View Full Version : Speedo Error

Rob Farmer
22-07-02, 14:04
I was browsing the micapeak bulletin board and found this posting:-

"I just installed a speedo hub from the R1100R. As noted previously,
it has a 2.875 ratio, compared to the 3.0 ratio stock on the Oilhead
GS. The changeover is quick, under 15 minutes including cleaning your
hands afterwards. Bottom line: it works just fine, and my speedometer
is now *much* more accurate.

Here are some data for 6th gear:

Tires: Metzler Tourance
38.5 rear, 35.5 front, cold pressures

With 3.0 speedo hub,
3000 RPM 65 mph indicated, 60.8 actual, 6.9% error
3500 RPM 75 mph indicated, 70.0 actual, 7.1% error
4000 RPM 85 mph indicated, 79.2 actual, 7.3% error

With 2.875 speedo hub,
3000 RPM 62 mph indicated, 61.2 actual, 1.3% error
3500 RPM 71 mph indicated, 70.0 actual, 1.4% error
4000 RPM 80 mph indicated, 79.0 actual, 1.3% error

I used a Magellan 315 GPS on flat straight sections of freeway at
steady throttle to allow the GPS time to settle. Obviously, my data
collection technique is imperfect, it's hard to read the analog
speedometer and the digital GPS at the same time while avoiding
trucks and SUVs. The GPS is only accurate at steady speeds on
straight roads.

The part number from BMW is 62 12 2 306 532. Maybe you'll find it for
less than I paid... but it's worth it for peace of mind.

Now I'm happy, a 1.4% error is much better than a 7% error. If I were
buying a new GS, I would demand that the dealer include the speedo
hub from the old Roadster in the package, out the door."

has anybody tried this mod?

Rob Farmer
95 R1100GS


City Slicker
22-07-02, 22:10

Read http://www.gsclubuk.net/forums/showthread.php?s=&threadid=979&highlight=speedo for the discussion on this a while back

I fitted the drive and it brings the speedo inaccuracy down to less than 5% and the odometer to about 2% comparing it to the GPS. That compares pretty favourably with the two cars I have tested with the GPS for accuracy.

It means that if you are on a highway doing 70 or whatever at least the car drivers aren't getting pissed off at you for not making progress.

It's an acceptable margin of error now and was worth the 20 odd that I paid for it from Motorworks. BTW, the Part No may be correct but the bike it is listed for is the R1150R, but in any case check for the 2.875:1 ratio.

The only problem with this mod is that the fabled 130 that other folks see on their speedo is in reality a more modest 120mph on the adjusted clock.

22-07-02, 23:54
City Slicker,

Do you know if the same applies to the R1100GS ?

I really need to know as I am being driven insane everytime I do an advanced ride I have to mentally adjust/compensate for the inaccuracy.

The most pronounced illustration is when I am doing 70mph on a Mway. The guy with a calibrated speedo and or GPS will swear I am doing 65mph. Given many people observe the same thing I am in no doubt they are correct.

What is worse is 30mph on the speedo is less by 3-4mph. And as you know I hate cars being close behind (ref the rear indicator mod)

Any hints/tips welcome.

Paul G (BHT)
23-07-02, 07:35
Yup, the GS 11's do have the same 6 - 7% error

23-07-02, 10:29
OK so this is not a lot of cash but I really am not too chuffed with Mr BMW. Just fit the damn thing in the first place instead of being all conservative on us poor customers.

The damage
Motorworks want 30 plus vat (new not in stock + don't have a used one )
Motobins want 25.50 plus vat

The real question is. Is the 2.875:1 the correct/best one ? Anyone thought of or tried the 2.6:1 ? I guess it may read high.

BTW : I have a '97 R1100GS and Metzeler E4s

City Slicker
23-07-02, 11:49
I think that there is a jump down to 2.5:1 from 2.875:1 so this would not be suitable. Working on my calculations the perfect ratio would be 2.7:1 but is not available, therefore the 2.85:1 ratio is probably better because the other one will under-read the true speed.

I have an 1150GS but have read posts by folks with 1100GS's with similar errors.

I discounted the tyres being an influence because in order to generate the kind of abberation we are looking at on the speedo there would have to be a significant change in the radius of the wheel; for example if there was a 10% error on a wheel with a circumference of 2000mm, and therefore a radius of 318mm, reducing the circumference by 10% would mean the new radius of the wheel would have to be 285mm - the difference would be 33mm, and there is no way that different tyres would vary by such a large amount.

It may be possible to have a variation in tyre tread from manufacturere to manufacturer, including the wear, of possibly as much as 15mm but this would be for something like an off-road tyre; most manufacturers will stick with a fairly standard radius from ground to axle to avoid problems with fit on bikes with close mudguards, braces etc, and most likely to ensure that the speedo will ready roughly the same.