Gift horse - part 12

It appears the charging system isn't working: the generator warning light (red) does not light up when the ignition key is turned, and, with the engine running, there is no charge at the battery. The starting circuit is fine and relies on a new Bosch battery, which is how the engine is able to start.

As with absolutely everything on these BMWs, the electrical system is very much its own thing and you just have to accept the logic that produced it and not get bogged down in "why did you do it this way" or "it should have been done like this", or you'll go nuts. Instead, as there are MANY checks to perform it's better to get stuck into troubleshooting and sort this sucker out.

If I understand correctly, the wiring as it relates strictly to the 'GEN' light goes something like this:

Current goes from the battery to the ignition switch, when you turn the key current can then flow to the lightbulb for the 'GEN' light, after which it goes to the voltage regulator under the tank, and from there it goes to one of the brushes on the stator; current then goes from the brush to its slip ring on the rotor, it goes through the rotor winding and out the other end on the second slip ring, up the second brush and then to ground. Circuit complete, lightbulb lights up. Makes sense to me, but of course I could be wrong. Also, I am fully aware that I just infuriated every electrical engineer on the planet with that "sketch". Sorry please.

Step 1: with the battery disconnected, I removed the front engine cover, then reconnected the battery, turned the ignition key and checked the dashboard. Nothing.

I tried moving the wiring loom that connects the rear of the instrument pod to the headlight but still no light. 

I then moved to the stator, disconnected the brushes' terminals and grounded the blue/black wire from the voltage regulator to the body of the stator: this time the light came on nice and bright, indicating that the bulb and all wiring, connections and components up to the brushes is in good working order.

The next thing I checked was the rotor, first by lifting the brushes and insulating them from their slip rings with a strip of rubber. I then tested for continuity from one slip ring to the other and got a reading of about 3.5 ohms, which is acceptable and would suggest that the rotor is also good. Another thing to test is that the rotor winding isn't shorting to its steel housing, so with a probe on one slip ring and the other on the steel casing, we get a reading of 1 meaning infinite resistance. One more thing off the list.

The engine fires up easily but now the 'GEN' light stays on and there's no reading at the battery terminals, meaning it isn't being charged.

Moving on, I removed the diode board and the wiring that connects it to the stator. I checked the wiring first, and the multimeter shows continuity along all three wires, that's good. The single wire that connects the negative output of the diode board to the Y center tap of the stator is also fine:

Time to check the diode board itself: the first thing we noticed was that we only had two mounting bolts, namely the two bottom ones, which are on the positive side (and therefore insulated) of the aluminum housing. I later refitted the diode board with four Allen bolts, and with clean connections to ground it to the engine housing:

There are eleven diodes in the diode board, with no easy way of determining a specific fault. It is however possible to determine if the component overall is working, by connecting one of the multimeter probes to the aluminum housing and the other to the underside of the spade terminals for the three phases. With three readings approximately in the same range, I'm satisfied that the diode board is "good enough". Had I found a fault somewhere, it wouldn't have mattered much which diode(s) specifically was fried, as it implies replacement of the whole board anyway.

I also checked the heavy gauge wire that feeds the battery + terminal from the diode board, again, all good.

I checked the stator spade connectors (UVW phases) and determined all were good, with no shorting to the stator housing.

I checked the brush holder to make sure it was assembled properly as the DF terminal must not find a ground to the stator housing. One more thing off the list.

Checked the wiring from diode board to voltage regulator; not ideal but good enough for now.

I also took a look at the starter relay and cleaned the spade connectors as it was (and still is) showing signs of corrosion. It seems to be working correctly, though as with many of the original components on this machine, it's probably a good idea to check this periodically and be prepared to replace it with a new one if it fails.

With all of the above checks performed, many components inspected, connections cleaned and everything properly back together, I connected the battery, turned the key and the bloody light still wouldn't come on.

I opened up the headlamp and checked the two fuses in the back of the noodle bowl: one was rated 6A, which is wrong, the other 8A, which is correct, except that it was blown. You don't just replace a blown fuse without knowing why it blew in the first place, however this is still a test phase so that's exactly what I did: two new 8A fuses, turn the key, 'GEN' light on like a Christmas tree, it turns off when the revs pick up, and the battery receives about 13.7v

It's possible that the rotor was completely de-magnetized and just needed a little helping hand to get things going. There's also the matter of that blown fuse I found, so I need to check if it happens again and why. As I'm not an expert at this, I'll keep looking into this but for now I'm satisfied that the system seems to work properly.