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
  • I have noticed that the power reported by my 3S pedals at the start of session on the trainer is usually noticeable higher (20-30W) than at the end of a session for the same cadence and gear combination. I have an old, magnetic brake, dumb trainer so I could well believe that it is the resistance of the magnetic brake when it is cold compared to when it is warm that is causing the difference.

    For reference, my bike and trainer are permanently set up in my garage where the temperature is cold at this time of year (around 0C) but consistent.
  • I just read through this thread and found that I have input here as well. Although my V3's have zero dropouts and is very consistent, I see the same "less watt picture" when comparing to NEO 2. My setup is V3 dual on DA 9100 cranks, indoor riding on NEO 2.

    Here's an example from todays training:

    Did a 5x1 min Z5 interval:

    - NEO showed = 451, 458, 454, 450, 454
    - V3 showed = 426, 438, 428, 431, 434

    So V3 approximate 20-25 W less than NEO 2. About 5% less

    Then I did a 2x10 min Z4 interval:

    - NEO showed = 315 and 317
    - V3 showed = 305 and 306

    Once again V3 was less with approximate 10 W. About 3%

    Then there was 2x20 min Z3 interval:

    - NEO showed = 280 and 284
    - V3 showed = 270 and 271

    Same picture; V3 was less with approximate 10-13 W. About 4%

    This picture, is what I have seen for a longer period. It's always the same - very consistent - V3 shows 5-20 W less than NEO 2. So yesterday I took of the pedals, cleaned the threading, mounted them again with a certain tightening. Picture remains the same, hence the numbers above.

    Off course the question here is; who is right? NEO 2 or V3 dual...? I have a pair of Assioma Duo's as well. I might do a test with them. At the moment, a friend of mine has borrowed them. So once their back, lets see about that test...

    But back to the start; my point here was, that I'm another one who see's the picture as others while riding indoor.

    So is there something about....?!
  • These are my findings
    Neo 2 Stages L Vector 3 Dual
    Avg Power 201 W 200 W 204 W
    Max Power 570 W 550 W 554 W
    Max Avg Power (20 min) 225 W 223 W 227 W
    Normalized Power 215 W 212 W 216 W

    Below 200 W the Neo 2 is less than the Vectors and above it tends to gain as the watts go up ( e.g. Max Power above). However as you can see the averages seem to even out.
  • Where a comparison of X and Y is being made, who is to say which one drifts or is more accurate?

    Most manufacturers offer up a +/- 2% accuracy...that means you could have a +2% powermeter comparing against a -2% powermeter....a difference of 4%.

    As a I just get on my bike and ride.....using the powermeter to regulate my output up climbs, on the levels and to gather data on Track efforts.......and to do HIIT sessions on the Ergo.....which gives a different reading to my powermeter....But as long as I only believe one of them is correct and use that one...then I will have comparative data to use in my training program.

  • The only way I could think of to truly know which is right, would be to use a control. In other words, you can't compare two different power meters to each other. I would think the easiest and most simple method would be to do the known weight hung from the pedal test both before a workout and at the end of the workout and see what the difference(if any) is.
  • So yesterday I did the static weight test. My V3 dual was within +/- 0,2% error...! However, V3 still show slightly less than NEO 2 (appr. 2-5%, higher wattage, higher diff.)

    So I also did a static weight test of my Assioma Duo's. Quite funny - they also measured within +/- 0,2% error. Today I'll do a test ride with them. I'll post the results here...

    So before any further comment in regards of you can't do this and you can't do that. Please just see this as a funny little experiment.

    One side note here; I have seen tests where V3 totally matches NEO (GP Lama on his 12 months review). In my case I have 2-5% less and a friend of mine has the totally opposite; NEO shows less than V3...?!

    So maybe my Assioma Duo test will reviel some interesting stuff...? I'll post info later...
  • So..... Here's the result. I swapped the pedals a bit before midway - that's why Assioma have data from the start and Vector in the end.

    How did it go? It's close! Both pedals measured slightly less than NEO2.

    Here's the data:

    - Interval..............12 min................10 min.................29 min....................50 min
    ----------------------------------------------------------------------------------------------------------------------------
    - Neo 2...............210 W................310 W..................311 W.....................275 W (low cad 62 rpm)
    - Assioma Duo...202 W................307 W....................-..............................-.......
    - Vector 3..................................................................305 W.....................269 W (low cad 62 rpm)
    ----------------------------------------------------------------------------------------------------------------------------
    = Diff.................... -8 W...................-3 W.....................-6 W........................-6 W
    = In %................-4,0 %................-1,0 %..................-2,0 %.....................-2,2 %

    Looking at a 10s power average all the way, I'd say the Vector had the Edge compared to Assioma. The numbers also confirm this. Vector beeing mostly 4-6 watts less than NEO2.

    What to do with this? You decide for yourself. Who is right, I don't know? But I had fun doing it... ;-)

    Cheers
  • Seems to me that this verifies that the NEO2 is the one that is not accurate. You have two separate direct measurement power meters that are within specs to each other and an independent weight. Both of those independently verified measurements are then consistently different to the trainer that doesn't really have a way to be independently verified.
  • Today's update....:

    Vector apparently settled in over night and now shows an error margin of 4,4 % (NEO 287 vs V3 275) ...... :-(

    So to conclude that NEO is wrong...? I don't think you can conclude that, with that drifting. What you can conclude; is that Vector pedals does drift. Whether how often or how much? Hard to tell...? Wonder what a static weight test would say now?

    Anyways...

    Since I got a new bike, I'm considering going back to crankbased powermeters. They're much less prone to errors / drift.

  • Here is a recent ride by tsmit. As you can see in the pictures the time alignment isn't perfect but the power agreement is very good at low and high levels. These pedals were installed last night and torqued in properly (so settling is minimal). Given this data the manufacturers in this test (Garmin vs SRM vs TACX) seem very much the same in reporting data. Of course if there is any doubt you can perform a hanging weight test. ciq.forums.garmin.com/.../1449946.png