TheIronWarrior
Well-Known Member
- Joined
- Apr 14, 2013
- Messages
- 423
- Reaction score
- 328
- Points
- 63
- Location
- Halifax, NS, Canada
ABS sensors use tone rings to count (fractions of) revolutions of the wheel over time. The output from the ABS sensor is a measure of the wheel rotational speed.
Converting between rotational speed and linear speed is done through the radius, not diameter (v=rw). The difference in unloaded radius of the tires is 0.35". The effect of this radial difference on effective rolling radius is not known to me, but I think we can assume that the effective rolling radius would be impacted by approximately the same 0.35". In general, treadwear can reduce the unloaded radius by 0.25" or so (varies with tire).
I wonder what the dimensional tolerance of tires generally is. Things coming out of a mold are usually pretty consistent, but there will still be some amount of variance. How much larger or smaller than nominal does the tire have to be before the manufacturer considers it a reject?
I think the DCT uses front wheel speed (rotational, from ABS sensor) and the vehicle speed sensor (same as speedometer, from the transmission out) so a rear wheel change might not even matter to the DCT at all as far as shift RPMs. ABS would compare the two wheel speeds to determine a wheel slip, so out of the two systems if I had to guess, a rear wheel size change would affect the performance of the ABS system more than the DCT, but I also think you'd have to be way outside the realm of design values.
Overall, I think there are way too many variables that can affect the readings, especially when underway (accelerating in a turn may load up the tires differently, for example) and causes of "normal" wheel speed variances (elevation changes, for one) that Honda would have designed their ABS system to spazz out because of a less than 3% change in tire size.
If I was asked to come up with the ABS algorithm, I'd just look to see if during braking one wheel was below a certain low speed value (maybe 2km/h equivalent rotation, or even fully stopped) and the other was above another certain speed value (maybe 8km/h equivalent rotation) to trigger the ABS. The only case where that wouldn't work is if both wheels locked (but the bike can never tell two locked wheels from stopped anyway) or if you were slower than the higher speed value, where you're effectively stopped anyway. You'd just have to decide the slowest speed you want/need ABS to work, and the tolerance you'd want on the slower wheel to read as "locked". A wheel size mismatch would only cause "problems" around the two target speeds, and a small percent of a small number is generally insignificant. The system would still work, just might drop into the "inactive" slow speed range a little faster or slower than design.
Converting between rotational speed and linear speed is done through the radius, not diameter (v=rw). The difference in unloaded radius of the tires is 0.35". The effect of this radial difference on effective rolling radius is not known to me, but I think we can assume that the effective rolling radius would be impacted by approximately the same 0.35". In general, treadwear can reduce the unloaded radius by 0.25" or so (varies with tire).
I wonder what the dimensional tolerance of tires generally is. Things coming out of a mold are usually pretty consistent, but there will still be some amount of variance. How much larger or smaller than nominal does the tire have to be before the manufacturer considers it a reject?
I think the DCT uses front wheel speed (rotational, from ABS sensor) and the vehicle speed sensor (same as speedometer, from the transmission out) so a rear wheel change might not even matter to the DCT at all as far as shift RPMs. ABS would compare the two wheel speeds to determine a wheel slip, so out of the two systems if I had to guess, a rear wheel size change would affect the performance of the ABS system more than the DCT, but I also think you'd have to be way outside the realm of design values.
Overall, I think there are way too many variables that can affect the readings, especially when underway (accelerating in a turn may load up the tires differently, for example) and causes of "normal" wheel speed variances (elevation changes, for one) that Honda would have designed their ABS system to spazz out because of a less than 3% change in tire size.
If I was asked to come up with the ABS algorithm, I'd just look to see if during braking one wheel was below a certain low speed value (maybe 2km/h equivalent rotation, or even fully stopped) and the other was above another certain speed value (maybe 8km/h equivalent rotation) to trigger the ABS. The only case where that wouldn't work is if both wheels locked (but the bike can never tell two locked wheels from stopped anyway) or if you were slower than the higher speed value, where you're effectively stopped anyway. You'd just have to decide the slowest speed you want/need ABS to work, and the tolerance you'd want on the slower wheel to read as "locked". A wheel size mismatch would only cause "problems" around the two target speeds, and a small percent of a small number is generally insignificant. The system would still work, just might drop into the "inactive" slow speed range a little faster or slower than design.