Anthracite XM 2.0 SEi Turbo Manual

[White Exec]

Just in case it's any help, I measured some of the pedal sensor voltages too (mine's H2, of course).

Your meter resistance reading of the pedal pot track is about right at 4k5 (within the rough manufacturing tolerances for 4k7 for this sort of device). The safety resistance is actually stated at 2k05 (a very odd value), to 1k7 isn't far off.

My ECU was supplying 4.43v to the pedal pot. With the pedal up, the wiper of the pot is at the ground end of the track (via the safety resistor), so effectively grounded via 2k05. I measured the voltage being fed back to the ECU (pedal up) as actually 0.225v - very nearly zero. Voltage from pot with pedal down was 2.757v. (Voltages read with sensor still connected to the ECU.)

In practice, these top-end and bottom-end (pedal down and pedal up respectively) voltages aren't critical at all. All the ECU is doing is looking for a CHANGE in voltage, and more to the point, actually a RATE OF CHANGE of voltage. Rapid voltage changes mean that the pedal is being used with gusto, i.e. high acceleration, and therefore the suspension needs to go firm, in order to prevent back-end squat. (Similarly, the brake-line pressure sensor/switch is looking for hard braking - which will also move things to firm, to prevent nose dive.)

As I understand it, although several sensors are monitoring the driver/vehicle for non-gentle behavior (acc.pedal, steering, body movement, brake pressure, vehicle speed...) simultaneously, and any one of these can prompt a move to firm (maybe modified by whatever vehicle speed is being detected), the sensors do operate independently. So, I assume, if (for example) the accelerator pedal sensor is bypassed (detached in your case; switchable-out in mine via a grounded 2k2 resistor [=pedal up]) in such a way as NOT to create an electrical problem, then the rest of the system should operate normally.

Your speed sensor problem obviously needs nailing. It's possible that the output from it, via black box, arriving at the ECU isn't clean or tidy (bad connection or whatever), and might need looking at with an oscilloscope. I presume these early sensors can fail/or be marginal: our S2 is on its third one (later type, of course), the first two having given up the ghost when the car was still under warranty! This wiped out not only the speedo reading, but also cruise control and trip computer, all dependent on its output.

Although you could expect the car to behave predictably with the acc.pedal sensor de-mounted (driver's foot might be off the pedal at almost any time), I guess the same couldn't be said for the speed sensor. I'm only guessing, but if the ECU sees, for example, body movement/steering movement at zero speed, it might well go "safe" or get confused!

Keep at it, and keep posting how things go.

[Aerodynamica]

Thanks Chris I think you're right about the speed sensor.

The pdf. Talks about the micro voltage produced by Hall effect from the speed sensor. I think this is separate from the supplied voltage to the sensor itself ( something like 8v) I wonder if I can test that off the car.

Anyway the car is out and is now in use. CX in Storage facility. Still have the problem with throttle response as the engine heats up in fact it cut out at a roundabout approach ( did this once before too) but it seems to clear and rubs well after say 10 miles.

Suspension is stuck in soft mode due to disconnected pedal sensor and faulty speed sensor system. Car thinks it is stationary.

The ride is pleasant with the rear trending to too soft but front is perfect - does dive under braking but not too excessive. The rear squat under acceleration in bad though. Body roll in minimal (compared with a CX that is!) so I can live with it for a while...

[Aerodynamica]

New weird behavior!

I keep hearing a rapid cycling of a relay in the dashboard - sounds exactly like the indicator but it is random and worrying...

[xmexclusive]

With the chatter present check if nudging the indicator switch alters the noise.
If so change or repair the indicator switch.
Voltage is tracking across the grease which has collected fine metal swarf from years of contact wear.
That makes the relay "chatter" but does not operate the indicators.
Leave it long enough and the conductive grease will eventually heat up and burn off.
Sadly this melts the indicator switch and fills the car with acrid smoke.

John

[Aerodynamica]

Ah, you're right - it is the indicator stalk. I guess it can be cleaned?

[MTXM]

I keep hearing a rapid cycling of a relay in the dashboard - sounds exactly like the indicator but it is random and worrying...

I have this on one of my XM at present and assumed it was indicator relay related. The car has been laid up for a year so that might be a factor. As the clicking comes and goes I think I will just live with it for the present! Regards, Matthew T.

[Aerodynamica]

I'd replace it Matthew, if it can cause it to melt like John says..

Mine seems to have stopped for now but I'm not waiting around - I'll see about a replacement ASAP

[White Exec]

The switch is still available from Citroen, pt.no. 6253 55 (light switch - complete stalk/switch).
Some on eBay, IIRC, some of which are Chinese copies.

It is possible to strip the switch, but the mechanical wear on the moving and fixed contacts will probably prevent anything other than a temporary repair. The contacts are not electrically stressed - they just wear each other out by the sliding action. Ripe for redesign, if anyone could be bothered!

[captainhaddock]

I took both of mine apart and it is very easy to do. Cleaned out, filed a bit aand re greased and had no more problems. Easily fixed. Wouldn't buy a new one just for that.

[Dieselman]

I have done the strip and clean procedure on a FIAT once, to good effect.