Early K Speedo Error Question

[98lee]

I have a question for anyone who KNOWS. (I can assume or guess all by myself thank you)


Is there a difference in the accuracy of the odometer vs the speedometer?

Now on most mechanically driven speedometers (like most older cars) any error in the odometer also shows up in the speedometer because they are mechanically linked.

But on our bikes, the electronic signal from the rear sensor gets split inside the speedometer. The odometer signal goes through one electronic circuit and the speedometer signal goes through a different electronic circuit. The electronic circuits are the source of "the standard BMW speedometer error".

The reason I ask is that in the process of evaluating different rear tires, I noticed the large difference in rolling circumference between different brands in 130/90-17.

I'm not all that concerned about speedometer error but I'm very concerned about odometer error. I ride a lot of poker runs at rallies and a lot of tours with turn by turn route sheets that list mileage for the turns.

What dawned on me was, in all these years of following route sheets that were prepared with Garmin Zumos, my odometer was usually only off by a few tenths after 100miles. VERY accurate!!!

What's the problem? Why don't I think my speedometer is correct also? Well it has to do with the Range of Light Gypsy Tour last Labor Day when I was chasing a Honda Turbo (650?) across the Nevada high dessert (5400ft elev.) with full bags and a top box at 122MPH indicated for over five minutes!!!!! I still had about 800rpm to redline but because of all the aero drag from the luggage and the altitude, it wouldn't pull it.

If the speedometer is also correct, that would mean that this is EASILY a 130MPH bike which most people agree is overly optimistic.

So, for those of you that KNOW, what is it? If the odometer is accurate, is the speedometer also accurate? Or in BMW's infinite wisdom, knowing how riders like to go "just a little over the speed limit" did they make the odometer accurate and the build in a fudge factor into the speedometer circuit.


I know, I could hook up with a buddy with a GPS and do some test runs (which I will when I get a chance but probably not for several weeks as life keeps getting in the way) but I'm curious NOW.


Anyone?


Bueller?



Bueller?

 

[rmarkr]

I know, I could hook up with a buddy with a GPS and do some test runs (which I will when I get a chance but probably not for several weeks as life keeps getting in the way) but I'm curious NOW.

Put one (any hand held) in the top of your tank bag!

FWIW My speedo is about 5 - 7 mph optimistic, the odo is spot on, with stock rear tire

 

[98lee]

FWIW My speedo is about 5 - 7 mph optimistic, the odo is spot on, with stock rear tire

That's kind of what I expected. That makes using the correct diameter rear tire very critical for odometer accuracy.

Thanks

For all you BT45 users out there: The new 130/90-17 BT45V rear that I just received is 3 1/4 INCHES less in circumference than the new 130/90-17 V rated Lasertec that I just received even though Bridgestones spec sheet says they are EXACTLY the same diameter---NOT. The Bridgestone is one full inch shorter than the Lasertec (and Bridgestones own specs).

 

[breyfogle]

German (maybe EU) new vehicle laws require that the speedometer NOT read low (i.e. the vehicle traveling faster than indicated) so BMW very carefully designs their speedometers to read comfortably high. It also helps with the advertising. BMW speedo's can be as much as 10% high. Odometers, on the otherhand, are required to be reasonably accurate and they are.

 

[98lee]

Thanks Brey.

Again, confirms my suspicions.

 

[58058D]

Well, hold on a minute on the Odo accuracy.....Not sure what the requirements are, and all my K Odos have been pretty darned accurate when out on the highway and coming across the test sections and now days with all the Post Mile (or Mile Post depending on your state's preference) markers. But several years ago when setting up the poker run for the Beemer Bash being run out of Plymouth, we had many divergent 'opinions' on how far we had gone....

CJ's airhead GS was the most generous, an 1100 GS (newest at the time) was only half as generous, and the two K bikes (my 100 RS and a K75 RT as I recall), were pretty close. We wrote down the running Odometer readings for each bike at each turn, and also the trip meter, then zeroed the trip meters. CJ's was as much as 10% higher than the Ks as I recall, and there was a Ducati in there, but I don't recall where it came in. As a result, we decided to list both running mileage and point to point mileage along with a disclaimer citing the above notes to help people along the way....To our knowledge, nobody got lost that year due to the routing, I think it was the stop at the winery and cigar shop that got them.... In retrospect, the Odo variance could just as easily been the tire circumference issue here as well, well worn knobbies on CJ's would result in more revolutions per mile recorded of course.

 

[BCKRider]

Sorry to weigh in late

58058 makes some interesting observations, especially about K bike odometer accuracy. The two I have owned (86K75 and 92K100) both seem to have absolutely accurate odometers. Haven't had the K100 down to the states yet to check it with the mile posts, but have measured the distance in miles of well-known bicycle routes here in Canada, multiplied the miles by 1.6 (it is a US bike) and it is spot on. Both older and newer air and oilheads frequently register more kilometers when I've toured with them. Kind of weird.

Checking speedometer error is more difficult - at least without a GPS. When I tried to it with my K75 on flat roads in Washington with little traffic and able to check my watch second hand every time I crossed a mileage marker trying to ride a steady 60 mph, I was never more than a second out of the one minute per mile. I think the K100 has a much more "optimistic" speedometer. Just not sure yet how optimistic.

And that sucks. I have owned several Japanese autos where BOTH the odometer and speedomer are right on the money. If I chose to speed, I know exactly how much I am over the limit. Same should be true for bikes with new tires. (As your tires wear, the circumfirance gradually decreases, and your indicated speed will be only slightly greater than real speed.)

Interesting observation from another poster that some tires are even smaller in circumferance, which would result in both higher registered miles and speed.

So, has anyone figured out a way to fix a high-reading odometer or a high-reading speedo when the odometer is right? Please share! I can't imagine that anyone prefers inaccurate readings.

The moderator of this forum may wish to move it, assuming there are some good solutions, because it most definitely is not just a K bike problem.

 

[pglaves]

I don't think the odo accuracy and the speedo accuracy relate in any meaningfully interpretable way. That opinion is based on the ownership of 7 K75s, a K100, and a K1100 and a total of about 500,000 miles combined on the bikes. And I generally run two GPS units on my bike. Voni runs one on hers.

The speedos typically read a little high, and the error is not linear. Most of the K75s read between 4 and 5 mph high between a real speed of about 25 up to 75 or 80. And that error increases slightly as the tire tread wears from new to toast.

Most of the K75 odometers read slightly low - 2% to 4% typically. That error tends to disappear as the tire tread wears.

Interestingly enough - other than the variation I find due to tread wear (and thus smaller diameter and circumference) both the odometer and speedometer on my '95 K75RTP former police bike seem to be very accurate when the tire is new. Funny how they can build precision when they want to, huh??

 

[bmw_rider]

how do you know the GPS figure is accurate?

 

[deilenberger]

how do you know the GPS figure is accurate?

Since it uses triangulation to locate you - and it does this on a regular basis - it has to be accurate (I believe Garmin quotes in single-digit percentages or less for error numbers.) If it wasn't - it would have no idea where you are.

 

[pglaves]

how do you know the GPS figure is accurate?

Well, when you run three GPSs all day, day after day, (for about 25K miles so far this year) and at the end of the day the moving average speeds and the accumulated miles match you get a pretty good indication that they are reasonably accurate. And on my bike, running two side by side, it is pretty easy to see if they vary in either speed or accumulated miles - and they almost always don't. Occasionally, in cities or in tall trees my old Garmin III+ will lose signal for an instant (less sensitive receiver in the older units) and the speed freezes for a moment, but other than that they work together virtually all the time.

I certainly, absolutely, trust my GPS data more than I trust the speedo or odo on any of my motorcycles - including the cop bike.

 

[cjack]

The early K bikes took pulses from a ring around the crown gear and after amplified in the instrument pod (on that little rectangular strip) the pulses went to a chip in the speedo/odo head. The chip divided the pulses by a suitable factor and stepped a motor which turned the odometers. Very repeatable and accurate depending on the rear wheel diameter.
As for the speedometer, the chip supplied standard pulse width pulses to a integrating circuit which resulted in an average dc voltage proportional to the pulse rate and this was what deflected the needle on the speedometer. Also this circuit has a potientiometer on the circuit board which can vary the speed readout for calibration, however this circuit and meter is not particularly linear.
So the two are not related at all...except for the source of the pulses on the bevel gear.

 

[pglaves]

The early K bikes took pulses from a ring around the crown gear and after amplified in the instrument pod (on that little rectangular strip) the pulses went to a chip in the speedo/odo head. The chip divided the pulses by a suitable factor and stepped a motor which turned the odometers. Very repeatable and accurate depending on the rear wheel diameter.
As for the speedometer, the chip supplied standard pulse width pulses to a integrating circuit which resulted in an average dc voltage proportional to the pulse rate and this was what deflected the needle on the speedometer. Also this circuit has a potientiometer on the circuit board which can vary the speed readout for calibration, however this circuit and meter is not particularly linear.
So the two are not related at all...except for the source of the pulses on the bevel gear.

What he said! I knew that - just not why!!

 

[BUDDINGGEEZER]

My 87 K100LT's odometer was almost dead on with GPS, of course the speedometer was fast like everyone elses'. I think it may be a motorcycle thing. I have a 91 Honda ST1100 and the speedo is around 8-10% fast just as the K bike was. the odometer is almost dead on with GPS. The ST1100 has a mechanical speedo drive on the front wheel???

Ralph Sims