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

Average power discrepancy between Garmin Connect and Strava

FloP.
6 months ago

Ever since swapping to the Magene Powermeter, I'm experiencing the issue that the avg. power between Strava and Garmin Edge 830/Connect is quite different from the values shown in Strava.

I saw in other threads that you've quite a lot of knowledge about this topic. I had a look at the fit files. It seems that the accumulated power(watts) that are saved in the fit file are causing this issue. As they are higher than what I have calculated for the accumulation.

Is there any solution for this issue?

  • 0 6 months ago

    Would you be willing to share an activity file?

  • 0 6 months ago in reply to aweatherall

    17277834946 (1).zipThat's the last one I made with the powermeter

  • 0 6 months ago in reply to FloP.

    There is definitely an issue between summing the 1 second power and using the reported accumulated power. This results in around 9% difference by the end of the ride.

    I would need to see what the power meter is actually sending. If you are prepared to collect some extra data let me know and I can give you instructions on how to do that. 

  • 0 6 months ago in reply to aweatherall

    Yeah, I'm curious about why the sum of all 1s power doesn't fit better to the accumulated power, as with another PM they're fitting way better, still not the same, but less than 1% difference

  • 0 6 months ago in reply to aweatherall

    Did a ride to day, have updated the software of the powermeter, problem didn't solve. Interestingly, in the interval where I didn't stop pedaling, the data matches and mostly the mistakes where power and accumulated power don't match is after short pauses of pedaling.

  • 0 6 months ago

    I think you may Zero Averaging turned off on the edge. The fit-file summation is fine until it comes to zero values.

  • 0 6 months ago in reply to mcbit

    No, unfortunately that's not the problem. 

    I had a look at the values and calculated it without zero, the average would be higher. And I also had a look at the settings in the garmin, avg power is calculated with zero 

  • 0 6 months ago in reply to FloP.

    This is where the summing starts to go wrong on row 28 of the data nad happens a few time after zero power/ cadence entries.

    Also you have quite a number of cadence entries from 120-180 rpm, not sure what you max out at.

  • 0 6 months ago in reply to mcbit

    Yeah I have seen this as well, I have sampled some data on the weekend and will take a look at the data later this evening. 

    I also realized that the problem only (more or less) occurs after zero cadence as in an interval in Friday where I pedaled all the time it was completly fine. Really curious where the error comes from

  • 0 6 months ago in reply to FloP.

    I suspect the power meter is somehow accumulating a torque value while coasting and dumping that into the first pedal stroke when you re-start. I had seen that most of the jumps on reported accumulated data occurred around a stop/start pedaling event. The problem I'm pretty sure is a power meter problem.