No announcement yet.

No +12V at coil - starting problem

  • Filter
  • Time
  • Show
Clear All
new posts

  • No +12V at coil - starting problem

    I couldn't post to the "FAQ --> No Start" section - I think that's where this post will eventually end up, but any help here is appreciated.

    I was out for a cruise a few weeks ago and the tach dropped to zero - pulled off to the side of the road and waited a few minutes, and she started back up. I thought I detected a bit of hesitation at lower RPMs, but it could have been paranoia. A few minutes later, she died again and wouldn't come back, so I had her flat bed towed back to the house.

    When I got back, I determined the following:

    - No spark at any plugs (pulled off leads and hooked up to grounded spark gap)
    - No spark at coil (pulled off secondary plug and hooked up to grounded spark gap)

    That told me that it was probably the coil, but the primary/secondary resistance checks out per other posts I've seen. I then looked for +12V at the coil while the ignition switch was in the RUN position and saw roughly +0.8V. I disconnected the battery and coil primary and checked continuity between the coil primary lead and the main relay socket, which checked out. I pulled the main relay, fuel pump relay, and OBC code relay and tested them all on my workbench - the relays all closed with +12V across their terminals, and continuity across the switched terminals checked out as it should. I reconnected the battery and saw +12V at the relay socket when the key was in the RUN position - this was in the same connector that leads to the coil that I checked earlier.

    Most of the other ignition diagnostic tricks I've seen by searching on the forum deal with ECU-side issues (i.e. the (-) side of the coil doesn't pulse to ground). To be sure, I checked my reference sensors (950 ohms on two of them, no reading on the third). I also cracked open the ECU to check for bad traces or burnt transistors and didn't see anything obvious. Regardless, it seems to me that I need +12V at the coil.

    Anyone have any ideas? Even though it clicks and bridges two of the terminals, should I still consider replacing the main relay? What about the OBC "black box" and relay located near the hood release? Could that third reference sensor (that seems to be open circuit) be an issue? During my searches, I saw that some people had run a seperate fused +12V lead directly to the coil - this seems risky, but is it worth trying?

  • #2
    I had the same problem and it turned out to be the ECU, but mine was obviously dirty and had rust spots on some contacts.


    • #3
      If you haven't got them already you can click here.... and select your year of M3 and
      download the full electrical diagrams for it.
      By the sounds of it you may have a poor connection or some
      corrosion along the line of the 12v supply to the coil.
      Mutlimeters can be a little deceptive when searching for faults
      like this. Although it may show 12v, a connector or coroded wire
      may only be able to pass very little current and hence not fire up the
      coil. Can be worthwhile using a test bulb to check for any current drop off.
      Might also be worth running a fused feed direct from the battery positive
      to the coil to see if the engine runs and confirm that the green wire supply
      is indeed your problem.


      • #4
        You seem to be an a good path in your diagnosing. Download the ETMs as recommended and identify the problem. I would stay away from short-circuiting around the problem...that just sounds like a good way to create new ones.

        So, check/replace the main relay (I always carry a spare). Check the OBC box relay. Check for a loose ground near the ECU (it happened to a friend). Also check that you are getting +12v on the secondary wire coming from the battery to the junction behind the coolant overflow tank.



        • #5

          I pulled the harness cover off (in front of the cowl) and found a few spots to tap into the green wire that terminates at terminal 15 on the ignition coil. Per the ETM, this wire should connect directly from the coil to the ignition switch. I exposed connections at the following points, working backwards from the coil:

          - Ignition coil
          - Cowl cover (pulled wire from harness and cut)
          - ~3cm dia. multi-pin connector (the one that screws together) - pulled back rubber guard on each side
          - OBC code relay connector (2-pin white plastic under dash) - stuck multimeter probes in each side
          - Multi-pin connector under steering column - stuck multimeter probes in each side
          - Ignition switch - not bad to get to if you pull the steering wheel

          I connected the battery, put the key in 'run', and carefully checked for voltage drops across wire lengths and connectors. I got ~10.5V across the OBC code relay connector, so I pulled the box and tested the traces on the board and the relay functionality - seemed to check out. That told me that it's probably an issue with the signal the OBC was sending to the box. My OBC has never had a functional backlight so I don't really use it; in any case, I tried a CODE reset (described elsewhere) to no avail.

          Following some advice I saw here, I pulled the 2-pin connector from the OBC code relay box and shorted the two green wires together on the male side. The car fired right up!

          I'm inclined to leave this as-is, since I don't really have much use for the CODE feature. I'll put in a permenant splice and solder/heat shrink everything back together - does anyone see any problem with this plan? Should I keep the relay box connected or just ditch it entirely?


          • #6
            I'd just just leave it as is with the two greens
            bridged. It works.
            Good fix.