Der Strom merkt immer, wenn kein Kabel da ist. - Ohm's 4th law
This page contains some information about the electrical system(s) of my BMW 2V motorcycles. Most topics on this page were written from experiences that I made on my R80GS, and some others relate more to the R100TIC. However, most of it should be applicable to all BMW "airheads" - 2V boxer engines - from the /7 era onwards.
The usual disclaimer: I did these things with my own bikes and on my own risk, so if you repeat this with your bike, you do it obviously on your own risk.
Technical note: Most of this was previously on the R80GS page on this website, but due to the amount of material that page was getting too slow to load. Early 2007 I decided to spread the material across multiple pages.
When working on the electrical system (except, perhaps, for changing a light bulb), there is one golden rule: The ground cable of the battery is always the first cable you disconnect and it is always the very last cable you re-connect. To paraphrase with the greek alphabet - it is the Alpha and the Omega.
Why? Well, in almost all cases the negative pole of the battery ("ground", although this term is not physically correct here) is connected to the frame, the engine block and the like. In other words, any metallic part of the bike is connected to battery ground.
Now, imagine you want to unscrew the positive pole of the battery while the negative is still connected. Sooner or later in this process, the wrench (which is tightly "clamped" to the positive pole) will accidently touch a part of the frame, a mounting bracket or something similar ... which means, you will have a nice short circuit between battery plus and battery ground! If you are lucky and react fast, you may be able to withdraw the wrench, but there have been reports of wrenches that were first "electrically soldered" to the frame and then melted within a few seconds. I'm pretty sure you would like to avoid that kind of experience ;-)
At this point, you might know that a fully charged battery of 25 Ah can deliver 25 A for one hour, or - approximately - 250 A for about 6 minutes. 250 Amperes is quite a huge current: the starter motor of a BMW 2V boxer can draw up to 130 A, which is the reason for such a fat wire between battery and starter.
It seems that my GS "consumes" an alternator about every 50000 to 60000 km.
The first time this happened was in 1993-10, at about 61'000 km. I had stopped somewhere for shopping, then went back to the bike, turned the key and pressed the starter button.
"klick-klick".
Well ... that was the noise of the starter relay. The battery was too weak to turn the starter motor and in turn to start the engine. Fortunately, I could start by pushing the 220 kg, but the charge control light was smiling at me. Riding to the BMW shop nearby, diagnosis was clear: The alternator winding was open.
Changing it is fairly easy, but the part itself costs a few bucks ... about 150 DM in 1993. This was a re-built exchange part, as the original part was not available within a reasonable delay.
The second time this happened was a only few months later, at about 68000 km. I was in Paris for a weekend, stopped at at little shop to buy something, then went back on the bike and pressed the start button ... "klick-klick".
Bad, very bad for your holidays. Well, what do you do? First of all, get the bike running again. I opted for the run-and-jump method ... of course with a well-loaded bike and bags. I never thought that these bags could hinder me, but if you have to run besides your own bike they are really, really in the way!
Later on I went for troubleshooting. I knew that the battery was not longer charged correctly, but I could not really find out what was wrong and the BMW shops are relatively rare in France. However, I managed to go for several hundreds of kilometers like that until I was home. The important thing is to make sure you use jump start every time, and that you do not consume too much current (during the day I rode with the parking light instead of the headlight to save energy). The problem with all modern bikes is that they are full of electronics and once the battery is empty you are lost. Magnet-induced ignition would be nicer ;-)
I have been searching the source of this problem for several weeks later on. The voltage changed considerably with engine temperature (13.6 V at 4000 rpm when running in 5th gear, dropping to 12.6 V at 4000 rpm with hot engine). The charging light would light up when I turned the ignition on, but would not light up when riding, so everything looked well - but it was sure that the battery was not charged correctly. I checked the voltage regulator, the diodes, but no clue ... no, it couldn't be the alternator winding, that part had been changed just a few months ago ... but then I obtained the electrical values for the alternator winding! It should have a resistance of 3.4 Ohms, but it actually was only 1.2 Ohms. In other words, this was a partial short circuit of the winding! - It turned out that the problem was due to the rebuilt part; I got a new one (partially under warranty) and this time it was the original BMW part for 340 DM (1994).
The really weird thing in this case was that the charge control light would not light up, but the battery would not be charged correctly either - in other words, no warning, but you loose energy. This is a situation when a voltmeter on the bike would have been truly useful ... indeed, a few years later I installed such a device.
As I said, the bike eats up one alternator about every 50...60000 km, so the next rotor was due at about 103900 km (1999-09). This time it was a "nice" failure, with the charging light coming up and the rotor having a real interrupted winding. Nothing special to tell, but it cost me about 320 CHF.
Do this with a cold engine. You will need three "special" tools: a hardened steel rod 6x50 mm, an allen screw M8x30 and ... a hammer. As an alternative to the steel rod/allen screw combination, you may get an M8x100 bolt and turn the "outermost" 50 mm of thread down to a diameter of 6 mm.
(Left) The engine cover removed. (Right) Alternator stator, alternator rotor, rotor holding screw, and the "special tool" to draw the rotor.
If your charging system is OK, you must get a voltage reading of 13.6 V or better above 2000 rpm with the headlight on (!). Not more than 14.5 V. The charging system is supposed to be working from 980 rpm onwards.
Alternator rotor, new, across the contacts: 3.4 Ohm (± 0.3 Ohm). The part number of the alternator rotor is: BMW 12-31-1-244-642 (Bosch 1-124-033-031), which fits all 2V boxer engines from 1969 to 1994. It seems that there exists a reinforced version (for police bikes etc.) with part number 12-31-4-257-539, but I don't know any details about this.
To control the stator: Measure contacts U, V, W against each other. Reading should be around 0.6 Ohm (less than 1 Ohm in any case). Must be isolated against ground.
You may test the diode board either with a small test lamp or with any Multitester. The wiring diagram for the diode board is shown in your rider's manual and the following picture shows the diode board of my GS at 110'000 km. The device was scanned from the back so that all connectors are visible; the colors were somewhat falsified by the scanner.
Note that the D+ connector has two pins, you can use either of them. Another important thing is that the upper part of the plate connects to battery ground, while the whole lower part is directly connected to the positive pole of the battery (B+). Take care to verify insulation of the rubber parts around the bolts!
The following table shows the values that you should obtain when you test the
diode board (which has of course to be removed from the bike to be measured).
O
means "open", X
means "conductive".
If you have a diode tester, the connections marked with XX
should show twice the voltage of the X
-rated ones
(yes, there are two diodes in the path).
Please note that some Multitesters inverse the voltage on the test probes
when they are operated as Ohmmeters, so before you conclude that a
diode is defective just inverse the leads and test again.
Positive test lead on | ||||||||
B+ | GND | U | V | W | D+ | Y | ||
---|---|---|---|---|---|---|---|---|
Negative test lead on | B+ | - | XX | X | X | X | O | X |
GND | O | - | O | O | O | O | O | |
U | O | X | - | O | O | O | O | |
V | O | X | O | - | O | O | O | |
W | O | X | O | O | - | O | O | |
D+ | O | XX | X | X | X | - | O | |
Y | O | X | O | O | O | O | - |
As a side note: The small diodes are BYW72, the big ones Bosch 1405-0241 and 1406-0241, or 405.426 and 406.424 (three each). Reportedly, these can be replaced by Motorola 1N3660 (cathode to case) and 1N3660R (anode to case), which are rated 100 V and 30 A at 100 °C.
The function and testing of the Bosch-type voltage regulator are described on its own page, also on this site.
A simple 2-LED "voltmeter" to survey the health state of the charging system, also on this site.
It seems that the grey ignition coil of the earlier GS models has a design flaw that causes the coil to crack after some years of usage.
In summer 2001 I found that mine had a crack, too, but I only noticed this in a heavy downpour (what else?), when the engine suddenly stopped. After a few minutes I could start again, arrived home and upon closer investigation I found a fine "line" running along the body of the coil.
I first mistook this for a residue from plastic moulding, but it was indeed a crack, allowing water to enter the coil. With the remaining heat of the engine, the water would evaporate and allow to start again after a few minutes.
This crack (nicely visible in the upper image) caused the secondary (HV) coil to break - I measured "infinite" resistance between connectors 4a and 4b - but the bike would still run without problems for several hundreds of kilometers, if weather was dry! It seems that the high voltage generated was enough to bridge the broken winding inside the coil and all that even without causing excessive heat (not more than hand warm). However, after a few hundred km the idle started to degrade and the replacement coil arrived just in time.
Looking for a spare part I found that BMW had redesigned the ignition coil several times and it seems that the new model is more durable. Price for the new coil was about 190 CHF in summer 2001, but I decided to get a used one. The photo shows the old design (left) side by side with the revised version (right).
Electrical data: For the "old" (gray) coil, resistance between 1 and 15 (primary) should be about 1.1 Ohms, between connector 4a and 4b (secondary) 7.5 ... 9.15 kOhms. The revised version (red) has 0.5 Ohms primary and 12.8 kOhms secondary. Spark plug with cable, 5 kOhm.
The BMW airhead ignition coils have a rather "special" design: the secondary coil is not connected to the primary coil but both ends are free floating. This works pretty well, but it has the disadvantage that the spark on one side jumps from the center electrode to ground and on the other one from ground to center. If one of the spark plugs gets dirty, if draws current off its counterpart.
A solution for this problem is to use the classical construction for ignition coils, where the primary and secondary winding are connected to each other. One such design is the "TwinCoil" from Silent Hektik, which I installed in 2001-09 (111'800 km). It uses one common primary winding and two separate secondary windings, both of which are connected in parallel to the primary coil. In this way, the spark always jumps from the center electrode towards ground. This change brought a better response to the throttle and I have the feeling that the cold starting behaviour has improved, too. - Electrical data (for quick testing): primary 2.1 Ohms, secondary 12.3 kOhms each, cables about 2 kOhms each. I don't know their factory tolerances.
At about 110 Mm, I noticed some unusual noise as the engine fired up. Everything essential was OK, but there was a loud "screetching" noise for about one second when the starter should release the flywheel. I thought that it might have been a lack of grease in the starter, causing it to stay with the flywheel when it should release and decided to disassemble it before things would get worse.
Special tools required: A very short, compact tubular wrench or ratchet 13 mm. If you want to dismantle and grease the gearing, get a "T20" torx screwdriver and silicon grease (see below).
My GS has a "Valeo" starter, type DR6RA-15. It is composed of three main parts: The starter motor itself - big black cylindrical part -, the magnet switch - smaller black cylindrical thing with two big copper contact screws - and the "gearbox", which is the conical aluminium cover that goes over the flywheel.
To inspect the coal brushes, unscrew the two little copper screws on the (big) motor housing and withdraw the "end cap". This part may seem to be stuck, but it isn't - it has a very tight rubber seal and comes off smoothly when you pull gently. Do not pry with a screwdriver or other tools - the plastic part that holds the brushes is rather fragile. The interior should be clean and almost free of dust. - Now, inspect the brushes. On my GS, they were in excellent shape and 2+ cm in length although the bike had gone 110000 km at that time. Nothing to do, so I put it together again.
To grease the starter "gearbox", remove the aluminium cone (three allen keys plus a T20 torx key. Unfortunately you cannot replace the latter with an Allen-head bolt as it has a special threading!). Clean the parts and grease the windings with silicon grease. The BMW manual specifies Bosch PZ2V3; as I did not have this at hand I took temperature-resistant silicon grease from the chemistry lab (Molykote 111). Then, reassemble the parts.
... and that's all. Reassembling goes exactly in the reverse order.
As for those infamous mounting bolts, you may want to replace them with two Allen-head bolts M8 x 45. Reportedly it is even more convenient to swap the two bolts that hold the starter motor with those that attach the bottom of the air filter housing, allowing you to remove both the starter motor and the bottom of the air filter housing much easier ;-)
Most Valeo starters tend to fail over time, due to "lost magnets". I experienced this on my GS in summer 2016.
These magnets are part of the starter motor itself and are glued to the outer part of the housing. Over time, they may simply fall off, probably due to thermal and/or mechanical stress. Loss of one or two magnets may go unnoticed at first, but the resulting debris will block the starter motor in the most inconvenient moments.
The symptoms for the "magnet failure" are somewhat similar to the symptoms of an empty battery, but there is a subtle difference of the noise when you press the starter button:
In the latter case, the starter motor is blocked and the magnet debris is blocking the rotation. Do not re-re-try to use the electric starter: it, or its cabling, will overheat! Note that you can still ride the bike — you just have to prepare for jump starting, e.g. by parking downhill. We spent a whole weekend doing this, two-up and loaded with luggage ;-)
It is possible to repair such a defective starter by reattaching the magnets (see the links at the end of this section). However, unless you are stuck in a remote place, it is probably more efficient to replace the complete starter motor, or to re-use parts from a car junkyard.
Probably the simplest alternative is to buy a new starter: as of 2016, prices for China copies of the original Valeo Starter have dropped to and below the 100-EUR range. Many of these replacement motors feature brackets that secure the magnets in place, in addition to the glue. You may want to look for those.
In 2016-06, I replaced the failed starter motor of my GS by such a brandnew motor, obtained from auto-elektrik.de (underlying domain is no longer active?). Carsten Tiedemann, the company's owner, takes the "raw" China copies and reworks them so that they should last.
If your Valeo starter is still working well, you can perform preventive maintenance: Open the housing, remove the rotor and then glue the four magnets in place with a strong, thermally stable glue. I have used epoxy resin for this - the same resin that I use for the weather-proof casting of electronic compounds.
Of course, the process had to be done in four steps, leaving enough time for the resin to cure and then rotating the housing 90° after each step.
Another possibility is to scavenge parts from another starter, obtained from a car junkyard.
Valeo starters of the type we need were used in a number of smaller cars, such as Opel Corsa, Astra, Kadett; Renault Clio; Citroën AX, BX and C15; Peugeot 205, 206, 306, 307, 309, 405, etc. Those who live in the USA may want to look for Saturn cars. Do not worry too much about the shape of the cast aluminium cone; the important point is that you ask for a Valeo DR6RA or DR6A. (It seems that the starter of the Renault Laguna is even a true 1:1 replacement for the complete BMW part?)
The part swap is pretty straightforward. On both starters, proceed as follows: