power meter drift issues?

hey all,

unfortunately i've fallen into the power meter accuracy snake pit.

recently i've been seeing what looks to me like power meter drift. you
can see a pretty good example of this from my last trainerroad workout
here (done on a pre-2017 wahoo kickr snap):

https://analyze.dcrainmaker.com/#/public/ade655fb-936b-462b-6037-ef06a2775681

one power meter is a set of garmin vector 3 pedals with a claimed
accuracy of +/- 1.0%. (i've followed the service advisory notes for the
vector and verified the spindle torque. after installing the vectors i
reset the install angles and i also did a static torque test, as
described on the official garmin support pages, using a rice lake
calibrated 20 kg weight). the second power meter is a powertap g3 hub
with a claimed accuracy of +/- 1.5%. both power meters were calibrated
(zero'd) before the workout, which was done indoors at a consistent
temperature of around 70 F.

at the start of the ride, the powertap numbers are a few watts below the
vector numbers, which makes sense given some drivetrain power loss. but
by the end of the ride the power meters have reversed, with the powertap
giving numbers about 10% above the vector numbers, which seems wrong.
(aside from the drift, the difference between the two power values seems
out of spec for claimed accuracy of the devices, especially when you
consider that there should be some drivetrain power losses).

has anyone else seen something like this before? to me it seems like
one of the power meters is drifting out of spec? any ideas what else
might be going on or how to root cause the problem (short of buying
a tacx neo or a third power meter ;)?

cheers,
ed
  • Comparing the two, they start out pretty close to the same and then drift apart over about 10 minutes or so. It may be that when you did the zero offset on the pedals, they were adjusting to an ambient temperature shift? It normally takes about 15 minutes for the pedals to achieve thermal equilibrium. If you're moving the bike from a warmer to cooler environment (or vice versa), either calibrate before moving it, or about 15-20 minutes after.
  • To be honest that's what I thought. Initially there seemed to be some parity with the outputs but as the event went on I was seeing 20-30 watts difference between Zwift and my Garmin head unit. There was no change in location for the bike, I was in the garage and it was fairly cold last night about 2 to 3 degrees Celsius at the start which is when the calibration would have been done and normally by the end of the session it warms up about 2 degrees just from bodyheat with no other heating needed so not enough for that drift I suggest. I will do a few checks, calibrate after a few sprints, and run a test tomorrow morning with the Neo in ERG mode and see what results I get.
  • When you see that kind of drift, you can sequentially allow the different devices to re-zero, and through that process you can figure out or at least get an idea of where the problem might be coming from. Many bike trainers will do a re-zero if you let the wheel spin down to zero; if you did that on the Neo, and then went back to a steady pedaling level, you could see if the relative output from the two changed. Then, you can separately re-zero the pedals, and then see if the relative output changed after that.
  • Last night I removed the pedals, degreased threads, cleaned faces, re-greased and installed with a good tighten.
    This morning I calibrated before the start, reset install angles and did a ten minute warm up with a few sprints and also turneed off the auto-zero function.
    I then re-calibrated the pedals and began a Step test, starting at 150watts increasing 25 watts every 2mins 30 seconds using the Tacx Neo ERG

    The results for each interval were of were as follows:
    Step 1 - Neo: 174watts / Vectors: 175watts +0.5%
    Step 2 - Neo: 200watts / Vectors 200watts 0%
    Step 3 - Neo: 225watts / Vectors: 222watts -1.3%
    Step 4 - Neo: 248watts / Vectors: 243watts -2.0%
    Step 5 - Neo: 274watts / Vectors: 267watts -2.6%
    Step 6 - Neo:300watts / Vectors: 289watts -3.7%
    Step 7 - Neo: 326watts / Vectors: 313watts -4.0%
    Step 8 - Neo: 347watts / Vectors: 323watts -6.9%

    I failed the next interval so killed off the ERG and decided to spin out the session at about 150 watts based on the Vectors thinking that this would coincide with the Neo once the power dropped off however, the last two remaining intervals which are 2:30 long and then 5:00 long were as follows:

    Neo 162 watts / Vectors 151 watts -6.8%
    Neo 160 watts / Vectors 150 watts -6.3%

    This would suggest that it is not necessarily the higher power outputs which the drift i occurring at as it seems to drift with time. Could this be somehow heat related within the pedal and the higher outputs are generating more heat which is throwing it out as it seems that dropping back to lower output started to reduce the difference? If I'd had more time I should have calibrated again at this point and continued to see what happened. The files from the Vectors is attached below.

    3207280854.zip
  • Last night I removed the pedals, degreased threads, cleaned faces, re-greased and installed with a good tighten.
    This morning I calibrated before the start, reset install angles and did a ten minute warm up with a few sprints and also turneed off the auto-zero function.
    I then re-calibrated the pedals and began a Step test, starting at 150watts increasing 25 watts every 2mins 30 seconds using the Tacx Neo ERG

    The results for each interval were of were as follows:
    Step 1 - Neo: 174watts / Vectors: 175watts +0.5%
    Step 2 - Neo: 200watts / Vectors 200watts 0%
    Step 3 - Neo: 225watts / Vectors: 222watts -1.3%
    Step 4 - Neo: 248watts / Vectors: 243watts -2.0%
    Step 5 - Neo: 274watts / Vectors: 267watts -2.6%
    Step 6 - Neo:300watts / Vectors: 289watts -3.7%
    Step 7 - Neo: 326watts / Vectors: 313watts -4.0%
    Step 8 - Neo: 347watts / Vectors: 323watts -6.9%

    I failed the next interval so killed off the ERG and decided to spin out the session at about 150 watts based on the Vectors thinking that this would coincide with the Neo once the power dropped off however, the last two remaining intervals which are 2:30 long and then 5:00 long were as follows:

    Neo 162 watts / Vectors 151 watts -6.8%
    Neo 160 watts / Vectors 150 watts -6.3%

    This would suggest that it is not necessarily the higher power outputs which the drift i occurring at as it seems to drift with time. Could this be somehow heat related within the pedal and the higher outputs are generating more heat which is throwing it out as it seems that dropping back to lower output started to reduce the difference? If I'd had more time I should have calibrated again at this point and continued to see what happened. The files from the Vectors is attached below.

    3207280854.zip

    i have the vector 3 for about 2 months.while in the first month only outside they were ok,when switching to indoor i have same problems with them drifting(to low values)compared to p2max negeco.also later in the ride . I have also noticed that when using them on rotor 3d24 cranks the numbers are more consistent from ride to ride.but when mounting them on dura ace 9100 the values are 6-10 watts lower.On the rotor they sit flusher on the crank arm and they get outside the crankarm 1-2mm more than on dura ace.
  • Further to my post yesterday, I went on again last night and started with no new calibration i.e. exactly where I left off yesterday morning and the Auto Zero turned off. I did a 20 minute warm up with a few sprint hard efforts in then stopped and calibrated. The data for the first 20 minutes reflects where I left off yesterday morning. Average power from the Neo was 220 watts with the Vectors reading 208 watts a 5.5% difference. I then did a one hour event after calibrating and noticed my readings seemed to be tracking reasonably well for the duration of the event with no drift. This is borne out by the results which showed the Neo reading 226 watts average and the Vectors reading 229 watts. Normalised readings were 231watts and 233 watts so within +/-1% accuracy of both devices and given the Vectors are at the front of the drivetrain I'd expect these to read a bit higher than the Neo which is my past experience of using both.

    I'm not entirely sure what to conclude. Perhaps they needed some time to bed back in again after servicing? Perhaps I need to leave a longer period of time from the start before calibrating. Ordinarily this wouldn't be a problem if I was leaving them on the same bike but I change them between bikes regularly. I will continue to monitor and see how they go.
  • Further to my post yesterday, I went on again last night and started with no new calibration i.e. exactly where I left off yesterday morning and the Auto Zero turned off. I did a 20 minute warm up with a few sprint hard efforts in then stopped and calibrated. The data for the first 20 minutes reflects where I left off yesterday morning. Average power from the Neo was 220 watts with the Vectors reading 208 watts a 5.5% difference. I then did a one hour event after calibrating and noticed my readings seemed to be tracking reasonably well for the duration of the event with no drift. This is borne out by the results which showed the Neo reading 226 watts average and the Vectors reading 229 watts. Normalised readings were 231watts and 233 watts so within +/-1% accuracy of both devices and given the Vectors are at the front of the drivetrain I'd expect these to read a bit higher than the Neo which is my past experience of using both.

    I'm not entirely sure what to conclude. Perhaps they needed some time to bed back in again after servicing? Perhaps I need to leave a longer period of time from the start before calibrating. Ordinarily this wouldn't be a problem if I was leaving them on the same bike but I change them between bikes regularly. I will continue to monitor and see how they go.


    Strange as i have experienced same thing.after switching them from bike to bike,after 3 days tested them against an elite direto(brand new)calibrated and the power was only in 1 watt diff for 1 hour ride.so they work very good when settled in.i was expecting them to read a little higher due to their position on bike than Direto but they were 1 watt lower for average.Good news
  • Two things I am wondering about here:

    (1) I've never ridden a Neo but at least my Flux gets pretty warm when I crank up the power - so it might be, that the thing drifting here is the Neo.
    (2) I am always a bit skeptical when I see power meter comparisons on a static trainer that is used in ERG mode: At least for the Flux the recorded power in ERG mode appears to be much smoother than the one recorded with my Vector 3 - not totally unrealistic but definitely smoother. Of course a noisier signal - probably closer to reality - can produce substantially different average power numbers.

    Maybe a good way to remove ERG mode from the equation would be to repeat the test on something like Alpe du Zwift and trying to roughly aim for certain average power levels by adjusting your personal output power and e.g. the gearing.
  • I am having the same issues but thought I'd add a data point in case this is relevant. Looking at my head unit, the room temperature rises up to 10 degrees F during a 60 minute workout at high intensities (I suspect from the trainer and my body, but I don't know). Are the vectors expected to be able to cope with that? I haven't been using my trainer's power meter to compare, but I notice "cadence drift" in later intervals whereby at the same cadence the vectors report less power. Sometimes this is dramatic enough that I have to shift up for intervals at the same intensity towards the end.
  • Are the vectors expected to be able to cope with that?

    Yes. All Vector pedals have continuous temperature compensation across the entire operating temperature range.