TPS reads 100% all the time.

[91GVR4]

Okey here is the story. I discoved last week that my O2 sensor voltage was jumping around alot and causeing the ECU to dump fuel. So I figure the O2 sensor was bad and picked up a new one.
Today I started to change it out when I noticed that the wires going into the plug were cracked and that all the wires around there were in bad shape. So I get out my sidecutters, strippers, cripper, ect. and spend a few hours patching up cracked wires on my thermostat sensors. While I was doing this I noticed some clicking from the relays under the hood. I thought that was strange because I didn't think any of the wires I was working on had power when the car was off.
Anyway I get the wires all fixed up and deside to start the car and see if my O2 voltage problem was just bad wires. Well the car won't start! It sputtered a few times, but mostly just cranked. So I look at my logger again and notice that the TPS is reading 100% and doesn't respond throttle inputs. Great! So I pull out the multimeter and test the TPS. Pin 1-4 4.5k ohms, pin 2-4 .8k ohms to 2.5k ohms as you open the trottle. Those values are good according to my manual. Now I am thinking I have more wiring problems. Next I go into the car and pull the 24 pin plug from the ECU. I check the resistance from the TPS and it is the same as when I check it at the sensor. Now I am guessing it is an ECU problem.
The only other thing I can think of is that the ECU reads voltage (not resistance) thru the TPS. So has anyone had a TPS that tested good, but when hooked up to voltage was bad?
Thanks for reading this and any sugestions you have are welcome.

 

You are correct. The TPS is a potentiometer which is fed from 5 volts from a regulator within the ECU, and taps off a proportion of that voltage to send back to the ECU, depending on the position of the lever. It is just like the volume knob on a radio which feeds a portion of the available signal on to the amplifier. The ECU does measure that voltage which is sent back from the TPS.
Test: set up your logger to show TPS voltage. You say it is 100%
Now unplug the TPS and see if it drops to zero. If it does, look at the GROUND connection for the TPS - if this comes off it will always send the whole 5 volts to the ECU.
If it stays at 100% you either have a damaged input on the ECU or another damaged wire in the loom which is shorting the 5 volts to the TPS input.
If you have a multimeter connect its positive lead to the TPS output and its negative wire to ground. See if the voltage on the meter varies from 0.5 volts at closed throttle to nearly 5 volts at WOT. If it does the TPS is OK and the ECU is damaged. If it doesn't and the ground to the TPS is good, the TPS itself is faulty or the wiring is shorting.
Hope this may help.
Edited one time to correct an error.

 

[91GVR4]

Thanks for the reply, I knew someone out there had the info I was looking for.
I already know that when I disconnect the TPS the logger shows 0%, so now I will check the output of the TPS and that should narrow it down. I just wasn't sure what the voltage was, but you helped with that. Thanks. Off I go.

 

[91GVR4]

Okey, I have figured out that the TPS has no ground, but the wires to the ECU are good because I can read the TPS resistance from the plug on the ECU. So, the TPS grounds thru the ECU, correct? Can I check the ECU to see if the ground path is bad? Are there resistors, caps, or diodes on the ECU that may have gone bad? Thanks for your help.

 

There is a track on the circuit board which blows if there's a short somewhere. (you may have accidentally shorted something while doing your other work - it only takes a momentary slip).
Open up the ECU and, using a multimeter or other continuity tester, check for continuity from pin 24 through to pin 101 or 106. If it is not continuous, solder a small wire between 24 and either one of those pins under the board in the ECU to repair that burnt track. That is all you should have to do. Use insulated wire so it can't short to anything else and take care while soldering, specially if soldering isn't something you're very familiar with. If that track is blown, according to my diagram you should also be seeing an incorrect coolant temperature reading.
Again, hope this helps you.
Tony

 

[91GVR4]

Thanks man. I figured it was the ground because I am seeing -75 deg. coolant temps. I noticed they both use the same ground. I will check the board today and probably be doing some soldering latter. Good thing I have some high quality, small gauge, wire laying around.

 

If your coolant sensor isn't showing proper temps your car is probably hard to start too.

-Josh

 

[91GVR4]

Yep I have had that problem before too. I conected the wires wrong once on the coolant sensor and got a CE light and bad running.
Well I check the ECU and sure enough the ground track from pins 24/17 to pin 101/106 is open. So I guess I will be doing some soldering at work tonight. We have some nice soldering stations there that are better then what I have at home.
Thanks for the the help with this. I hope to have her back on the road this week. Pending my original O2 voltage problems.

 

I just did a search and I have this exact same problem. I read this thread and I just figured out my ECU ground circuit is blown after checking for continuity, found none, alligator clipped from the 106 to 24 terminal on the board, came up with 0.5. Then tried plugging the ecu back into the car with the aligator clips still hooked up, TPS reads normal on my AFC, no more sputtering or black smoke, no more CEL. Ok so my question now is this, I dont really trust myself too much soldering the board, so is it possible to solder a jumper wire from the 24 to 106 terminal on the HARNESS and NOT the board? Or will this not work properly or cause some other type of problem? Just wondering if its necessary to mess with the actual circuit board and if the ground must jump through the ECU or not, thanks for any help.

 

[powerplay]

Here this might help you. Please invest in the FSM on CD so you know what you are trying to do.

 

[91GVR4]

I think as long as the TPS sees a ground it will be fine. I soldered a 26AWG wire between the pins on the ECU so my ground doesn't actually go thru the board. It all works good for me.
On a side note, I think I burned the ground out when I was working on the car with out disconnecting the battery. I had the coolant sensor wire off and I think it grounded out in the engine bay.

 

[stevep]

Fixing a blown ground trace on the ECU is trivial.

Steve