Hi all,

I have a Citroen Xantia 2.0HDi which drains the battery over 2 days. I have tracked down the culprit. Even when the ignition is off and the key withdrawn, the ECU and injection relay are both still live. I have checked the ignition switch and the line leading from the switch to the ecu (arriving at the ecu on terminal 69) is dead. The ECU is a Citroen Part number 9636321680 which I think is a Bosch 0218010749 EDC15.

The injection relay has a permanent live connection through F4 (in the engine compartment fuse box). The coil of the injection relay is therefore switched on the negative side by the ecu.

Which terminal on the ecu determines whether the ignition is switched on or not? I can see a switched live arriving on terminal 69. Are there any others?

The question is,
Is the ECU faulty and keeping the Injection Relay energised when it shouldn't? In which case I need a new ecu.
Is there a fault elsewhere which makes the ecu think that the ignition switch is on? In which case, which terminal does it use to sense that?

I plugged in the Lexia and went to the actuator test page. I can successfully switch the Injection Relay on and off with the ecu so there's nothing wrong with that function. I also disconnected the ecu with the ignition switched off and the injection relay deactivated. I can therefore conclude that there is not a short in the line leading from the Injection Relay to ecu.

Does anyone have a spare ecu that I can use to swap out to determine if the fault is with the ecu or the remainder of the wiring?

Cheers
noz

Hi Noz

You have an Hdi we all succomb eventually The only experience I have had with this problem is on one of the C5s, that was due to the fitting of an aftermarket radio which was sending spurious signals to the ECU and switching it on, as yours presumably doesn't have multiplex this won't be the cause of your problem although the fact that spurious signals can switch things on could be a possibility. I hope that someone who knows the problem can help.

Peter

Thanks Peter,

When I say "I have" a Xantia, what I really mean is that "I have someone else's" Xantia which I'm trying to repair. It's not mine.

I don't know if you downloaded the attatched diagram but I would be obliged if you would and have a look at the circuit going from the battery, through fuse A, through the iginition switch, through BF00 and into terminal 69 on the ecu. This is the only way I can see that the ecu can tell the difference between the ignition switch being off or on. Can you see any other way?

LEGEND
1320 - Injection ECU
1304 - Injection Relay
BF01 - Under bonnet Fuse Box
BF00 - Passenger Comparment Fuse Box
BMF1 - Underbonnet Supplementary Fuse Box
1150 - Glow Plug Relay
CA00 - Ignition Switch
C001 - Diagnostic Connector
1220 - Engine Coolant Temperature Sensor
0004 - Instrument Panel


Cheers

noz

Hi Noz

I'm busy swapping some seats over at the moment, I will have a look later in the day.

Peter

Right I'm back. The only other connection is to '1150' don't know what it is but looks quite heavy. Is it possible to disconnect the feed to the Ecu and see if the voltage is still present, that would eliminate it, or whether its actually being fed in. I know its not easy on a car but what I used to do with TVs was disconnect each circiut one at a time until I found where the voltage was 'leaking' in. Sorry I can't be of more assistance

Peter

Hi Noz..

Unfortunally a simple ECU swap on a HDi is not a straight swap...

You can only fit a NEW HDi ecu alone.

If you want to try another ECU you also need to swap the Matching Body Computer which is located right up behind the Glove box and Passenger air bag...

This is because they are matched for the Transponder unlocking etc.

If you fit an ECU and body computer you then only need to Lexia them to learn the Transponder Pickup and match the Key to them...

I have got a Match ECU and Body computer together from a T reg HDi 110...

Ive looked closly into this HDI ECU swapping because Im wanting a Later X or Y reg ECU for my V reg, as the later ECU's have got Cruise Control built into them.

Let me know if you still want..

Paul

Hi Paul,

That's not good news. I had diagnosed a faulty ecu and Allan has gone and bought another one. He was coming to my house on Sunday to have it fitted. I'll need to put him off until I can get a better handle on this.

To be honest, although all the tests and results lead to a faulty ecu I was never convinced since the ecu is fairly robust and from my (albeit limited) experience they rarely go wrong. That's not to say that they never can go faulty. I was really hoping someone would spot another way in which the ecu is fooled into thinking that the ignition switch was switched on.

I was really confident I was going to be able to solve this but now I'm not so sure. I have no working experience of this particular ecu so I'm in the dark. Thanks for the heads up on the ecu swapping since I would have been stumped if I'd tried it.

Back to the drawing board.

Cheers

noz

Oh, I suppose I should really ask, are there any members with a matched ecu and body computer (CPH or Protection Control unit) which they could lend me in order to prove whether the ecu is faulty or not?

It is my understanding that one way in which modern diesel cars start so easily is that the glow plugs are energised when you unlock the car with the remote. Therefore, by the time you get the key in the ignition the glow plugs have been on for 10s of seconds so when the ignition key is turned the engine starts immediately. In order for that to happen the controller for the door locking has to communicate with the circuit that operates the glow plug relay and the glow plug relay needs a supply that is not routed through the ignition switch. Now there's none of that circuitry evident on the attached diagram since the glow plug relay (1150) is supplied via Fuse Box BF00 and the ignition switch. So there's no way for the body computer (CPH) to bring on the glow plugs. However, does anyone know of any other similar functions whereby the CPH would cause the engine ecu and associated double relay to be energised?

Yes noz.. see my post above ive got a matchng pair I offered to you..

However after install you will still need a lexia to match the key transponder up to the replaced ecu...

Has your man bought a brand new ecu? Or a second hand one?

Peter,

Thanks for the advice. All greatfully received. Item 1150 is the glow plug relay. It gets its main power through the ignition switch and fuse box BF00 arriving at terminal 3. The same circuit arrives at terminal 69 on the ECU which I presume is the way the ecu knows the ignition switch is on. I have disconnected the glow plug relay and tested terminal 3 but it was dead. I therefore assume that terminal 69 on the ECU is also dead. Therefore, by deduction, the ecu is not being signalled (at least by the normal route) that the ignition switch is on when it it not. If the ecu is not faulty and it does indeed think that the ignition is actually on then it must be receiving a signal on another terminal to make it think that this is so. However, on the diagram attached, all other circuits leaving the ignition switch which have nothing to do with the engine have been omitted for clarity. Therefore, the only circuits enabled by the ignition switch are the line to the glow plug relay and ecu, Fuse F2 and Fuse F12. These were also dead when I tested them with the ignition off.

I'm thoroughly stumped.

cheers

noz

Another thing to listen for noz.. after an engne switch off and key removal after about 20 seconds or so there should be a click and pressure release noise as the system shuts down.... do you hear this.. theres nothing to do.. just open the bonnet stop the engine then listen under the bonnet...

Hi Paul,

Sorry, I wasn't paying attention. It might be useful to try the matched pair even if only for elimination purposes. I do have a Lexia although I've never done the matching before. I presume its all menu driven and reasonably straightforward? I don't know what Allan has bought so I'll need to contact him to find out.

Regardless of what you do or how long the ignition has been off, the ECU relay remains permanently energised. There's also a slight high pitched hum which I assume is from the stop solenoid on the HDi rack. The fact that the injection relay is permanently energised results in a constant draw of 0.75A. This is enough to drain the battery over 2 days.

To let you understand, I obtained a copy of Citroen Sedre and printed off all drawings with Fuse 25 on them. The reason for this was because Allan told me that he pulls Fuse 25 when the car is parked to stop the battery drain. I then disconnected every circuit coming from Fuse F25 one at a time whilst watching an inine ammeter. None of them resulted in the elimination of the 0.7A current draw. I then started looking around the engine bay and heard the high pitched hum above. I relaised it was the stop solenoid and figured the ECU must believe the ignition was on. I dismantled the ignition switch and tested its operation to make sure it wasn't stcuk on. No suck luck. The circuits leading from the ignition switch were dead if no key was inserted. So I drew a blank. After a bit of searching I found that Fuse F4 fe the injection relay. I pulled the fuse and the 0.7A current draw ceased. So did the stop solenoid and the injection relay dropped out. In other words, the circuit coming from Fuse F4 was the one causing the current drain.

However, the injection relay is switched by the ecu on the negative side so it is permanently powered on the live side directly from the battery va Fuse F4. In order for it to be energised the ecu must be earthing the return circuit from the coil in the injection relay. The question is, how can it do this if the ignition is off and terminal 69 is dead? I assume that the circuit inside the ecu for the injection relay circuit is earthed with via an NPN transistor. It could be that this transistor is burned out on the PCB of the ecu. Not being my own car, I'm reluctant to open up someone else's ecu. I risk bricking the ecu and rendering the car immobile.

I take it that if I replace the ecu and CPH and reprogramme then I can always go back to the original ecu and CPH at any time simply by reprogramming again?

Thanks for all your help.

Cheers

noz

Hi Noz

It might be worth checking the Injection relay.
Two thoughts:
The critical relay contacts might have welded themselves closed.
There are lots of different variants of the injection relay.
There is usually a part number and circuit diagram on the top of the case.
Only found this out when I tried some recovered ones in a 2.5 I could not start.
Fitting the wrong type bypassed the locked out keypad and got it running.

John

John,

Thats a fair point. Allan has replaced the injection relay with a new one as part of the process of tracing this fault. I have assumed until now that the relay he bought was the correct one. I'll need to go back and ask him where he got it although I think I remember him saying he got it form Citroen.

In terms of stuck contacts, I am able to use the functions in the Lexia to test all actuators including the Injection Relay. When I activate the test the Injection Relay clicks on and off at about 1s intervals for about 10s. Therefore the ecu is able to switch that relay on and off. This suggests that there are no shorts in the circuitry and that the circuit is working as it should. Faults in this circuit would either prevent the relay from energising or keeping the relay energised all the time. The fact that it turns on and off proves that the circuit works as it should. Therefore it suggests that if the ECU is keeping the injection relay energised then it must be because it thinks that the ignition is on. To me, terminal 69 on the ecu is the terminal which becomes live once the ignition key is turned. When this terminal goes live then the ecu goes from slumber mode to active mode thus energising the Injection Relay. You can't easily get into the multiplug on the ecu but according to the wiring diagram terminal 3 on the Glow Plug Relay is wired in parallel with terminal 69 on the ecu. With the ignition off I have tested terminal 3 on the Glow Plug Relay socket and it is dead.

Therefore, either there is another way in which the ecu thinks that the ignition is on other than the way I've described or there is a fault in the ecu which is keeping it energised when it shouldn't be.

I have spoken to Allan and advised him that he needs to get the CPH from the same car as the ECU he has bought. The ecu came from a scrap car in a breakers inthe south of England somewhere. He is waiting on a response whether the scrappy can find and extract the CPH and send it to him.

If it had been my own car I would have taken the ecu to bits by now to try to identify the transistor which takes its feed from terminal 69 and check it for faults. However, when its not your own car its a bit more difficult.

Thanks for all the suggestions.

cheers

noz

Hi Noz,

Not sure if this is at all relevant, however, my 2005 C4 (100% CAN) has the 2.0 HDI engine and I have gleaned quite a lot of Citroen info on this engine (via Docbackup) and it's control (only a small proportion of which I fully understand!!)

I'm not sure if the earlier Xantia variant of this engine is directly comparable.

Anyway as I understand it there is a unit called the Built-in Systems Interface (BSI) which quote 'is the heart of the mutiplexed architecture system' unquote. The BSI has four operating modes:

Inactive mode (all BSI outputs on standby)
Standby mode (corresponds to the absence of the ignition plus signal (relay switched by the engine relay unit) and the accessories positive signal
Reawakened mode (all of the functions are active)
Reawakening mode (software initiation)

One of the BSI's many functions is the 'distribution and protection of the supplies to components connected to the BSI' (which probably includes the engine ECU).

The point I am getting to is that the BSI is a programmable unit and corruption of the software is a possibility. My understanding is that a dealer would download the appropriate vehicle software (from Citroen) and then re-programme the BSI using Lexia.

Hope this helps!

Ken