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

Accuracy issue (resloping?)

Hello,
Is there a way to reslope Garmin Rally? I've been using them for about 9 months (2 sided). LR balance is really bad. My partner tested them and got the same issue, so it is very unlikely to be my legs fault. I get about 58L-42R to 57L-43R on an average ride. She got 55L-45R. We both have dominant... right legs. Since these are in fact 2 independent power meters, it seems there is serious accuracy issues that would need resloping. Thanks!

  • L power phase is 243, R is 214

    L peak is 55, R is 62.

    Also did single leg drills as suggested on my elite direto xr. Right leg and direto were within 1W of each other (avg). Left leg was 8 watts higher on a 90W average (82W on the direto). As much as I would have liked the other way around, it seems its the left power meter that is overeading, by almost 10%. 

  • Any tip on how to deal with this issue would be appreciated. Thanks!

  • You dont need to buy weights, just hang a weight on both pedals, if one of the side is showing bigger/smaller number than the other then the 2 pedals are not even (so just a simple balance test without knowing the accurate weight of the used dumbbell or any other weight) and You can send a RMA request for the replacement.

  • ahh, i now read your single leg test. You can send the RMA email :) 

    or if you dont wanna lose training days without your powermeter, you can do a rescaling in the head unit (scale factor ~0.96 on left and test again, or in the Connect app in the Vectors setting), should be ok after it (i had old Vector1 in the past, i had to rescale them to 0.98%, both sides but i made it for a friend too , he had L0.98 and R0.94 if i remember good)

  • As much as I would have liked the other way around, it seems its the left power meter that is overeading, by almost 10%. 

    If you are in the middle of the cassette (no significant cross-chaining) with a clean but lubricated drive train you should expect about 3 to 5 percent loss. So it seems like the right is maybe a few watts low, and the left about five watts high, at 90 watts.

    In order to determine whether this was a scale factor or offset issue, you would have to test against the Direto at multiple power levels (preferably at a relatively cadence).  If it is an offset (zero calibration) issue then you would have a constant wattage difference relative to the Direto at 100, 200, 300W; if it is a scale factor problem then you'd expect to see, say, 10W error at 100, 20W error at 200W, etcetera.

    This also assumes that your Direto is holding its accuracy over time and temperature. Many trainers have a significant drift as they warm up, and I don't know how the Direto stacks up in that regard. While you can (relatively easily) check the torque calibration on the pedals, there is no easy way to validate the accuracy of a bike trainer.

    The 8 to 10 percent difference you're seeing, assuming the Direto is spot-on, is higher than expected for the pedals, but it is within the range that you can adjust the scale factor (that range is +/- 25%). If you feel that there is a problem with the pedals, based on your results here (or a hanging weight test, even with a filled one-gallon milk jug hung off the pedal toe loop, would establish the difference between the pedals as Luko79 suggests), then you need to contact Garmin Customer Service.

  • Thanks. I will do more testing this week-end, in a more controlled environment, at different power levels. The (relatively) good news is that, worst case scenario, the left pedal always overeads by 10%, if it is a scale factor problem. Assuming the right pedal doesn't underead, (if it does, it is a better scenario anyway in terms of total power inaccuracy because it would somewhat "cancel" the left side inaccuracy) the total power would be off by about 5% max. That explains why it took me so long to figure it out, as I always thought the discrepancy in total power to be due to drivetrain loss.

    Now, unfortunately, if the right side undereads and the left side overeads, the only way I can figure that out is with weight standards where I would know exactly the expected value. I do not have access to such standards.

    I do warm-up the Direto for 15-20 minutes before doing any such test. You're correct, there is significant drift.

    • the drivetrain loss is never so high, usually 2-3watts on small ring and middle rear @200w. the direto has no temp drift, that is accurate trainer. the left pedal measures too high, so simple (probably this direto has a small offset issue /a very few watts/ due 2 old calibration or loosened belt but thats all)
  • one of my friend "made" a known weight with kitchen scale (probably it had 3kg limit and 1gr accuracy), and he measured everything piece by piece, and he had a 15kgs "known weight" by free Smile

    it is the simplest and most accurate method to valide the pedal's slope value. (of course, You can validate it with an indoor trainer too /it is the dynamic torque calibration/  but there should calculate a tiny drivetrain loss and somehow You have 2 validate the trainer before this test...)

    jpansy.at/.../