My Citroen XM 3.0 PR-Vestige

[White Exec]

The TPS wiper must only be travelling across a limited part of the resistance track.
It obviously never gets to 'bottom', otherwise it would output 0v.
It doesn't get to 'top' either, when it would output +5v.

You say it needs 0.5v (and no more) to register idle.

Could you persuade the wiper to get into the very bottom part of the track?
If you can, then you'll have access to voltages very close to zero (and certainly between zero and 0.5v), and then be free to up the TPS supply voltage to achieve a WOT figure.
_______

There is another approach:

Raise the TPS supply voltage a bit, to give proper WOT functioning**.
This will zap your Idle figure, because it will have drifted too high.
You could then trim the TPS Idle figure down, by putting a suitable resistor between the TPS wiper (output) and Gnd.
A temporary pot would allow you to adjust.
By winding the pot down in value, the Idle voltage will fall.

** You might get away with clocking/rotating it, which will also skew the Idle voltage up, but then be able to trim it back down again with the temporary pot.

[Dean]

Chris is right, im going in circles, i need to think about this properly as swapping the wiring about isn't going to do anything.

I wired it up as the wiring diagram suggests, Throttle plate closed it shows 3.2v 'MID' and as you open the throttle the voltage decreases to 1.4v with an 'IDLE' status at WOT, this is wired exactly as the diagram says.
Behaviour and voltage is the same with all 3 TPS' and both V6 throttle bodies but if i plug in my 2.0i throttle body it works perfectly correctly

I can wire the TPS to show 4.5v at Idle and .5v at WOT, i get the full range of voltage just in reverse, the car runs great but cuts out at full throttle because of the trailing throttle fuel cutoff strategy.

plugged in the reference voltage to the TPS is 4.6v unplugged it is 5.3v


D

[sdelasal]

Some suggestions Dean...
1. "if i plug in my 2.0i throttle body it works perfectly correctly" - so that's a good start point. As i'm sure you know, the tps is simply a potentiometer with three connections for Voltage supply, voltage ground and signal.
2. So electrically disconnect the 2.0l sensor tps and measure the resistance between the supply & gnd pin, then measure resistance between signal pin to gnd pin; at both closed and then open throttle posiitons. That's three measurements. The last two should be some fraction of the first one, e.g. 25% and 75%.
3. Repeat the measurements for a v6 tps and compare. They can't be the same otherwise both would work. It'd be interesting to see what values you get and how they compare to the book values if available.
4. Other thoughts, is the tps connector on the 2.ol tps and the v6 tps the same? Identical pin configuration e.g. is pin 1 for ground on both? Any chance the factory has swopped gnd for supply between the two applciations, but used the same connector? Steve

[Dean]

I was running the 2.0i TPS on the V6 throttle body and only put the original V6 one back on earlier this week.

The part numbers differ slightly by one digit, resistance values are the same as is all sensor functions although i have not tested resistance change/degree of rotation, however this would not cause my problem as both throttle body's although opening in different directions still have a 90 degree travel and min/max resistances the same. The V6 throttle body uses a crank and rod arrangement to give slower throttle plate movement for a given input in its first 40 degrees or rotation and then a much faster throttle plate response around the mid range onwards. The sensors differ by colour and the white V6 part has an extra tang on the plug the black 2.0i part does not.

Sorry i was a little unclear with that post, the 2.0i throttle body opens in a clockwise direction, the V6 opens in an anti clockwise direction.
I can wire the V6 to work with the correct spec voltages but it is back to front (4.5v @ idle and .5v at WOT) if i reverse the wiring i cannot get the full 4.5v @ WOT and the ECU does not detect the correct idle state even though the signal voltage is correct at idle.

Every time i unplug the TPS with ignition on the idle/mid switching threshold voltage moves up and so far i have not found a way (other than accidental and unrepeatable) to move it back down.

The TPS wired as the wiring diagram suggests is working backwards and with an idle voltage of 3.2v 'MID' and decreases to 1.4v at WOT where it switches to 'IDLE'. The only shared wire with other sensors is the shielded ground which is looped in with Pin A on the MAP sensor which is the MAP sensor ground, unplugging the MAP sensor makes no difference to TPS function, the 5v supply and signal are wired directly to the corresponding pins on the ECU.

Could i have 2 duff ECU's, 2 incorrectly setup throttle bodies and 3 defective TPS'? that all read the same resistances.........highly unlikely, the wiring checks out with no continuity to any other places other than the two ECU pins, i know from others info the ECU should see .5v at idle up to 4.5v at WOT as Cit documentation suggests and that the WOT 'FULL' status isn't as important to running as the Idle position is due to the fuel cutoff playing havoc with driveability if the throttle plate is not closed when the ECU detects an 'IDLE' state.

I need to look for other information regarding how the ECU determines these voltages into idle, mid and full throttle states, does it work with the MAP sensor and crank sensor? or is it all down to the TPS, i know that the TPS has a massive effect on idle speed and engine performance, i can vary idle speed by 600rpm or get the car to start instantly or need a few seconds cranking just by clocking the TPS by a few degrees out of spec.

This is very frustrating because i get the feeling the answer to this issue is simple and is staring me in the face, i am tempted to dissect a TPS to see what it actually is, is it simply an old school wiper track variable resistor and nothing more in which case the ECU gets it's throttle state info from a lookup table of voltages (so why do those thresholds move?) or another sensor is used to confirm these (like the MAP).

D

[White Exec]

The two TPS's evidently differ in their direction of movement/rotation for the 2.0 and V6.
And you say they have the same end-to-end resistance value.
But, are both resistance tracks linear, or something else? Could one/both be non-linear? And if operated "backwards" might that be a problem?
Just wondering.

[JD01JD]

A few months ago I adjusted my TPS as at idle it was around 0.7v when closed. Lexia still reported it as "idle" though.

Can't remember the process 100% now but to adjust it I first adjusted the sensor to read 0.5v closed throttle, then to get the ECU to recognised the new voltage as idle i think I had to switch the ignition off, unplug the TPS, switch the ignition on then reconnect the sensor (sorry can't remember the exact process but I think that was it)

If I didn't do that process the ECU would remember the 0.7v as idle - it would stay at "idle" until the throttle was opened enough to reach above 0.7v. Now it switches to "mid" slightly above 0.5v.

The ECU is able to adjust itself to the idle voltage - it will take the voltage it has learnt for "idle" and switch to "mid" when it sees a voltage above that (with a 0.1v or so margin of error). How it determines when to switch to "full" I have no idea. There are a few possibilities - using a voltage above the set idle voltage, using a fixed voltage independent of the set idle voltage or somehow learning it from the full voltage sweep like a modern car would.

[White Exec]

Maybe it learns a voltage when first connected to its sensor, and then always works with that. Only re-learns if it has to, eg by finding no sensor connected, or voltages completely out-of-range. Who knows? This is what documentation is for!!!

[Dean]

Chris the two sensors work the same way but the different engines throttle plates rotate opposite directions, the TPS needs to be wired differently to read correctly operating in a different direction, following the V6 wiring diagram would mean it is wired for a throttle plate that opens clockwise, it does not though, the V6 plate opens anti clockwise when viewed from the throttle cable side of the TB.
I have found no documentation on the web or in the items in my possession on the subject other than that posted by John on the operational voltages and recall mod to the sensor mounting holes.

JD01JD that's helpful, im going to try that once i've finished sorting out the garden and see if i can get that 'IDLE' / 'MID' threshold to set correctly.

Many Thanks

[JD01JD]

Having a look at Citroen service website three different TPSs fitted to the 12v V6s.

TPS.jpg

Citroen's parts lists can be a bit confusing sometimes but it appears that replacements for the early cars (until RPO 5124) item "3" is listed which consists of item "2" and "A" which looks like a redesigned connector to be spliced into the loom.

I assume the later cars have the different connector from factory. Cars from RPO 6048 appear to use a completely different sensor (P/N 16281E). This has the connector directly on the TPS and appears to be a Bosch part shared by many PSA cars of the era.

This site shows the two different sensors on 12 valve throttle bodies.

No idea about the differences between the three but I assume they are different? Something to think about anyway.

What's confusing is the correct voltages are able to be obtained but in reverse - in theory if you switch the ground and supply to the TPS (so they are on opposite sides of the track) the signal voltage should reverse.

[Dean]

I think that loom section is to swap the pinouts and terminate in a different plug, i suspect this was done to be able to fit the Bosch TPS (of which all of mine are).
I have put the wiring back to how i did it which gives .5v up to 4.1v now with a little throttle stop tweaking, the ecu shows all three states of idle, mid and full throttle.
I managed to bring the idle/mid switching threshold down from .89v to .68v by disconnecting the TPS, then the battery for a few minutes, then plugging the TPS back in before reconnecting the battery, i then clocked the TPS up to an idle voltage of .60v which is still within spec.

It drives nicely now with no jerking and a good steady idle at the right speed but there is still a hint of mid range stutter but over a narrower rev range than before.

I will need to dig and explore these TPS changes, i am confident it is just the TPS itself and what the ECU expects to see which is the issue, i knew about the early TPS which was huge and mounted in its own frame but i didn't know about the 3rd type with a throttle switch, i need to confirm how old my ECU is as the case would suggest it was built after july 94.

I do now know that to use the early manual ECU, i have to re-wire the TPS so both of the two ECU's i have could be incompatible with my TPS setup.

D