Preparing to renovate the charging system on my ’84 R100RS, I assumed that the bike had an electronic voltage regulator because of the year it was built. The metal case however led me to mistakenly think I had a mechanical voltage regulator. I learned that early electronic voltage regulators had metal cases like the mechanical versions – just with a somewhat shorter metal case. Opening the metal case (by simply removing the red tape) I confirmed having an electronic voltage regulator. Notice that this model has a small potentiometer – presumably that controls the output voltage. It’s marked with a white paint stripe to show its set to the factor setting. A new OEM electronic version:
Changing the rear tire on my ’84 R100RS I discovered a bad rear wheel bearing.
The process of learning how to remove, install and adjust new bearings was aided by Duane Asherman and Robert Fleischer. Both men have made exhaustive studies of this service, documented them on the web and were very kind to assist me via email. And Robert Fleicher’s encyclopedic www.bmwmotorcycletech.info covered both setting preload and removing and installing bearings in extraordinary detail.
I was fortunate to discover that my 1984 model R100RS has wheels that allow removal and installation of wheel bearing without first heating the hub. Removal was a simple matter of using a 30mm blind bearing puller, and installation using a standard race installation driver – both quite inexpensive tools on eBay or Harbor Freight.
The bearings on my R100RS’s wheels are taper bearings – similar to the ones I used to lube on my first car – a ’67 Ford Mustang. The preload setting on the old Ford however was a simple process of tightening the castle axle nut and then backing it off a quarter turn, and installing a cotter pin through the castle nut.
I learned from Mr. Asherman that the taper bearings on my R100RS are so over designed for this application, that I probably didn’t need to be much more scientific to get this job done and still get good service life from new bearings. But I couldn’t resist doing the preload adjustment to the specification of 15 to 30 newton-centimeters. (about 21 to 42 ounce-inchs)
In a nutshell, adjusting the taper bearing preload on my motorcycles wheels was a simple matter of selecting the proper size spacer (referred to as the “wedding band” spacer) that controls the space available between the inner and outer bearings’ races once the axle nut is tightened.
There are two tricks involved with measuring the bearing preload. First spacers must be used to allow tightening the axle nut while the wheel is not installed on the bike. I elected to use one of the suggested methods – I bought a dozen “top hat” spacers employed on the axles that could be stacked along the axle to take up the necessary space. This might have been more expensive than one of the other suggested methods such as cutting a tube to length. But the “top hat” spacers were cheap enough – and super easy to use.
The second trick is to measure the torque required to rotate the axle after the wheel was assembled with new bearings and spacers and the axle nut properly torqued. Both Asherman and Fleischer advocated different methods – and I came upon a third after searching YouTube for videos about setting taper bearing preload.
My method was to use a dial torque meter. I was able to find a used one on eBay for under $100. The Snap-On TQS-025 is scaled in inch-ounce from 0 to 48 – perfect for this job. It’s 1/4″ drive allowed me to attach a socket to fit the axle nut, and rotate the axle to take the torque measurement. The “tell-tale” on the TQS-025 recorded the maximum torque registered while rotating the axle. This video shows how I used the TQS-025 to measure preload – indicating 26 Inch-Ounces. Note the stack of “top hat” spacers to allow tightening the axle nut.
I also tried a different tool – a Proto 6104, a torque limiting screwdriver. This tool allows setting a maximum torque on the inch-ounce scale (0 to 100) using a setting similar to a “clicking style” torque wrench. This tool can be used to bracket the torque of the axle by choosing settings lower and higher than the actual preload torque.
The Proto 6104 was a little more expensive than the Snap-On TQS-025, and less precise in measuring preload torque. But it probably has more useful applications in the workshop – specifically not allowing drive shaft filler plugs to be over-torqued.
This video shows the Proto 6104 set to 22 inch-ounce and not rotating the axle because the preload exceeds that torque value. Note the stack of “top hat” spacers used as a single spacer to allow tightening of the axle nut:
This video shows the Proto 6104 set to 24 inch-ounce and rotating the axle – indicating that the bearing preload is less than this value:
These two tests allow “bracketing” to test the preload of the taper bearings.
These photos are of diode boards used on BMW motorcycles made in the late 70’s and early 80’s – in particular, my ’84 R100RS. I’ve read that these boards are also used on some Moto Guzzis from these years.
The right board in the top 3 photos and the top board in the 4th photo are of the diode board that I just removed on my bike. I’m guessing it may be original – or one installed in the bike’s first few years. The left and bottom boards are of a new part.
I am evaluating the old board for signs of wear – particularly heat stress. And I’m comparing the two boards for differences. Doing simple electrical tests, both diode boards appear to function properly.
Photo 1: The new board is obviously cleaner, and has different markings on the 6 large diodes. On the new board (left), three of the large diodes are labeled 041-N and three are labeled 042-N. (the difference is these diodes pass current in different directions.) On the old board, the diodes are labeled E1110 and E1120. I’m trying to find out if these different diode types have different capabilities in terms of their rated load and heat capacity.
I have been able to locate datasheets (in German) for the E1110 and E1120 diodes on my bike’s diode board: e1110-data-sheet e1120-data-sheet
I’ve not yet found datasheets for the new diode board’s diodes marked 041-N and 042-N.
Photo 2: This photo shows some differences in the physical construction of the diodes on the two boards. The new board (left) has somewhat smaller diodes. Besides the 6 large diodes on each board, there are small diodes inside the board. Theses are physically much smaller in the new board. The older board (right) also has wires spot welded to the large diodes to connect them to the PC board.
Photo 3: The old board (right) has grey paint that is flaking. I’ve read that the grey paint signifies that this board was made by BOSCH. The new board (left) is made by Wehrle. A version of this board made by Wehrle were defective. The diode leads didn’t make proper contact with the PC board, resulting in heat build up that melted the solder. In the early 80’s, I replaced the diode board 2 or 3 times on the new ’84 R100RS I owned because of this problem. In this photo, I can’t detect any failure of the diode solder connections.
Photo 4: This photo shows the two part number stamps. The board from the bike shows part number: 1 244 063 3. The new board shows part number 11 244 063 04/14. I’m guessing that these are the same part number – the trailing “3” and the “04/14” are possibly date stamps, and the older board is missing “11 2” at the beginning because BMW enhanced it’s part numbering system.
Opening the metal case (by simply removing the red tape) I confirmed having an electronic voltage regulator.
Notice that this model has a small potentiometer – presumably that controls the output voltage. It’s marked with a white paint stripe to show its set to the factor setting.
A new OEM electronic version:
