. Negative to the red wire with the black stripe. The voltage measured here will tell you what the resistor voltage is (record voltage). Then keep the negative multimeter lead on the red/black wire and place the positive multimeter lead on the yellow (signal) wire. This will tell you the voltage going to the ECU. You do not need the engine running. Just in the second key position. Sample and record the values at idle and at full throttle. I have an extra TPS connector that I use. But you can just take three spare wires about twelve inch's long and unplug the TPS and insert the wires in the terminals of the plug. Then reconnect to the TPS. Be sure the wires do not touch. You could have a hickup area somewhere. This test will let you know where. With the negative multimeter lead on the red/black wire and the positive multimeter lead on the yellow wire, read the multimeter voltage from idle (moving slowly) pull the linkage towards full throttle. The voltage should gradually climb. If it climbs and then drops in a certain spot and then climbs again, than it is your TPS thats bad. If it moves up like it should than it is not the TPS problem. After you do this I will tell you how to test your ECU circuit that sends and recieves the TPS voltage. You can isolate the problem with these checks. If the TPS and ECU check fine. Then I would look at the throttle body for defects. Another common defect in the TPS is the resistor strip negative terminal inside the TPS. This area disintigrates from the resistance. To check this spot you will need to have the positive lead on the red wire and the negative lead on the red/black wire. Move the throttle from idle to full throttle in a rapid back and forth motion. The voltage should always go back to the idle voltage. If you loose voltage at any time then it is the defect that is causing it.

 

I appreciate all the detailed knowledge regarding the TPS.

I would like to have some further details on the ECU. What is your source on these?

It sounds like we need some better quality components for those critical inputs.

 

You can find the information by performing tests and research. Simple really. How does it work, what values should it have? Things like that. Basic electronics. The ECU is a simple computer that acts as an input and output device. A signal out to a component results in a signal input to the ECU. You can tell by the PC board which components are effected by certain input/output voltage. Example. The TPS sends back certain voltage within a known range to the ECU. Paths are opened/closed based off of this voltage range. Which in turn add or subtract voltage to other components, which gives you things like more fuel, timeing speed for injectors. ETC. ETC. I took my old ECU which was good and first identified all the components and paths. Wires go into a circuit which is controlled by resistors, diodes, gates, chips ETC and then come right back out as a wire to specific components. I have found a few areas that I'm concerned with which seem to be factory defects. Also concerned with the way people are waterproofing the ECU. If you look at the design, it is DC voltage and requires grounding. Waterproofing them in the wrong way will cause bad grounding, which will cause component failure and erratic operability.

 

Could you please explain this procedure taken from your above post:

"After you do this I will tell you how to test your ECU circuit that sends and receives the TPS voltage."

Thanks.

 

You most likely won't hear from him, he's been off the site, sold his Disco, etc. If you want more in depth factory manuals on the GEMS ECU and the sensors, go to the link in my signature for D1 GEMS.

 

After reviewing all of LRD2&ME's tps postings, I failed to read exactly which generic replacement he decided upon and how the casing/blade issues were addressed. I did find this link from the rover sd1 site but am unclear as to how to DIY. Anyone? Thanks.