How to: Calibrate your speedo after a sprocket change

[CHEMIKER]

A couple of weeks ago, I changed my front sprocket to 15T. I know this leads to speedo and odo error, and until I get a speedohealer, I wanted to know mentally what the speedo error was. Here's the setup I used:

First, tape a friend's GPS to your instrument panel. To get the tape behind the instruments, you have to remove your inner fairing (easy and takes 10 min). Be sure to use a friend's GPS! How crappy would you feel if your own GPS fell off and got ruined?

Next, place a video camera on a tank mount to video the whole instrument panel. Make sure the GPS, speedometer, and tachometer are all clearly visible in the field of view.

Go for a ride and video what the instruments are doing. Be sure to level your speed at approximately 10mph increments, and hold for several seconds. The instruments (GPS and speedometer) don't refresh instantly, and holding speeds allows the instruments to catch up.

Watch the video when you get home and input actual versus displayed speed in a spreadsheet. Plot actual speed versus displayed speed to get an idea of the error.

I will post my offset after I collect it tomorrow.

 

[MyFirstBike]

What Happened... This was back in June... I want to see the results... lol

 

[reiobard]

Be sure to use a friend's GPS! How crappy would you feel if your own GPS fell off and got ruined? QUOTE]

What Happened... This was back in June... I want to see the results... lol

His Friend's GPS fell off and got ruined

 

[FZ1inNH]

Duct tape! Velcro???

 

[CHEMIKER]

LOL thanks for the reminder. I did this back in June and forgot to post the results. This actually worked very well. The video turned out to be a great idea as I was able to compare GPS vs speedo readings at many different speeds and not have to worry about remembering all of them (or writing them down while on the highway ). I just plotted several speeds from 0mph to 100mph then did a linear fit in excel to get some key speeds. Here are the results (WITH a 15T sprocket, WITHOUT a speedohealer):

Speedo.....Actual

37...........35
48...........45
60...........55
71...........65
97...........88
100..........91

It turns out that below 25 there is very little error (1% at 20mph), so I focused on speeds above that.

 

[CHEMIKER]

His Friend's GPS fell off and got ruined

It actually survived the ride just fine, much to my friend's chagrin. Me killing it would have been a good excuse for him to replace it (it's long overdue).

 

[Wavex]

LOL thanks for the reminder. I did this back in June and forgot to post the results. This actually worked very well. The video turned out to be a great idea as I was able to compare GPS vs speedo readings at many different speeds and not have to worry about remembering all of them (or writing them down while on the highway ). I just plotted several speeds from 0mph to 100mph then did a linear fit in excel to get some key speeds. Here are the results (WITH a 15T sprocket, WITHOUT a speedohealer):

Speedo.....Actual

37...........35
48...........45
60...........55
71...........65
97...........88
100..........91

It turns out that below 25 there is very little error (1% at 20mph), so I focused on speeds above that.

That's some serious setup there Travis! I did the same thing but without the camera... glance at your speedo and GPS while maintaining the same speed, make the math in your head, and quickly realize that the speedo is about 10% fast across the range.

60mph on speedo? = 54mph real
70mph on speedo? = 63mph real
80 = 72
etc...

1% @ 20mph? you mean 0.2mph difference? I think the 10% is valid below 25mph too, and don't see why it wouldn't. Now of course, 10% of 20mph is only 2mph, which indeed is negligible due to margin of error.

:thumbup:

 

[CHEMIKER]

That's some serious setup there Travis! I did the same thing but without the camera... glance at your speedo and GPS while maintaining the same speed, make the math in your head, and quickly realize that the speedo is about 10% fast across the range.

60mph on speedo? = 54mph real
70mph on speedo? = 63mph real
80 = 72
etc...

1% @ 20mph? you mean 0.2mph difference? I think the 10% is valid below 25mph too, and don't see why it wouldn't. Now of course, 10% of 20mph is only 2mph, which indeed is negligible due to margin of error.

:thumbup:

It depends on whether or not you use the 0mph speedo = 0mph GPS data point. Including it and not including it both give a very linear fit to the data, but the equivalent values between about 25mph and 0mph vary depending on the case. Either way, any differences between speedo and actual speed should be ignored under 25mph because the values are very close. Besides, if a cop busts you for speeding in a 5mph or 25mph zone, my guess is that you were going more than 10% over the posted speed anyway.

 

[reiobard]

It actually survived the ride just fine, much to my friend's chagrin. Me killing it would have been a good excuse for him to replace it (it's long overdue).

I know.... I know..... I have the exact same GPS...needs replacing hardcore...

 

[SirIsaac]

LOL thanks for the reminder. I did this back in June and forgot to post the results. This actually worked very well. The video turned out to be a great idea as I was able to compare GPS vs speedo readings at many different speeds and not have to worry about remembering all of them (or writing them down while on the highway ). I just plotted several speeds from 0mph to 100mph then did a linear fit in excel to get some key speeds. Here are the results (WITH a 15T sprocket, WITHOUT a speedohealer):

Speedo.....Actual

37...........35
48...........45
60...........55
71...........65
97...........88
100..........91

It turns out that below 25 there is very little error (1% at 20mph), so I focused on speeds above that.

Based on the above data, the following table would apply if you still had the 16 tooth sprocket:

..............speedo...speedo % error
Actual....16 tooth...error 16 T
35...........35............0
45...........45............0
55...........56............1.8
65...........67............3.1
88...........91............3.4
91...........94............3.3

Maybe the stock set-up is not as inaccurate as many think it is (or want it to be. )

 

[CHEMIKER]

Based on the above data, the following table would apply if you still had the 16 tooth sprocket:

..............speedo...speedo % error
Actual....16 tooth...error 16 T
35...........35............0
45...........45............0
55...........56............1.8
65...........67............3.1
88...........91............3.4
91...........94............3.3

Maybe the stock set-up is not as inaccurate as many think it is (or want it to be. )

I don't think the stock setup is completely accurate. I didn't test it before I changed my sprocket, but I have heard others talk about the inherent inaccuracy of the stock speedo.

Anyone have a 16T sprocket, a GPS, a video camera, and some electrical tape???

 

[phil737]

As a side note, if you can't find a GPS unit, many cell phones have "speedometer" apps you can download. Some even give a free week trial period, which is enough to go out and verify your bike/car/boat speeds. It worked like a charm when I calibrated my Speedohealer.

 

[Doorag]

Can it be done on a rolling road? If so, then I would expect to be able to get the figures exactly.

 

[Shinleung]

out of curiosity, why does everyone use GPS as a reference?
As the satellite is so far away, I doubt how many % error is expected?

and also, as i remember, when i was slowing down, the GPS tend not to pick up the change as quickly as the speedo?
so i assume there is a lag?

 

[CHEMIKER]

out of curiosity, why does everyone use GPS as a reference?
As the satellite is so far away, I doubt how many % error is expected?

and also, as i remember, when i was slowing down, the GPS tend not to pick up the change as quickly as the speedo?
so i assume there is a lag?

Good point. I honestly don't know how accurate a GPS is. I'm assuming that it's pretty accurate as even though the satellites are far away, my GPS knows if I'm in my driveway or at my neighbor's house.

As far as slowing down or speeding up, yes, there is a non-instantaneous refresh rate with the GPS. There is with your speedo too for that matter. Everything digital will have to refresh at some interval. I held my speed constant for several seconds at several speeds in order to let the instruments come to equilibrium.

 

[Andz]

The standard setup isn't accurate at all, I had my bike on a dyno once and the dyno top speed was 249km/h while the speedo showed 271km/h