2014 Stelvio Speedometer Accuracy

[Cheifhbw]

I've seen posts from over the past years that noted that Moto Guzzi speedos are not overly accurate. Has there been found a way to increase the accuracy? My 2014 Stelvio, even though I love it, is almost 10% off. At least according to my GPS. My Peterbilt is a whole lot closer.{Only 1-2% on the downside} On new cars you can have the dealer get in the computer{ECM}and make sure the correct tire size is installed.

[ChuckH]

My '12 is closer than yours.  It is 1-2 MPH off up to about 60MPH, 2-3 MPH off up to about 75 MPH and about 4 MPH off about that.  The bike's speedometer is optimistic, meaning when my GPS (Garmin 665) is reading 70 MPH, the bike is reading 73 MPH.

I am currently running Shinko 705 tires but have had about the same differentials with Michelin Anakee 3's and the OEM Pirelli's.

[beetle]

Speedo correction is part of the map. It can be modified.

[Spuddy]

Speedo correction is part of the map. It can be modified.

Obviously by the dealer, yes?

Spud

[beetle]

No. You have to edit the map itself. Anyone can do it with the GuzziDiag tools and TunerPro.

[Vasco DG]

Yup, PADS has no such generous provision....

Pete

[Kiwi_Roy]

I would expect a modern speedo to be digital and dead accurate, is the error just due to tire selection, just curious.

[beetle]

I would expect a modern speedo to be digital and dead accurate, is the error just due to tire selection, just curious.

Tee hee!

They newer bikes all use the same sensor, but the ABS models have dual sensors and use different correction factors. The 1200 Sport has a front wheel sensor that is usually out by about 10%. The Griso sensor is about 5% out.

No. Not accurate.

[pyoungbl]

There is a, what I consider to be complicated, speedometer accuracy formula for EU certification of vehicles.  Basically, the speedo has to be within a % or two for low speeds and a greater variation at higher speeds and must read greater than the actual speed.  I suspect this is to make darn sure you are going below the posted speed limit (of course you are intending to do that, right?), while still accounting for tire wear.  As the tires wear the wheel will spin faster for a given speed over the ground and thus a reading that is 100% accurate for a new tire will be reading faster as the tire wears.  This is much like telling what time it is...close is OK.

Now, before we go off on a tangent about staying within some magical allowance before you get a speeding ticket, I know of at least two jurisdictions where you get ticketed for as little as 2 mph over the limit.  Be very careful when driving through Emporia, VA on either I 95 or US 58; also just west of Emporia on US 58.


Peter Y.

[rocker59]

I would expect a modern speedo to be digital and dead accurate, is the error just due to tire selection, just curious.

The European OEMs build them with optimistic readouts on purpose.

[Kiwi_Roy]

Tee hee!

They newer bikes all use the same sensor, but the ABS models have dual sensors and use different correction factors. The 1200 Sport has a front wheel sensor that is usually out by about 10%. The Griso sensor is about 5% out.

No. Not accurate.

The old Veglias are just a magnet spinning eddy currents into an aluminum disk creating drag opposed by a hairspring.
The modern speedos are pulse driven, the microprocessor counts pulses per time unit and drives a stepper motor so there is really no excuse for them to be in-accurate. They should be at least as accurate as an old Smiths Chronometric based on clockworks.
My trade is instrumentation, I learn how to calibrate instruments to better than 0.1%, it seems to me that a simple task like this should be very accurate more so if the pulses are coming off the front wheel not subject to slip from power applied.
The biggest variable should be tire circumference.
I assume they are perfectly linear i.e. same error at any speed, at 60 MPH the pulse rate should be double that of 30 MPH.

Is it just a matter of changing one parameter to correct them?

[rocker59]

The old Veglias are just a magnet spinning eddy currents into an aluminum disk creating drag opposed by a hairspring.
The modern speedos are pulse driven, the microprocessor counts pulses per time unit and drives a stepper motor so there is really no excuse for them to be in-accurate. They should be at least as accurate as an old Smiths Chronometric based on clockworks.
My trade is instrumentation, I learn how to calibrate instruments to better than 0.1%, it seems to me that a simple task like this should be very accurate more so if the pulses are coming off the front wheel not subject to slip from power applied.
I assume they are perfectly linear, is it just a matter of changing one parameter?

Except if legislation and penalties in EU mean it's a safer bet to program the speedos to be slightly optimistic.

[Kiwi_Roy]

Except if legislation and penalties in EU mean it's a safer bet to program the speedos to be slightly optimistic.

I can see that they might be optimistic to make you think the bike is faster than it really is but does that mean they over read miles as well?
The two go hand in hand.

I'm not trying to be difficult, I just think it's an interesting topic.

[rocker59]

I can see that they might be optimistic to make you think the bike is faster than it really is but does that mean they over read miles as well?
The two go hand in hand.

I'm not trying to be difficult, I just think it's an interesting topic.

Odometer and Speedometer are two different circuits.  Even on mechanical units.

For example, I tested my ITI and though it was optimistic with speed, the odometer was correct.

When disassembling a broken unit, it was interesting how two different systems were being driven off the input from the speedo cable.

I took the opportunity to move my needle a little bit on its shaft so the speed would be closer.  It then became a little pessimistic at city streets speeds, and much closer to correct at highway speeds.

[Kiwi_Roy]

Odometer and Speedometer are two different circuits.  Even on mechanical units.

They both use the same revs per mile input, I think it's 1,000 per mile.
So for every 100 revolutions of the cable the Odometer moves 0.1
While the old Veglias just rely on eddy current drag the new Digital ones count how many pulses arrive and scale them for miles and time them for speed.

For example if we were getting 1000 pulses per mile it would count to 100 and increment the Oddo by 0.1
At 50 MPH you should be getting 50,000 pulses per hour 13.88 per second so wind the stepper up 14 steps which should co-respond to 50

In reality we are getting a lot more than 1000 pulses per mile and the stepper motor is divided into hundreds of steps It probably uses a longer time period therefore the resolution should be much better. Speedos with a digital display rather than the stepper motor have even less reason to be inaccurate.

The instruments employ a microprocessor, that's what they do best.

[rocker59]

I don't doubt that the instruments have the ability to be more accurate.  It's just my understanding that speed is programmed to be optimistic because of EU laws/regulations

[pauldaytona]

In europe they have to be reading  until 10% more. Partly to compensate for different tyres and tyre state, And to compensate that they were itself also not always accurate, so you can never tell the cop that it was only 98 when you were doing 103.

My Volvo car from 2002 has the same positive difference.

I don't mind it's to much, it's good to know what margin you have while driving.

[beetle]

The modern speedos are pulse driven, the microprocessor counts pulses per time unit and drives a stepper motor so there is really no excuse for them to be in-accurate. 60 MPH the pulse rate should be double that of 30 MPH.

The pulses from the sensors go to the ECU, which applies the correction factor. The signal is then set to the dash, which has its own microprocessor. It's unknown whether the dash simply dumbly drives the speedo via the ECU signal or does its own crunching as well.

Is it just a matter of changing one parameter to correct them?

Yes.

[beetle]

Odometer and Speedometer are two different circuits.  Even on mechanical units.

In the case of the 5AM and later models, the dash does the mileage calculation. I wonder whether mileage calc is affected by the change in speedo correction factor? Obviously, they don't care if the speedo is optimistic, but the mileage has to be accurate.

[Cheifhbw]

/Thanks Riders, The general consensus seems to be they built it that way but with the right tools it can be changed. Maybe somebody needs to start a sideline adjusting MG's speedos. At least it's nice to know that the mileage is accurate. Thanks again ya'll.     Chief head bottle washer    Cheifhbw

[pauldaytona]

I see no other way then that adjusting the speedo, will change milage too. The multiplier that we use, will work for both. Since it is made to adjust wheel/pulse differences.

[Kiwi_Roy]

It's interesting how Speedhut do their speedometers, they accept anything between 1k and 250k pulses per mile but you don't have to think about it. To calibrate them all you do is press Start, drive 2 miles and press End, it's done.
I used a runners GPS accurate to about 3 ft when I set mine.