I like to use a speed sensor rather than just relying on my head unit, to account for GPS drop-out around tall buildings, tree cover, tunnels etc. I recently switched from using a Bontrager DuoTrap S chain stay/magnet sensor to the new TRACKR Speed. For both sensors, I use the same custom circumference measurement which I’ve carefully measured (valve vertical on the ground and roll 3 full revolutions along a straight, level line with my weight on the bike and tyres at my riding pressure, measure with a metal rule and divide by 3).
With the DuoTrap, my recorded ride distances aligned with my route plan distances almost exactly - typically to within a few 10s of metres even over 100km+ rides. Since changing to the TRACKR Speed I have a consistent variance of +2% recorded distance over the route plan (past 4 rides have been +1.96% to +2.07%). I’m on the latest firmware on my Bolt V2 and TRACK Speed, and after contacting Wahoo they say the TRACKR Speed is ‘working as expected’ and agree that, as the variance is consistent, a workaround is to reduce my circumference setting to eliminate the variance.
Obviously that’s a practical solution but I’m still puzzled as to why a sensor should deliberately mis-record. Surely speed/distance is a super-simple metric to calculate. Has anyone else experienced similar? I wonder whether Wahoo may have deliberately set the sensor to overread in the expectation of riders under-measuring their wheel circumference by 2%, or of ANT+ drop-outs meaning 2% of wheel revolutions won’t get captured or something! Can’t think why else the sensor isn’t just counting the wheel revolutions and multiplying by the circumference setting, without bumping the number up 2%.