Citroen XM Forum

I need to check the start up sequence on the 2.5 with its engine management light, perhaps is it behaves in a similar manner. The 2.1 having a manual pump is devoid of these things.

The other idea I had, was wiring up the duff sensors to the working sensors

Now the front sensors are working. If I wire up the cables that go from the ECU plug to the rear sensors, to the wiring for the working front sensors, will my ECU think that all sensors are working and test as normal, and report that all is working Ok?

It is cheating, and something I would normally not do, but the brakes are working perfectly on the car without the ABS, and I know to drive accordingly with no ABS.

Hi Robert

Might be that you could feed both rear inputs from the one good sensor as a temporary fix.
I seem to recall that the ECU combines the rear signals anyway.
That might be easiest done at the ECU plug because at that point the wires are easily identified.
The alternative if you have the loom from that scrapper is to make up a Y lead from two 2 pin plugs and a socket.
That could be plugged in on the back subframe.

John

Measure the resistance of a good sensor, and replace the knackered sensors with resistors of that value

Hi Johnathan

Is that just a possible option or something tried and works.
My interest is that it could be ideal made up as a temporary plug in lead to aid fault finding.

John

I have known it to work on a BX, at least allowing the ABS to pass it's 'self test' for the MOT.

Hi Jonathan

That is useful to know.
I realise that it will not pass the ABS normal running test routine.
To be able to use the ABS start up routine for fault loction is something I had not thought of.
I will sort out the bits to make up a test lead and check it out on both Bendix and Teves.

John

Interesting, interesting, some very good food for thought there.

I will get cracking tomorrow and try and find out if wiring the 2 failed rear sensor supplys at the ECU plug to the working fronts supply works. I might also see if I can track down some resistors as well, the good ones both measured 2550 Ohms or there abouts.

I like your idea John, about creating a test lead.

One question about the self testing. We know that the ECU does a start up check, and then another check when the car is running. What are the parameters, or when does the second round of testing start. What triggers it? My concern being that I don't want it to start doing its second testing when the car is being driven in the test centre between the different check stations.

Robert

robert_e_smart wrote:
One question about the self testing. We know that the ECU does a start up check, and then another check when the car is running. What are the parameters, or when does the second round of testing start. What triggers it? My concern being that I don't want it to start doing its second testing when the car is being driven in the test centre between the different check stations.

Robert

I would expect that once the ECU sees vehicle motion over 5mph it then looks for varying inputs from the ABS sensors and throws an error is there are miscounts of pulses.

IMO, you are going to have to fix the system properly.

Hi Robert

The ABS start up checks are done every time the ignition is turned on and the engine is started.
This is a done once static check that checks that all components are present and respond.
As Johnathan has pointed out the static check on the sensors is simply measuring the resistance of the sensor coil.
The other system components also get static checks.
Both main and slave microcontrollers run these programmes and also compute the same test programme storing the results.
The main microcontroller then checks that both sets of results agree.
The flashing light sequence is the physical indication of the various stages of this process.

Dummy resistances it seems may well get you through this static check.

Documentation on the moving tests is far harder to come by. I expect it to vary with each ABS type.
We know that the ABS can lock out due to seeing system malfunctions while the car is on the move.
Most people assume that the lock out comes from sensor missreadings during braking.
In practice this is not the ideal time to self check for missreading as it is the only time the system is looking to operate from reading differences.
So why not wait until there is no brake application, the ABS micros are then free of work, so set them to count ABS pulses, then compare the results.
It will take less time than even the start up routine so why not do it regularly. Say every 10 or maybe 20 seconds would seem reasonable.

In mainland UK MOT the examiner is only required to check the ABS light to ensure that the startup routine completes correctly.
Passing this test then getting an ABS light while moving the car to complete the test might present the examiner with a problem if he notices,
particularly if he has already marked the ABS box as a pass.

Despite the above, I am with Dieselman, fix it properly, you never know when you are going to need the ABS.

John

Ive an opposite problem on my V6 12v auto

Self test - then the light illuminates... however the ABS STILL WORKS!! REALLY It does Ive had this argument on the old fourm..

Anyway, what Im getting to is, driving along, if I flick the ket to start (Which if course in an auto does NOT activate the starter) it resets the elctronics, and sends the ABS into self test again, and as long as the car is moving the ABS light now DOES go out!!!

Then when I come to a junction, and just about to stop, the light will come back on!!

Paul