Accuracy issue (resloping?)

If you are having suspected accuracy issues, then first ensure that the pedals are tight in the cranks. The recommended torque range is 25 to 30 foot-pounds (34 to 40 N-m). Secondly, it helps to do your manual zero calibration(s) when the system is not in the middle of a temperature transient - if the pedals and bike have been in a consistent temperature environment for 15 minutes then you should be fine.

If you think that your pedals are not measuring accurately, you can perform the hanging weight torque test as described in this link:

https://support.garmin.com/en-US/?faq=AVZeQisZvi3oBYZUUZdPV6

Then you can program corresponding scale factors into the pedals using an Edge, or through Garmin Connect Mobile on your phone.

Good morning,

Thank you for your quick answer. To be honest, I didn't expect to have to buy weights and do this kind of manual calibration on pedals that are this expensive. How come sloping is not done in the factory?

Thank you

The pedals are calibrated and tested to 1% slope accuracy at the factory over the full spindle rotation.

Most of the time, power measurement errors on Rally pedals end up being due to incorrect zero offset, which is why I suggested to look at that side of things as well.

Thanks for the precision. These pedals have been put on 3 different bikes over the last 9 months, and this accuracy issue has always been present. At what point would it be safe to assume that any incorrect zero offset at a given point in time cannot explain a constant discrepancy between the 2 power meters? The discrepancy is almost 20% between the 2 here.

Basically, I think Garmin should offer to recalibrate pedals free of charge when clearly something is odd. I like the product, it has been reliable so far, I just think a mistake was probably done during factory calibration.

There are a couple of different possibilities here, that you can investigate without going out to get weights.

You mentioned that you and your partner are both right leg dominant. I have no reason to doubt that, but it may be that while you think you can push harder with a given leg, you also have more trouble getting it out of the way on the upstroke.

Have you looked at the power phase numbers (or graphs) on Garmin Connect or on the Edge itself? If the amount of crank arc where your torque is positive for the right leg is substantially smaller with the right than the left, then this indicates either a zero offset problem or a physiological issue.

You can also check to some extent what your functional ability is with each leg, by enabling TE/PS (torque effectiveness/pedal smoothness) from your Edge, and adding a datafield to display torque effectiveness. The "torque effectiveness" is a ratio of the average torque in a crank cycle, to the positive torque in that crank cycle. When the torque effectiveness is lower, it implies that the negative torque on the upstroke is high and is effectively negating the work being done by the other leg.

If you (or a friend) have a stationary trainer that measures power then you could do single-legged drills, bearing in mind that there will be some loss through the drivetrain. If you have a stationary trainer that doesn't measure power, but has pretty high inertia and a speed sensor on the rear wheel, then you can do a single legged test where you ride at a steady speed. This will not give you super high precision but you should be able to see if the discrepancy is as large as what your records suggest.

You should contact Garmin Customer Service in your region to arrange anything that involves shipping the pedals.

Thanks. I haven`t looked at torque effectiveness, will do. Just so that I understand what I'm looking for, would I expect a higher or lower torque effectiveness for the leg with the smaller balance number (right leg)?

My understanding is a significantly lower torque effectiveness for a given leg would mean I'm pushing on the pedal against its movement (when its going up) harder with one leg than the other. Therefore, would the leg with the lower torque effectiveness be expected to have a higher power balance value as more force is theoretically applied on the pedal, even if some of this force doesn`t propel the bike (all else being equal, meaning I assume the positive force is similar between the two)? Or would it be the other way around?

In your case I'd expect to see lower torque effectiveness from the right leg. 

On a 20 minutes ride, torque effectiveness was 86% L/ 78% R, with a 58% L balance and average watt of 237 (so net watt 137.46 L and 99.54 R). If I understand correctly, it means a left leg P+ of 159.84 W and P- of -22.38 W. Right leg P+ of 127.62 W and P- of -28.08 W.