View Full Version : K1200RS speedometer error....

02-08-2004, 06:12 PM
Have seen a number of references in ON and on the web regarding speedometer error on various BMW motorcycles. I own a 2002 K1200RS and believe mine may be off judging by some pretty non-scientific, seat o' the pants estimations and observations. Any feedback on this would be welcomed as would suggestions on how to remedy the problem.


02-08-2004, 07:19 PM
I've read and heard that up to 10% error is common, and regularly see advertisements in the ON for speedometer calibration services.

Calibration notwithstanding, perhaps the surest and quickest way to pin down a take on your bike is to borrow (if you do not already own) a GPS that will give you your actual speed. Get out on the big road and make notes on your margin at 40 -- 50 -- 60 -- 70 mph, or whatever, etc., until you are satisfied.

The GPS does not have to be a pricey unit like the highend models. A friend locally has an inexpensive handheld unit that he bought years ago when hiking was his passion. Recently, he hung it on the handlebar of his K100 by the wrist strap to get a read on his actual speed and speedo error.

Just food for thought. And btw, welcome to the boards.


02-09-2004, 05:14 PM
I wanted a more accurate speedometer and odometer (also a back-up) for my 1988 K100LT. I read various accounts on-line of installing a Sigma bicycle computer on a motorcycle, so I did that. The bike computer can be calibrated exactly to match your tire circumferance and is very accurate. It reads up to 180 plus MPH, includes odometer, stopwatch, trip odometer, miles per day, average speed, highest speed, etc.

The cost of the computer is about $25-$30. I bought some tiny rare earth magnets from Radio Shack for a few bucks and epoxied them to the front wheel for the sensor trigger, as the unit comes with a spoke mounted magnet for the trigger.
Installation tips can be found in the IBMWR tech pages as well as other sites (do a Google search).

I am very satisfied with the unit. Installation and calibration took about an hour if that. I mounted the readout directly on the handlebar, but you can mount it elsewhere if desired.

Calibration instructions are included in about 8 languages with several diagrams but still are a bit vague.

02-10-2004, 12:51 AM
Thanks Rick and Tom. I think I can lay hands on a GPS to run a check on the speedometer. While the Sigma computer sounds interesting and effective, I'm not sure I want to do an add-on if I can find a fix for the stock speedo. For the price of admission, one would think BMW could come up with a speedometer that's accurate. I say this with all due respect for the finer qualities of my KRS (but while still recovering from the sticker shock of the very pricey 12K maintenance.


02-14-2004, 12:14 AM
I have done the GPS (SPIII) test with my 2002 K12RS. I set the cruise control to hold steady on a flat straight road. Straight is theoretically important because the GPS takes fixes at a max of once per second - so if you round a curve, the track will not follow the curve but will make a series of short straight lines - or cords across the part circles. The sum of the cords will be less than the actual distance around the curve. It's a small error, buy hey ... Flat is more impotant since the cruise control holds a more steady speed when the road is flat. You need to watch the GPS for awhile to be sure the measured speed is steady.

This data is from memory (not good lab technique) but is not far off.

100 indicated 98 GPS
80 indicated 78 GPS
60 indicated 59 GPS

In my book, that's damn good. With old tires the difference increases slightly. As always, YMMV or all Beemers may not be born equal.

02-14-2004, 12:38 AM
Wow. That sounds pretty close and I hope mine is in the same range. I have to admit it has seemed to be further off, but again I'm basing this on how it relates to my normal pace and how it compares to the traffic I'm traveling in. Will definitely try to get may hands on a GPS. Temps in the low 50's tomorrow....maybe a good time to track one down!

02-16-2004, 08:31 PM
Public and commercial non-DGPS units do not use position location algorithms to determine the speed estimate reported by the unit. Instead fancy math based on doppler shift algorithms is used. So curved roads or straight and up or down or flat is not an issue when comparing your speedometer to a GPS to whatever your real speed is.

The 1 second sampling rate and the processor capability of the unit in use are the two bigger factors in the accuracy of the speed reading. Generally, GPS speed lags reality by somewhere in the range of 2-5 seconds.

The technique of riding straight, flat, and steady to stabilize the readings is the best way to use an imperfect GPS speed to compare to an imperfect speedometer. You will still not know your real speed. You will know the difference between the two devices.

02-16-2004, 11:32 PM
Hi Camfarm,

No doubt a GPS cannot provide a fully accurate speed but what is the degree of error expected? 10% 5% 1%? My WAG is that it is closer to the latter number.

Since all measuring devices of all types have some error, knowing error exists is not helpful unless one has an idea of the size of that error.


02-17-2004, 02:38 AM
But aren't we splitting hairs just a bit?

In my state, the maximum speed allowed on interstates is 70 mph.

If I am showing 85 mph on the speedometer, and top a hill to find myself caught by radar, three percent or five percent below or above 85 is mostly irrelevant -- I have a ticket either way.

Maybe the guy who commented he would spend his money on a radar detector first had the right idea.

Just food for thought ...

02-17-2004, 07:26 AM
Actually, I was tempted to edit my post to start: "As Allen stated..." since I do agree with your assessment and merely added my own [mis]understanding of the electronics.

In a series of observations over the past several weeks, my feeling is the same as yours. GPS speed is a more accurate and precise method of knowing what's what. The residual error for GPS is probably in the 'so what' range. My speedo is showing the same 5% to 8% error reported by others for the K1200RS. My Ford truck, likewise, when running winter tires versus stock tires.

Saw a reply from an auto engineer somewhere, addressing the speedo error. EU regs for manufacturers permit up to a 10% error on the high side but 0% error on the low side for road going speedometers. If I were calibrating a speedo at the factory, I'd feel safer doing what appears to be done by BMW.