This discussion has been locked.

You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Power calibration 820 vs 935?

I have 820 and 935 connected with Powertap P1 pedals through ANT. Why is the successful calibration number different between 820 and 935 -> I first calibrate with 820 getting usually number between 12 and 14, and thereafter with 935 getting usually number 4. No movement in pedals in-between, time difference between calibrations few seconds + inside garage where environment is pretty steady.
There is also a difference in average watts and NP for the same ride, today 4 watts between the devices (but same diff in both avg power and NP) for 45min ride on a trainer -> is this the consequence of different calibration number? Funny as they should be reading same data transfer from the pedals....
I started and stopped recording manually on both devices, but really with maybe 1s difference (steady pedalling so no powerspikes).
Is there a way to say which is device is more correct and/or should I just calibrate with one device?
Kai

0 aweatherall over 7 years ago

The offset is applied at the pedals so both units should be getting the same data, though this could be impact radio reception.

Could you post both files?

You only need to do the one calibration. What happens if you do the 935 first. It could be that the difference is due to doing back to back calibrations.

0 KBCap over 7 years ago

I already deleted 820 ride as not to get duplicates in GC/TrainingPeaks. I'll ride tomorrow again with 935 calibration only and let's see if wattage is aligned. If not, I'll post both files.
I still would have assumed that calibration software piece is the same for both 820 and 935 and the returned value (offset) would be the same (both running latest firmware checked as per few days ago). Could it be linked to "observed" temperature as 935 is on my wrist and thus generally shows too high temperature (mixing body heat and environment)?

Kai

0 aweatherall over 7 years ago

One thing to check is that the crank length is set to the same value on both units. The offset value being returned is being reported in Nm. (Torque)

The only information from the head unit that is provided to the power meter that would affect the calibration process is crank length.

0 KBCap over 7 years ago

Took a while longer than expected....
Anyway, checked that crank length is correct in all three instances (820, 935 both Ant and BT variants). Noticed though that apparently last time 935 was connected over BT to the pedals (assuming that it always tries first the power meter it was previously connected with), but now changed it to ANT connection.
And now only calibrated once with 935 and actually the value returned was 10 (when previously around 14 for 820 and 4 for 935).
Results: excellent, now power data matches between the two devices (avg, max, NP)! problem solved...

But two differing values surfaced IF and TSS: 820 reported TSS 35.9 and 935 41.8. Quite a difference, although both now seem to have the same power (and heart rate) data.
They are pulling the same FTP etc. data from GC, so no difference there, are they somehow having different mathematics?
I didn't think that training history as such would impact that calculation? In this case 935 knows my different runs and one spinning from the previous days which 820 is unaware of.

0 aweatherall over 7 years ago

Could you post the FIT files from both units and I will look at the IF and NP values.

The things that will impact those is the FTP value and whether auto-pause is enable or disabled.

0 KBCap over 7 years ago

Ok, it appears that 820 had different FTP number anyway. I don't know where it was coming from, presumably some kind of auto update number based on actual rides (slightly higher number than I have used based on 20min tests). Funny that it hadn't got either the old or new FTP number I had been using in 935 during past two months (and recently updated), and which were visible in GC.
In any case, everything is working now, so thanks for all the advice.