C3 Earthing problem runs with obd2 scanner attached

[ozvtr]

Impedance looks Ok but the voltages are all wrong. Do you have good contact and ground on your multimeter?

If you close the doors and let the car go into sleep mode you should get about 2.5 Volts on each line. A bit more on pin 6, a bit less on pin 14 (which you are not getting).

Some of your reported readings are erroneous. If the resistance from one point to another point is 0 Ohms you cant develop a voltage across it. Do you mean open circuit (meter shows OL or no reading) on the ohms readings?

If you can remove the OBDII port from the mount, you might be able to back probe the connector while you have the scanner plugged in?

With the engine running and the car doing "stuff" you should get about 3V on pin 6 and about 2V on pin 14.

Looks like a BUS problem but what? Both voltages being dragged down? How does adding the scanner bring the voltages up?

[Meannamate]

Ok. I thought those readings were dodgy. Sorry where I quoted ‘0 Ohms’ I meant OL. I’ve taken the glove box out so that I can remove the OBD connector from its mount today. I’ll post a full list of voltages/resistance with the scanner connected/disconnected as soon as I can get some daylight. Thanks again for all your help!

[Meannamate]

Update - ok, it’s taken some time to pull the glove box and undo the OBD port so I can spike it from the back. I made sure got a solid earth on the multimeter and measured both voltage and resistance on every pin. So, out of the 16 pins 2,9,10,11,13 and 15 are blank (not wired). With pin 1 bottom right/pin 16 top left (using pin out diagrams from tinterweb) pin 16 always reads 12.5 v. Most pins read 0 v and OL. I’m just listing the ones with readings using the various scenarios below. Doors closed and everything switched off.
1. Key out, scanner disconnected;
4 and 5 - 0v, 41 ohms
14 - 0v, 131 ohms

2. Key out, scanner connected;
1 - 0.01v, OL
4,5,6 - 0v, 0.7 ohms
7 - 0v, 134 ohms
8 - 0v, 0.4 ohms
12- 0v, OL
14- 0v, 92 ohms

3. Key in, ign on, scanner disconnected
1- 12v, OL
3- 2.4v, OL
4- 0v, 0.9 ohms
5- 0v, 1.3 ohms
6- 0.5v, OL
7- 0.34v, OL
8- 2.4v, OL
12- 10.78v, OL
14- 0.25v, OL

4. Key in, ign on, scanner connected
1- 12.3v, OL
3- 0v, OL
4- 2.4v, OL
5- 0v, 1.4 ohms
6- 0.4v, 0.9 ohms
7- 11.84v, OL
8- 2.4v, OL
12- 10.75v, OL
14-0.26v, OL

So, certainly not getting the expected 2.5v at 6 and 14

Sorry it took a while to post. Hope you can help?

[ozvtr]

The voltages don't really change with the scanner attached. You would probably need an oscilloscope to spot the difference.

The only thing I can suggest now is to remove all the ECUs from the system and see if the BUS is being loaded down. You'll need to disconnect the scanner and check pins 6 and 14 for the following tests.

This is a list of ECUs (that I can think of). Check if you do or don't have them.

ABS (ESP), air bags, aircon, center display, COM2000, engine ECU, BSM, BSI, power steering, radio, reverse sensors, alarm.

Disconnecting the ABS ECU will also disconnect the engine ECU as the engine ECU passes through the ABS module. Remember the "ends" of the BUS are the BSI and the engine ECU (by Wheelers reckoning). But I would inspect the ABS connector as they have proved to be problematic in the past. Salty water can get into the connector and damage the pins and electrical continuity.

I would disconnect all of the ECUs from the BUS. Inspect the sockets and receptacles for corrosion. If they look OK, connect the BSI and check voltages of OBDII pins 6 and 14. They might not be 2.5 Volts but they should more than the 0.25V that you are getting. Add the ABS ECU and check OBDII voltages. Add engine ECU. You should get the 2.5V on 6 and 14 now. If they are OK keep adding ECUs until the voltages changes.

I am assuming one of the ECUs is causing the BUS error. So unplugging it should bring the BUS back up, but we have to find it. It's not an impedance problem, that is there is no short in the system and the pins appear isolated from ground. The BUS is not shorted because you get the requisite 120 and 60 Ohms when the BUS is static. It looks like one of the ECUs is "talking out of turn" and upsetting the BUS. If I'm wrong...well I don't know where to go from there.
Sorry but it's all that I've got.

[Meannamate]

Ozvtr- thanks again for all your help. I’ll start unplugging and report back. I’m starting to think it’s an internal fault in a module as every connector so far has been clean and dry.

[Meannamate]

Finally got around to unplugging the ECUs as ozvtr suggested. Weather has been a big problem for a few weeks. Today is the first day I can spend time. So all ECUs unplugged except the radio and AC. I would have to pull the radio out and I couldn’t get the AC plug out. So all unplugged except the above two and the ABS plug. Still getting 0.1 volts at pins 6 and 14. Unplugged the ABS and voila! 2.97 at pin 14 and 2.0 at pin 6! Everything works with the scanner connected. With the scanner out it won’t start and flags ‘braking system fault’ on dash. Based on this do I commit to a replacement ABS ECU? Thanks again for your help!!!

[wheeler]

You could try bridging out the CAN lines on the ABS connector which will effectively bypass the ABS unit, If the ABS unit is at fault this should allow the engine to start normally.
Would need your VIN or Reg to get a wiring diagram for the ABS system to get the pin numbers to do this.

[Meannamate]

Good idea wheeler! Thanks! Should I pm the vin and reg to you?

[wheeler]

Whatever you prefer, the VIN gets automatically hidden from public view anyway.

[ozvtr]

Unplugged the ABS and voila! 2.97 at pin 14 and 2.0 at pin 6!

Good to hear, nice work!

Everything works with the scanner connected. With the scanner out it won’t start and flags ‘braking system fault’ on dash.

You could try bridging out the CAN lines on the ABS connector which will effectively bypass the ABS unit

With the ABS disconnected but the scanner plugged in, the engine runs (and starts)? Is that what you are saying? Confused. To my knowledge the CAN BUS should pass through the ABS ECU to the engine ECU. Meaning if the ABS ECU is disconnected, the CAN BUS is disconnected from the engine ECU too! As I have quote Wheeler saying. We might be wrong!
Hmmm very interested to see what the circuit diagram looks like. CAN BUS might be paralleled off the ABS and not passed through it!? Could that be the difference between the different ABS ECU BLOCK versions? So many questions!!

Rhetorical Question:
What is being transmitted over the CAN line that the engine ECU needs to run the engine? I know that at start up the BSI passes on the engine ECU security code and transponder ID over the CAN BUS to the engine ECU. And that the engine will not start without them. But once the engine is running, I know that the engine ECU does NOT require this information. Well...I know it doesn't for my TU3JP engine!! So why is the engine shutting down when the CAN BUS gets stuffed up, while the engine is running? It just fouls up the ECU and that's it? I would have thought the CAN BUS handler and the engine management would be two isolated components of the engine ECU? So I can understand the engine not starting but I can't understand how it stops running. Yes, there is a lot of information going back and forth from the engine ECU but I would have thought loss of the CAN BUS, AT WORSE, would only cause the engine to go into limp mode, if effect it's operation at all? All the CRITICAL sensors and enablers required for the engine to run are wired DIRECTLY to the engine ECU!
Am I wrong? Does the engine ECU poll the BSI for the security code and transponder ID while the engine is running? Confused!

[Meannamate]

Hi ozvtr! Thanks! Yes. With the scanner connected and the ABS disconnected the engine starts and runs. Scanner and ABS disconnected it flags brake system fault and won’t start.

[wheeler]

I would agree that the CAN connetion should only be needed for initial starting. With the ABS unit unplugged this should interrupt the CAN & it shouldnt start. I can only think that the scanner is somehow bypassing this. As far as I remember the only ECU on the high speed CAN that is in ‘parallel’ on the C3 is the power steering.
Really need the vehicle details to get a wiring diagram for this. Next step for me would be to check the resistance of the CAN lines at the ABS connector. If that is OK i would bridge over the CAN lines to bypass the ABS unit & see if it starts OK.

[Meannamate]

Hi wheeler! I sent a pm to you with the VIN and registration number. Hope you can trace the connections and I’ll give it a go!

[ozvtr]

I would agree that the CAN connetion should only be needed for initial starting. With the ABS unit unplugged this should interrupt the CAN & it shouldnt start. I can only think that the scanner is somehow bypassing this.

We are talking about a connection literally at the other end of the CAN BUS from the OBD2 connector and (without looking at the wiring diagram) is physically severed from the CAN BUS by disconnecting the ABS ECU. I can't wait to see the circuit diagram!!!!!!

As far as I remember the only ECU on the high speed CAN that is in ‘parallel’ on the C3 is the power steering.

On the CAN network, the BSI, the auto gearbox (AL4 or sensodrive), the steering ECU and the ABS ECU were in parallel. Then the engine ECU branched off the ABS ECU. All the other ECUs were on a VAN (Vehicle Area Network). Pre-facelift CAN-VAN system.

If that is OK i would bridge over the CAN lines to bypass the ABS unit & see if it starts OK.

I'm putting money on the engine ECU now being on the CAN BUS proper and not 'branched off' the ABS ECU. I don't see why it was in the first place. Ooooooh! Hurry up with that circuit diagram.

[wheeler]

OK i have attached including an enlarged section of the ABS ECU for better clarity.
As i suspected the ABS system is in 'series' on the CAN network so unplugging it will break the CAN lines.
How i would now proceed is to resistance check the CAN lines at the ABS ECU first.
Make sure that there is nothing plugged into the OBD socket and that all ECU's are plugged back in as the should be apart from the ABS ECU.
Measure the resistance across pins 19 & 25 of the ABS plug then same again across pins 21 & 23. You should get 120 ohms on each.
If the resistance checks are good next i would bridge a piece of wire (ideally with a fuse) across the CAN high wires pins 23 & 25 then another separate wire across the CAN low wires on pins 19 &21. This wil bypass the ABS unit. In theory if the ABS unit is the cause of the problem the car should start normally although expect multiple warning lights/beeps about the ABS & braking system faults.
I would recommend that you check the codes printed on the wires correspond to the pin numbers on the ECU plug.