Well, 200k miles of using a crappy tail light design are what I think did it in. Current draw from the LCM to the tail light was not excessive when I tested the wiring. I have had to repair the tail lights a couple of times because of melted plastic and burns/pits in the metal. They really went cheap on these things but BMW is not alone....

I already ordered some chips off ebay, thanks for the link! I spent an extra $20 on shipping to get it here faster. Kinda pissed that I could not find a local source (USA) that could ship it faster. This is the first time Avnet has let me down.

I have done this type of work before, it does not look too bad. I have not had to deal with a waterproof coating before so that should keep things interesting. I will post more pics when I am done and let you guys know how it goes.

Well, I received the replacement chip and soldered it in. Not my best work but the waterproof protective coating was a pain to deal with. I had to do a lot of tip cleaning to make sure the joints were clean. I also damaged the trace at pin 6 so I had to add a wire to make up for it. It is one of the chip grounds.

Unfortunately this didn't 100% repair the issue and I think when this chip died it damaged the chip that controls the setting of the taillight between marker and brake. Right now the repair is working intermittently, about 50% of the time. On several occasions I have heard an electric humming like a transformer by the taillight right before it throws the warning light. This makes me think that there is a high-frequency switching part of the assembly that has been damaged. Is anyone familiar with the basic circuity of the LCM and what part I should look for next?

Well done with the repair. :thumbup:

The "buzzing" is something often you get when the driver circuit is overloaded, or at least working VERY hard. :confused:

You don't have an intermittent short to ground in your loom somewhere near the rear tails? That would fit both the cause of the original chip failure and the new symptoms... :dunno:

Thanks! I though it looked kinda sloppy from the pic and I was waiting for someone to hate on the job I did. Thanks for the positive encouragement!

There might be something going on with the light housing, I have repaired it a couple of times and it looks like there might be some rust/corrosion under where it connects to the wiring harness. Unless someone has a better idea I will try to isolate the light socket and see if that fixes the intermittent issue. Otherwise I will run a temp wire from the LCM to see if that helps.

Since the circuit trace got black and it wasn't just the chip that burnt I would say there must have been a short and still could be at times.

That is a good point. Does anyone know the layout of these boards better or know how the brake/marker light dimming is controlled? I am going to have to do more work to isolate the circuit.

Full disclosure - I don't know how this circuit works - this is guess work.

But, given that the IC driving the lamps doesn't have a heat sink or look to be able to dissipate much excess heat, any dimming of the lamp circuit must be via PWM (yes, same PWM as the control of the aux. fan - although used in a different way in this case). The noise you hear when it's faulty supports this theory. I'm guessing the LCM varies the mark/space ratio of a square wave going to the tail lamp to modulate the brightness. The PWM used for lamp dimming is usually in the 1-2kHz range, right smack in the middle of our hearing range. When your tail lamp assembly is faulting (short or similar to ground etc.) the IC is delivering a high current at 1-2kHz and you can hear this, via the protesting components. :yikes:

I should drag my oscilloscope out into the car park and check the tail lamp circuit- this would confirm the PWM operation. But it's cold and dark outside! :confused:

This fault just HAS to be an intermittent fault path to ground in the lamp assembly somewhere...

I'm assuming that the light circuit is connected to the solder pad at the end of the burnt trace ? These chips must have a really low on resistance, I recently replaced the chip on my rear wiper board and it was very similar to what your working on. The tiny SMD can carry 30 amps, albeit over two pins (just like yours) and had no heat sink! Has one now though ;)

Looking at the picture, is that a burnt resistor in lower RH corner ?

The picture does look like it is damaged but it is just the circuit board coating and the picture quality.