6/94 Brake Light Circuit

[DavesNotHere]

So, the dreaded and usually simple brake light circuit problem. Except it seems to be not so simple this time.

Fuse 8 (7.5A) blows after pressing the brake pedal. It's not always immediate, but it's within a press or two or three - usually.
When fuse 8 blows, the MID shows "Brake Light Circuit"... and with a mechanical brake switch installed, the brake lights do not work.

With an electronic brake switch installed, all three brake lights work - until fuse 8 blows.

The chassis (CC89633) is a 6/94 build, and I cannot find an electrical diagram for that build month. (It would be awesome if someone has a link)
It is supposed to use an electronic (hall-effect?) brake switch. It *has* worked with a mechanical switch before. With the current (mechanical) switch: - pins 1-2 and 3-4 are closed with the pedal depressed and both open with the pedal not depressed. Both the '94 and '95 schematics that I have access to show mechanical brake switches.

When Fuse 8 blows with the mechanical switch installed, the car is driveable without brake lights. With the electronic switch installed, the transmission can't be moved from park (and this latter is the correct behavior according to the dealer).

As best I can tell from a 5/94 manual (which may not be correct for a 6/94 build), section 6325.0-00 (page 394) Fuse 8 feeds both the LKM directly (pin 7) on the downstream (normally open) side of the brake switch, put 12v on pin 7 of the LKM and pin 53 of the EML when the brake switch is depressed. That should light the brake lights and disengage the cruise control. The 9/94 manual may be "more" correct, but I'm not certain due to the switch issue.

Link to the 5/94 manual: http://www.armchair.mb.ca/~dave/BMW/e31/e31_94.pdf
Link to the 9/94 manual: http://www.armchair.mb.ca/~dave/BMW/e31/e31_95.pdf

Four things:
1) Without the manual showing the electronic brake switch and how it is wired, it's not clear to me how to troubleshoot this.
2) Does anyone have a pinout of the LKM? I can't find that yet. Yes, I can get under the dash and trace the wires if I have to, but not having a month-correct schematic...
3) What normally causes Fuse 8 to blow? I know that this error is usually the switch, but I'd prefer not to replace a $130 electronic switch before checking other usual suspects as well, or first.
4) How does one test the electronic brake switch? IMO, it should have 12v and ground and toggle the remaining two pins, but I don't want to take it apart to create the schematic for it if I can help it - in case it isn't actually failed.

Fwiw, I'm the latest owner of this car, working through some obvious sins of others, and I intend to keep it. It would be nice if I could drive it again...

Best,
Dave

[Timm]

The brake switch is a double-pole, double-throw switch. One pole is used by a number of systems (EML, AGS and TPS on the 840) to detect if the brake pedal has been pressed - and reduce power/select a different program in the EGS etc if it has been. The other pole is used to switch the brake lights on (powered by Fuse 8). BOTH poles are parsed by the LKM to ensure that the brake pedal switch is working, it expects that if one pole is closed the other one must be - and if one pole is open then the other should be - if either of those is not true you will see the 'Brake Circuit' warning in the MID (and that is different to the 'Brake Light' message which detects if a BULB has failed).

So, first of all, stop worrying about the circuit diagrams, they are all pretty similar and show the same circuit between years but have been sectioned differently. There are only a few diagrams available and they cover a number of years is some cases (the 95/96 covers 1995-1999) The 12V from Fuse 8 is used to power the brake lights via one pole, the 0V connection is used by the other pole to signal switch position to other circuits.

Circuit here: https://www.meeknet.co.uk/e31/e31_94.pdf#page=394

You need to separate the problems - does the 0V side blow the fuse (very unlikely) or does the 12V side blow Fuse 8 (sure to be) - so remove the switch and supply the BLACK/RED 12V output with 12V - does it take excessive current - if it does then check all the brake lights including the high-level one for an errant bulb!

[DavesNotHere]

Thank you, Timm.
I'll look into it this afternoon and post back here.
I'm curious about the difference in brake switches. According to the VIN, mine is supposed to have the 61318356527 switch (white collar, electronic innards, fat aluminum heat sink), not the 61318360417 version (red collar, DPST mechanical). I'm not aware that they're interchangeable (at least not according to my local BMW dealer). It would be refreshing knowing that it's one of those expensive bits and bobs that can be replaced with a more common part if need be.

41ccXCh1T5L.jpg

Hopefully it's just an errant bulb. There are several modules triggered by the 12v on the switch output; hopefully it doesn't come to that. Should just bypass the LKM... lol (no, not really).

Best,
Dave

[Timm]

They are exchangeable parts - the electronic switch supersedes the mechanical one but they both do exactly the same job. The electronic one is a Hall-Effect device but mimics the mechanical action of the two-pole two-throw device.

[DavesNotHere]

Thank you for the info on the electronic switch.

Well, here's what I've found out so far:

- I should invest in the Buss Fuse company. I'm going to need more 7.5A fuses before this is done.

- Pin 2 (blue/red) from the mechanical switch toggles 0-12-0 (pedal up, brakes applied, pedal up) as expected. On the electronic switch, it toggles 9v-12-9v. Both switches seemed to work correctly - more on this later. The carrier for the LKM II, GM2, and EKM is loose in the footwell, the modules are installed in the carrier.
- The three brake lighbulbs are Osram 7506. They read 0.6 ohm each, cold. For 21w, they should draw 1.75A each.
I do not have a suitable power supply for where the car is at right now, so I looked at it a little differently than you suggested, Timm, but I think that it will make sense. I can get one if it comes down to it. The three bulbs together should draw 5.25A. That leaves 2.25 A of headroom for other loads on Fuse 8.


- With NO bulbs installed, I could not get the Brake Light Circuit error to flag on the MID. It also did not trigger the Brake Lights Failure error either.
- With ONE bulb installed, the Brake Lights Failure error triggers. I could not get the Brake Light Circuit error to trigger, and the fuse didn't blow.
- Same with TWO bulbs...
- With THREE bulbs installed, the Brake Light Circuit error did not trigger for several cycles of stepping on the brake, checking that the lower and high brake lights worked, and then releasing the brake. The Auto Trans pin release worked as expected.
Now the weirdness (of course...):
As soon as I shut the hood, stepped on the brake, and shut the driver's door, the Brake Light Circuit error triggered. Fuse 8 was blown.

I swapped the mechanical switch back for the electronic version, and it did the same thing. Interestingly, when F8 blows, the Transmission pin does not release with the electronic switch, but it does release with the mechanical switch. Weirdness...

So, being out of 7.5A fuses at this point and it getting too dark to see, I'll have to wait until tomorrow to look at it again. And of course I buggered-up the mounting clips on the electronic switch, so that will need to get replaced at some point, or the tie-wrap thing, assuming that it matters.

I'll get new bulbs first, just in case one is a problem. The high-mount bulb is clearly older than the others. Maybe that's a clue. dunno...

Thoughts?

[Timm]

There's a lot of current going somewhere! A 7.5A fuse will fail at twice the rated current within 1 second - so that's a 15A draw somewhere. You can try disconnecting the cruise control actuator as that is powered by Fuse 8 as well - but my guess is the brale light bulbs....

[DavesNotHere]

So... yeah... It's something else than bulbs or the cruise control actuator. Meh.

I replaced the two 7506 taillights with Sylvania LEDs (7506SL.BP2, supposedly compatible with bulb sensing), the high lamp with an 1141, and disconnected the cruise control actuator in the engine compartment as follows:

- Replaced Fuse 8. Replaced old 7506 High stop light with 1141 (1.44A vs 1.75A). Started car, stepped on brake. Fuse 8 blows right away.
- Replaced Fuse 8. Replaced Right stop light with LED. Started car. Stepped on brake. Fuse 8 blows right away.
- Replaced Fuse 8. Replaced Left stop light with LED. Started card. Stepped on brake. No problem. Ran car a few minutes. stepped on brake. No problem. Stepped on brake 1x/second 10x. No problem. Wedged brake pedal "depressed" for 30 secs. No problem. Stopped and started car. Stepped on brake. Fuse 8 blew.
- Replaced Fuse 8. Disconnected the cruise control actuator in the engine compartment (top connector of the cruise control). Stepped on brake. No problem. Drove the car 50' and stepped on thr brake. Fuse 8 blew.
Decided it was late and something stronger than tea was in order. Parked until tomorrow.

I hate to think that it's the LKM, but it's not a bulb or the cruise control actuator. It *is* definitely a load fed by Fuse 8 unless it's a frayed or pinched wire in one of the harness bits. Time to spend more time in the schematic. I don't want to swap LKMs (and possibly damage the new one) unless everything else has been eliminated.

FWIW:
- The BMW dealer gave me three Eico 1156 bulbs because that's what they are currently using. I didn't install them as they're 27 watt and 2.1 running amps - which would obviously contribute to the load problem. I also didn't say anything to the dealer about it; no doubt it can wait for another day.
- The Sylvania 7506SL LEDs are visibly dimmer than the stock 7506 OSRAM or Sylvania bulbs. They're coming out as soon as this issue is addressed. I don't know if the Sylvania ZX2 LEDs are any better. They look like they have larger heat sinks, so maybe the chips are bigger and they're closer to stock brightness. Might be worth a shot. The LEDs did not trip the failed bulb sensing circuit.

Thoughts?

[DavesNotHere]

Wrapping it up - more or less,

Disconnected the black connector from the LKM. Performed the following tests:
- Ignition off. Brake pedal up. Checked resistance to ground of the following pins IN the black connector:
Pin 7: 2k ohms +/-. Seems good. In a perfect world it would be "infinity", there are loads on it (ASC+T, AGS, Cruise Control Tempomat). 12v/2k won't make a 15A load, so that's not the problem.
Pin 6: >2M ohms. Left rear brake light. With LEDs installed, this is fine. No short to ground at least.
Pin 8: >2M ohms. Right rear brake light. As good as pin 6.
Pin 17: 1.1 ohms. High stop light. This is about right for an 1141 bulb.
Well, no obvious problem there. Therefore the problem should be "upstream" of the LKM's black connector or the LKM itself.

- Test to see if the problem is upstream of the LKM:
Verified Fuse 8 is good. LKM black connector disconnected. Brake pedal up. Key in Run. (engine running - might as well charge the car while I'm at it...)
Pin 7: 0v with pedal up, 12.2v with pedal down. Seems okay. If the car is charging at 13.4v, that's 1.2v of "drop", which isn't a 15-amp load. Good enough for now to move-on...
What I should do is check the voltage on pin 1 of the brake switch to see how much of a load there actually is. I'll get around to that tomorrow - at least to know what the voltage there is without the ASC+T load on it (that happens when the brake pedal is pressed).

- Test to see if the problem is in the LKM:
Verified Fuse 8 is good. All three brake light bulbs removed. LKM black connector connected. Brake pedal up. Key in Run. (engine running, as above).
Waited a minute to see if the fuse would blow. It did not. The Brake Lights Failure MID error flagged and was expected.
Pressed the brake pedal several times. Waited again, pressed the pedal several times. Blocked the brake pedal down. Waited a minute or so. Fuse 8 did not blow.
Installed a new 7506 bulb in the Right brake light. Repeated the above test. The light lit and Fuse 8 did not blow.
Moved the bulb to the Left brake light. Repeated the test. The light lit and Fuse 8 did not blow.
Moved the bulb to the High brake light. Repeated the test. The light lit and Fuse 8 did not blow.
Installed the Left and Right brake lights. Repeated the test. The lights lit and Fuse 8 did not blow.
Installed the High brake light. Repeated the test. The lights lit and Fuse 8 did not blow.
So, it looks like the LKM does not have an internal fault, or if it does, it's intermittent.

Decided to drive the car because it seems that the problem has occurred (sometimes) around the same time as the ASC+T test fails (which it often does and I get the two ASC lights... I do need to dig into the electric motor for it, but haven't done that yet). So, I shut the trunk. The brake lights go out and Fuse 8 blows. Woot! at least it's something "mechanical" or affected by vibration.

I'll finish editing this post in a bit - have to take the dog out...

[DavesNotHere]

Continuing...

I did recall that other posts have written about frayed wires in the trunk hinge area, but neither the lower or high brake light wires go through the looms near the trunk hinges. They *are* part of the looms that the trunk lights branch off of, so perhaps stresses from the hinges moving the loom branches get back into the main looms that wind by the batteries and across the lower trunk latch bit.

Next tests: Isolate the loom area where the problem is suspected. To do this, I removed the three brake light bulbs and replaced them one-at-a-time with 7506SL LEDs. The point of this was to reduce current so that only an obvious mechanically-induced issue should cause a load that blows Fuse 8, not a cumulative light bulb load.

Of interest during this: The 7506SL LED would not work at all in the High brake light position. I swapped it with a different socket that I knew worked because it had a working 1141 bulb in it (that I installed and watched light before using the socket for the LED). I also measured 12.2 V across the socket terminals with the LED installed and the socket locked into position in the light. I'm guessing that there's some sort of mechanical interference between the LED itself and the High brake light plastic lens that pushes on the LED and makes it lose contact inside the socket. The same LED and socket work fine in the left or right brake light position, and the 1141 bulb in the same socket works in the High brake light position. Anyway, something to look into later. Back to troubleshooting...

So, Verified Fuse 8 is good. Brake pedal wedged down. Key in Run (engine running). Right brake light 7506SL LED installed. Opened and shut the trunk numerous times. No problem.
Rinse and repeat with the Left brake light. No problem.
Rinse and repeat with the High brake light (1141 bulb installed). No problem.

Obviously time again for more than just tea, but it's still light out and I really want this over and done...

I unplugged the rectangular connectors on the lower brake lights, took out the High brake light bulb and started tracing the wiring looms by hand, looking for the SW/GN (black/green) wire on the left loom and SW/BL (black/blue) wire on the right loom (which is also on the left loom). I didn't find any obvious problem where the wires were contacting a ground or the body - no marks from arcing or gases or exposed wires that I could see, but there were a few questionable areas:
Some tight spots on the left loom around body areas and some metal bits, screw threads, and nuts that stick out behind the jack, the loom by the left trunk hinge has frayed sheathing where it flexes, but the wires look fine, and the left battery negative cable is sandwiching a bit of bundled loom between it and the body*. <-- My guess is it's this last thing, but for the life of me I can't find a point of electrical contact to ground - at least not yet.

*I installed a pair of H8's earlier this year and they take up just about all the available space. The left battery's negative cable doesn't have much room between the battery and the body before it connects to the common ground for both batteries nearby. Yeah, I'm thinking that the root cause of this is trying to do a good thing and install better batteries. Go figure that no good deed goes unpunished... lol. Well, we'll see eventually, I guess.

I un-sandwiched the loom from the battery cable and looked it over. It had a pretty good crush going on, but I didn't see any electrical marks on the body or frayed areas in the wiring. I'll look again tomorrow when it's lighter out. I'm guessing this is where the problem is, but it may not be - perhaps it was anchoring the loom and it was abrading somewhere else that was moving. Who knows?

I looked over the trunk lid loom on both sides, removing the rubber grommets from the body and checking the wiring on both sides. No obvious problems with the wires themselves and the grommets are supple and intact. I made sure that tight areas in the looms were loosened as best as I could manage.

I could not get to the High brake light loom to check the SW/RT (black/red) wire. I suppose that's under the card above the trunk metal. Saving that job for another day, and maybe I won't need to deal with it this time.

I plugged the rectangular connectors back into the brake lights and noticed the LEDs doing that nasty flicker thing - that they hadn't done before.
Looking up front: The brake pedal wedge had come loose, but also Fuse 8 was blown. I guess that happened while moving loom wires around, but I might never know.

Fuse 8 replaced and the LEDs stop flashing. That's new...
Brake pedal wedged down and the LEDs are lit.
1141 bulb installed in the High brake light, and everything is fine.
Several presses on the brake, and everything is fine.
Driving the car around for 5 minutes, on and off the brake, and everything is fine. Also, the ASC+T failure hasn't come on, which is new... Can't complain, but it's new...
I realized that the Cruise Control Actuator was still disconnected, so I connected it. No problem.
Several presses of the brake, and no problem.
I might as well test it with 7506 bulbs, so out come the LEDs and in go the two new 7506 bulbs.
Another 5 minutes of driving and braking and everything is fine. Granted, there's still an 1141 bulb in the High brake light, but it's plenty bright - better than the old 7506 that was there.

The housing holding the LKM, GM2, and EKM modules is still dangling in the footwell, but if that was going to be a problem, I would think it would have been one by now. I'll put it back tomorrow if all is well - I still have to fix a horn that only blows at 2 o'clock and it appears to be a wire issue and not the slip-ring or pogo (grrr), so I might let it hang a bit longer for access. Fun, Fun, Fun!

I'll go over the loom again before I tape it up better and call it a day. If I find anything (or nothing), I'll post an update. Hopefully whatever needed to get fixed got fixed, but for the life of me I didn't see anything absolutely definitive.
Timm, I much appreciate the help you gave me on this one. Kudos and Cheers!

[Timm]

My pleasure - I will warn you though that the Cruise Control will not work if you have LED's fitted in the brake lights - the Cruise Control actuator uses the brake light bulbs as a 0V return for the actuator clutch. This seems a bit weird, but this is an extra safety system - if the brake lights are ON then the actuator clutch disengages turning off the Cruise Control! More info here:

[DavesNotHere]

Thanks, Timm. No LEDs in the brake lights, it is.
FWIW, your point as I understand it is that the TEMPOMAT uses the pin 7 (12v brake switch output side) LKM connection as a ground pull-down. Makes sense from the schematic.

Just a thought, but it seems to me that as long as one of the three brake bulbs isn't LED, that there should be at least a 1 ohm connection to ground and there can't be much current from the cruise control otherwise the bulb(s) would glow. Might be worth a try when I'm not needing to fix something else. I suppose it could be addressed with a voltage divider and pull-up diode as well (relocate the TEMPOMAT connection to the center of the voltage divider).

Well, let's first see if I can't find out where the current was going in the first place so I'm not doing this again at an inconvenient-er time.

[Timm]

I tried it - and it didn't work. Tungsten bulbs have a very low cold resistance and you can stick a lot of current through them before they glow - the LKM does this all the time to check the bulbs are in place. With just the high-level bulb replaced with an LED the CC would engage, and a few seconds later it would cut out again....

[DavesNotHere]

That would suck... lol
I think I'll put some thought in on this. Someone might want a workable mod for LEDs; I wouldn't mind one. But first - gotta pass the biannual vehicle inspection here - time to find out what's up with the horn wire and get the the ASC pump motor running. Joys...

[DavesNotHere]

Update:
No issues since getting whatever was shorting in the left taillight wiring area un-shorted and replacing the brake light switch.
I never did find out what was causing the short, but the weather's been unpleasant and I haven't put much effort into looking further. I'll get around to it eventually, but I have bigger fish to fry this month (ABS pump motor failure or something related like ABS module wiring - hopefully it's just some corrosion in a connector).

The horn was fixed with a piece of 12ga stranded wire bridging two of the four bolts on the steering column rag-joint (loosen two adjacent nuts, wrap the wire around one bolt and under a nut that comes from the engine-side of the rag joint, slide the other end between the rag joint rubber and under the steering wheel-side of the steering shaft, leaving some slack for movement when the steering wheel is turned or the column up/down geometry is changed. Do this on one side of the rag joint since there doesn't seem to be much room to jumper across the outer edge of the joint. That gives a flexible and positive ground for 360-degrees and the horn now works in any steering column position and any turn angle). I'm sure the "correct" way to fix the horn is to find the wiring ground that has failed and trace it to wherever it was supposed to go, but since nothing is obvious, this fix will get it through the vehicle safety inspection here and should last for quite a while. I do have a concern about creating a ground-loop and currents going where they don't belong, but I need a horn "now" and a concourse level repair will have to wait.

A $12 URO brake switch for a '95 850Ci without any modifications worked just as well as the $130 BMW hall-effect switch, so I bought a spare URO 4-pin switch for the next time this happens. Yes, it sucks holding the brake pedal down with your head whilst getting the brake switch to lock-in properly. It also sucks trying to get the single 6mm nut back on the module frame - the steering column motors are in a totally bad place for hands or a socket extension. Gonna need more than tea to get that nut back on...