TL; DR: My ~5 year old Wahoo Kickr Core seems to report power that is ~5% higher than that measured by 3 different bike-mounted PMs (Assioma MX2 and 2 different Qarq spider PMs; all up to date firmware, have been calibrated/spindown performed on trainer). As I have been using the Kickr exclusively for indoor structured training, this probably means my AI-estimated FTP is 5% too high, and since that number is a direct measure of my value as a cyclist/human I’m in crisis!
OK I’m kidding about that, but this is annoying, if for no other reason than that it means outdoor rides will have incorrect IF and TSS values, OR if I manually correct my FTP indoor rides will feel too easy. I don’t think there’s a software solution here (e.g. direct ability to set a 5% offset for the Kickr in either Wahoo or TR software.) The bike I normally use on the trainer does not have a PM, so PowerMatch is not an ideal solution either.
Two questions:
Is this something that happens to trainers as they age? I checked this trainer once a few years ago with a different Quarq and it seemed accurate or slightly low, consistent with drivetrain losses.
Any suggestions to fix this, beyond PowerMatch? I can’t think of a solution that doesn’t yield an inflated indoor FTP.
I got my start in the cycling game on a Peloton. I did their 20min FTP test and ended up training off of a 335 FTP. Being 38yrs old and 68kg I thought I was hot $hit with just 24months of training.
Long story less long…bought a road bike and a Kickr, 20min test showed 261. Dagger.
You mentioned spindowns which is great, but have you tried the “Advanced / Factory Spindown”. That is not supposed to be done often, but may be worth a try here since it is a more comprehensive operation than the common version.
Not sure about the ageing part but my kickr core was off by about the same. its just the accuracy of their budget product…I recently got a kickr v5 which reads the same as my assiomas within a watt or two.
5% is quite a significant discrepancy so I would either scale up your bike power meters by 5%, sell your core and get a kickr v5, or I can see in intervals.icu they have a function for indoor/outdoor ftp (never used it).
As usual @mcneese.chad with the super helpful suggestion. Did the factory spin down, now seems very close to at least the one other PM I’ve compared to. Thanks!
Question for @mcneese.chad: one change I’ve made in recent months is to add a rocker system to the trainer. While I’m sure a strain gauge-based PM can measure just fine even with some motion, do we know if putting a trainer on a rocker plate accelerates bearing wear or requires more frequent calibration? Seems like that heavy flywheel would be fighting gyroscopic forces as it moves, which might wear things out or add some friction or something.
I have no data to back up my thoughts, but I feel the amount of motion seen in a typical rocker won’t impact the trainer negatively. The actual lean angle change is less than 5* per side even when standing and around half that for most seated riders.
The physical displacement in space is minor. I did a CAD sketch related to a similar question years ago and think the delta at the flywheel was less than 1" [25mm] max, likely closer to half that aligned with the angle info above. All that is from my shaky memory, so subject to correction
I just don’t see the amount of motion as a potential new source of wear or concern.
Totally agree. The bearings don’t care how they are oriented with respect to gravity. They only care about the angle of forces relative to the bearing, which isn’t changing since the whole trainer is moving together.
For what it’s worth, my Kickr Core on my rocker is matching my Assioma Duos within 1-2 W, just tested 2 days ago, despite both being several years old.
My thought would be that the heavy flywheel is a pretty good gyroscope when in use. Making a gyroscope wobble back and forth, even a little, will repeatedly put lateral loads on the bearing that they aren’t really designed for.
To be clear I’m 100% on board for rocker plates and plan to continue using mine. I’m just wondering if they might apply some additional strain to a trainer that would cause some degradation in accuracy and therefore more frequent calibration.
