[lenshead]

I just had this part made for me! :grin2:

I've measured a speedo error of about 4% on my Avons with 4000 miles on them. This sensor wheel should lower speedo readings by exactly 4%.

My goal is to be accurate at any speed +/-1mph up to 100mph. I'll report back with test results.

I don't have ABS and my traction control is always off, but i don't anticipate any hiccups. I guess I'll find out!

 

[ukzx10r]

I just had this part made for me! :grin2:

I've measured a speedo error of about 4% on my Avons with 4000 miles on them. This sensor wheel should lower speedo readings by exactly 4%.

My goal is to be accurate at any speed +/-1mph up to 100mph. I'll report back with test results.

I don't have ABS and my traction control is always off, but i don't anticipate any hiccups. I guess I'll find out!

Is this for the rear wheel?
I noticed when i fitted a bsb rear slick that my speedo was reading lower than usual.
I wondered how this affected the traction because the front tyre is basically standard size so maybe it calculates each rotation and compensates for tyre size differences

 

[lenshead]

It is to replace the rear wheel sensor. Tire circumference changes between brands/models as well as over the life of any given tire. Even turning uses a part of the tire that has a smaller radius. The CPU does not need the wheels to turn at exactly the same speed to do traction control. It think it's based more on sudden changes in actual rear wheel speed that is not paralleled on the front, for example.

 

[lenshead]

OK!

So I currently have a speedo error of 0.092% !!

This was obtained by setting my speedo to KPH (a finer scale than MPH) and using a GPS app that logs peak speed to the first decimal. Three runs averaged.

Of course, this is coincidental, since, at the current state of tire wear, the uncorrected error fell right at about 4%, the exact correction factor of the new sensor.
This error will change as the tire continues to wear and again as it get's replaced. But I'm confident I can trust my speedometer now!

I have a second prototype, and I'll be seeking a volunteer to test it? Maybe someone who rides more aggressively than I do and can see if it affects ABS or TC?

 

[SkyDork]

I remember this topic from a long time ago. What I can't remember is if it was you who originally posted the idea of doing this or not lenshead. :dontknow: :hello:

Either way, this is the most viable solution to getting an accurate speed input. But questions do arise that I'm interested in. Nice job on the disk. My concern with that would apply to the TC and ABS systems. If they aren't matched up, the system can cause errors. At the very least it can be more restrictive when it senses a larger speed difference sooner than normal. That's irrelevant if you ride with it off all the time like both of us do, but anyone else copying this might see some weird behaviors.

The other thing is how are you getting that error with GPS? I know how, but I wanted to point out something about that testing on it. Using max speed readings is problematic. GPS speed is as accurate as radar....over steady-state speeds only. Accelerations do not allow enough time to record an accurate speed at all times. If you don't know, the GPS signal is basically only an atomic clock time signal. So the app receiving the time needs to do it quick enough to be able to see a difference in times between locations, which determines the speed of the receiver. There's usually a pretty decent delay in that when accelerating. So I'm just curious if you've done any steady-state testing with it to see if that accuracy is accurate. :lol:

 

[ukzx10r]

Has this got more holes or less holes?

 

[lenshead]

It has 48 holes down 2 from the stock parts 50. As far as measurements go they were taken at a steady speed maintained for about 3 seconds. Results were very reputable. I know the number sounds made up but that really is the result I got!

 

[lun471k]

Really curious about this (it's interesting) ! If you find someone who wants to try it out with ABS and KTRC I'd really like to see the results.

 

[lenshead]

Results were very reputable.

Repeatable! Dammit talk to text!