Completely different subject.. this calibration check, is it worth doing and how....?
If you want to compare the bkool power with your PM, I would start with a completely flat course (eliminates weight as a factor) and then ride at a fixed speed for a few minutes to compare power output (I use Golden Cheetah but Strava can also be used).
Do this at several different speeds, 10mph, 20mph, 25mph and perhaps 30mph.
Then one final test, find a steep and unrelenting slope (like Crowcombe or Salcombe) and ride up it at around FTP and compare the average power over the actual climb only, as the Bkool will probably struggle to give enough resistance and hence the PM will probably show lower figures. You can then estimate roughly how much gradient the bkool is actually managing to simulate, assuming your PM is good to within a few percentage (should be).
The bkool seems to vary from firmware version to firmware version, and on the pre-release at least, mine is pretty good at moderate powers (below about 200 Watts), but then seems to exaggerate power (or rather under applies resistance) by around 10% now (used to be more like 25%).
There is generally a few percent difference between any power meters, and of course it depends where it is being measured, as it will be higher at the crank than the rear wheel.
Unfortunately Bkool doesn't actually measure power at all, it estimates it from the resistance applied by the electro magnets and the speed you are rotating the flywheel, neither of which are particularly calibrated.
From comments they made when I reported the wide difference, they are relaxed about differences up to about + or - 10%.
I'd love to try these tests on a KICKR and see if the Bkool software and the KICKR resistance unit make use of the calibrated power meter in the KICKR.
Geoff
that's really disappointing.
