My ’94 R1100RS had been “pulling” to the right since I bought it in 1998 with 13,300 km on the clock. Every front tire was wearing lopsided. Apparently I am not the only one who has noticed this, many of you mentioned the same. To better the situation I loaded the left hand side system case much heavier than the right hand side one. Later I found out that most ’94 RS owners on the list are doing the same thing. I rode over 150,000 km with an over-loaded left-hand-side system-case.
Some people say that the rotation of the engine causes the “pulling”. To convince myself that this was not so, I rode the bike 50 mph (80 km/h), put the gear shifter into neutral and turned the engine off. The bike was still pulling in the same direction as it had before, and to the same extent.
Finally I got to the point where I had had enough of all that and decided to do something about it.
The first thing I did was to take off the right hand side case and ride with only the left hand side case installed.
The result was great. The bike ran straight. I could take my hands off the handle bar without sitting out to the left side of the saddle.
In order to have some idea about the magnitude of the extra load I needed on the left to make the bike run straight I placed 25 pounds of lead into the left hand side case but nothing into the R.H.S. one. These lead pieces, 5 of them, have about a 3x3 inch base and about 2 inches of height. The result: The bike ran straight.
To find out the minimum weight I needed to balance the bike I removed one of the 5 pound lead piece. The bike was pulling to the right again, slightly. The result: 25 pounds of extra load is needed to balance the bike.
The extra weight on the left hand side of the bike produces a tilting effect, a “moment towards the left” in engineering terms.
This “moment” is the weight multiplied by its distance from the centre of the rear wheel. If the distance is measured in inches weight in pounds, the moment is in “pound-inch”. In order to obtain an equivalent “moment towards the left” by using the weight of the bike, the rider and the side cases act on the rear wheel only. I would have to move the rear wheel towards the right hand side of the bike, towards the final drive.
But how much???
The followings are approximate values:
I measured the distance between the centre of the rear wheel and the centre of the 25 lbs weight. It was about 9 inches.
The moment produced is:
25 lbs x 9 in. = 225 in-lbs.
The bike weighs 285 kg + avg. gas of 15 kg = 300 kg = 660 pounds. The rider (it’s I) 170 lbs, the total weight 830 pounds, without the system cases. I estimate that about 5/8-th of that load is on the rear wheel, therefore (5/8) x 830 = 518.75, say 520 lbs. without loading the rear wheel by the system cases. My system-cases are pretty heavy, say 40 lbs each, times two =80 lbs. Therefore the rear wheel carries 520 + 80 = 600 pounds as I estimated.
To establish the necessary offset of the rear wheel I have to divide the 225 in-lbs moment (which is caused by the lead) by the weight of 600 lbs. (the load falls onto the rear wheel). Therefore: 225 in-lbs. divided by 600 lbs. = 0.375 inches = 9.5 mm, about.
Since this 9.5 mm is the result of estimated values I decided to move the rear wheel towards the right hand side, hence closer to the final drive, by 8 mm only. By removing the factory installed 2 mm thick shim from in between the hub of the rim and the final drive, I will gain 2 mm, clearance therefore I only have to machine 6 mm off the hub of the rim. The result: the rear wheel will be closer to the final drive unit by a total of 8 mm, relative to its original position.
I had no problem with machining some of the hub off; the hub is thick enough and I also have a second set of rims.
From previous experiments I still have two pieces of 2 mm thick factory shims and a home made 6 mm shim too, in case I would want to revert to the previous setting of the rear wheel. In addition to that I made four pieces of 6 mm thick steel washers for the adjustment of the length of the wheel bolts.
Before going into a radical modification of machining off some of the original hub I had to make sure that the wheels were in parallel planes and I also wanted to have the means to see how these parallel planes would line up with each other.
A wheel Alignment Jig had to be built.
The simplest way to check “parallel” is in the vertical plane by using a 2 ft. spirit level.
First: I put the bike onto the lift and checked if the rear wheel was in the vertical plane. It was not. The centre stand had to be shimmed until the rear rim was in the vertical position.
Next, the front wheel had to be checked if it was in the vertical plane. This check is only accurate if the front wheel is parallel with the rear wheel.
Along the bike a reference line has to project the plane of the rear wheel right along the front wheel. This line of reference is provided by a 2” x 1” x 8 foot long aluminums tube which is clamped to the rear rim at two places and runs, just under the muffler, passing by the front wheel. These clamps of equal length hold the rectangular tube 200 mm away and parallel with the rear rim.
Next, the front wheel has to be set parallel to the aluminums tube, and by doing so, it will be parallel with the rear wheel too.
The tube passes the front rim at two places. At both places the distance between the rim and the tube has to be the same in order for the front rim to be parallel with the rear one.
The adjustment of the front rim into a parallel position with the tube was done by attaching soft tie-wires from the ends of the handle bar to the system case supports, on each side of the bike. The fine adjustment is done by small turn-buckles which are installed at mid length of the wires.
After the parallel position of the front rim is achieved the checking for tilting of the front rim can be accurate. The same spirit level used for setting the rear rim into the vertical plane was also used to do the same check on the front rim.
The front rim was also in a vertical plane. If the planes of front and rear rims are in the same vertical plane then both wheels are running in the same track. The vertical plane for reference was chosen for convenience of measurement.
Since the reference tube is at a distance of 200 mm away from the rear rim, the front rim has to be 213 mm from the tube for the wheels to run in the same track. The 200 mm is a distance that was chosen for convenience; it could have been any other length too. The 13 mm additional length, measured at the front rim, is half of the difference between the width of front and rear rims.
The rear rim is 136 mm wide and the front rim is 110 mm of width.
136 – 110 = 26 and 26/2=13.
I had 6 mm machined off the rear hub and I omitted the 2 mm factory shim and I installed the rear wheel. At this point the wheel moved to the right by 8 mm.
Tire and rim cleared the swing arm but spokes hit the brake calliper. By sanding I removed some material from one of the legs of the “U” shaped spokes. One of the legs of the “U” is longer than the other. Because of the interference between the calliper and the spokes, and the fact that I did not want to remove too much from the spokes, I had no choice but to replace the factory shim and by doing that the offset was reduced to 6 mm.
At this point I set up the reference line, the aluminums tube, again set the front wheel parallel with it and took measurements. The rear rim was 200 mm and the front rim was 203.5 mm from my reference line. These dimensions indicated that the centre of the rear wheel is 9.5 mm to the right from the centre of the front wheel.
To have common centres the measurement at the front rim should have been 213 mm.
213 – 203.5 = 9.5, however the rear wheel was offset by 6 mm only. The 9.5 mm actual offset is very visible from the back of the bike.
The final location of the rim now is 6 mm closer to the final drive than it was when the bike left the factory.
The bike still pulls slightly towards the right, so I have to remove the shim to obtain a full 8 mm offset.
My intention is to mill off some material from the rear side of the calliper. I do not destroy original parts so I ordered a used calliper from www.bmwboneyard.com with which I will experiment.
Modification to the second-hand calliper purchased for experimental reasons:
Presently the “2 mm shim” is installed so, to start with, I removed this shim, reinstalled the rim without the shim plate, removed the active rear calliper (did not disconnect it from the pressure hose, just let it hang), installed the second hand calliper. Marked this calliper where spokes were touching it. Removed the calliper.
The two halves of the calliper were separated,
From the half, which faces the rim, the piston was removed,
Wall thickness of the brake-cylinder was measured at the centre and at the edge of the cylinder’s diameter. At the centre the thickness is 4.7 mm, at the edges 2.3 mm.
The back of the cylinder has a “lens” shape. From the lens shape 2.2 mm thickness was machined off. Still, this did not provide enough clearance for the spokes to pass the caliper.
From the front half of the calliper, from its two mounting surfaces, 0.6 mm were removed. Therefore the whole calliper became offset from the centre of the brake-rotor by 0.6 mm.
This did not affect the pads, just changed the position of the pistons, a bit. Total clearance gained is 2.8 mm.
The two halves were reconnected and the modified calliper was installed and bled, and tested. The rear wheel was installed and there was no more interference between spokes and callipers.
As a final inspection, the wheel alignment jig was re-assembled and all previously taken measurements were taken again.
According to the latest measurements the rear wheel is offset by 7.5 mm from the front wheel towards the right hand side of the bike; however relative to the original position of the rear wheel, the offset was 8.0 mm.
On November 17, in +3 C. = 37 F. temperature, I test rode the bike for 34 km, at highway speed and small turns around street corners.
The test results are beyond expectation. The bike is not what it used to be. On the highway I could release the grips and it runs straight, still reacting to the crown of the road. Tilting of my head brought it back to the straight. At the sides of the road’s crown the bike followed the slope, a bit to the left or a bit to the right. Before it, always pulled to the right no matter on which side of the road I was riding. While riding now I constantly have the feeling that there is no need to hold the bar; it runs straight and it has to be forced out of straight running into turns, unlike before when the bike had to be forced to stay in straight running. It turns around corners very easily, it actually tries to turn more than I want. (I am still accustomed to the heavy turns).
Practically, this heavy bike has become nimble. At starting the lightness of the steering is immediately noticeable.
The test was done using new tires and without the system-cases. In case of a rider much heavier than I, especially with a heavy passenger or in a situation when riding with unevenly loaded side-boxes, the result might be different than my case.
The described measurements are not obtained by high-tech equipment. Due to the fact that rims with a lot of mileage on them are not perfect, that the reference line may not be 100% straight and some human error could be considered, I would say that a (+ or -) 0.5 to 1.0 mm in accuracy is possible.