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
  • You're killing me with this. So because the NEO and Vector are now 4% off today, you are concluding that the Vector is drifting? Why would you think the Vector is drifting rather than the NEO?

    It seems to me that result is completely in line with your previous test where the NEO was all over the place and the Assioma and Vector where within specs the entire ride. You can independently verify whether the Vector is correct or not by doing the static weight test. You have no way to independently verify the NEO is accurate that I can think of.

    Frankly, if I were you I'd be contacting TacX about why their trainer power reading is drifting and how you can go about independently verifying that it is not drifting. Garmin has already provided a capability to do so and a procedure to follow. When you did that, you found it was well within specs.
  • But that's exactly my point Browner40 I don't know who to trust?

    I have done a static weight test - with both Vector and Assioma - both beeing very good. And yesterday V3 was the one closest to NEO (Assioma very close to too). Both V3 and Assioma agreed on NEO showing to much. Maybe they're right - I can't tell...? What disturbs me here is that V3 the day after just drops another 2%.... WTf...? How can that be reliable? Well it isn't! And this is my issue. Call it settling in. I'm not sure what I call it...

    Check out this video - ALERT - 40 min tech stuff about powermeter principles. What is solid priciples and what is less good. I think some of the explanation lies here....!

    https://www.youtube.com/watch?v=zZ9fh6srmnk

    I have found out a way to test my NEO and I will do so in a few days (currently I'm out). I'll post the result here.

    GP Lama in comparing powermeters:

    https://www.youtube.com/watch?v=JhheDLXRM7I
  • The static weight test is the control. I'm confused why you don't know who to trust? If both pedal meters agree with each other, AND THE CONTROL, why would you still doubt them compared to the NEO? You say that the next day the V3 dropped 2%, but based on your description I would not make that assumption. Instead I would be assuming that the NEO increased its measurement by 2% from the actual measurement. Since the control showed that the pedal based meters had no real drift during the ride, while the NEO fluctuated by as much as 4%.

    Seems to me that there is a very consistent result on all of your tests and it does not show the V3 as unreliable. Instead, it shows the NEO starting out with up to 4% higher power measurement than a direct measurement pedal system, and then it settles down throughout the ride to be ~2% higher than the pedals.
  • So a bit more testing today....

    Did an FTP test(!) I'd say I know my body quite good, since I have done quite a few FTP tests. I'm usually around 330'ish W in off season which typically goes up to 350-365 in "hot" season. Current shape is quite good - building up for La Marmotte in France this summer.

    So in a test right now, I'd expect around 350-360'ish W.

    I also found a video on how to reset the NEO 2 to factory standards. So I did that. I then did the test according to Zwifts protocol (the long one, 20 min).

    So NEO 2 came out with.......= 350 W
    Vector 3 came out with......... = 340 W

    Hmmmm.... Throughout the warmup phase, with 10 s rolling average, V3 was consistently 5-15 W lower than NEO. The higher the wattage, the higher the offset.... Hmmm that's too much for me!

    I have not done another static weight - I might do that this week. But....

    But based on my own feeling and X numbers of FTP tests, I don't believe V3 is right here.

    Edit
    I did reset installation angles and zero'ed twice after 10 min burn in on V3.

    It's not that I have concluded I'm done with V3. As I have written in another post, I really like that pedals. But I am considering a Quarq Dzero. I am tending toward more belief in a crank based system...

    Edit 2
    Just saying
    https://support.tacx.com/hc/en-us/articles/360000455705-Calibration-and-power-measurement
  • Todays ride....

    45 min low cad (avg. 63) hard effort:

    - NEO 2.......... = 320 W avg cad 63
    - V3 ............... = 305 W avg cad 63

    My best ever outside is 333 W for 1 hour (Assioma Duo, 2018/05). Like I said the other day, I'm currently in a good shape. My FTP test yesterday is actually on par with my PR FTP from last year, since I have lost a bit of weight.

    So with this in mind, I have a hard time believing V3. I'd say I'd expect 320-325 W for 1 hour on fresh legs. Definitely not 305'ish.

    BTW - in the above, V3 is 4,9 % off.

    Take it for what is is. Some are happy and have results in the expected area. I just don't. Too bad for me. But don't make a big story out of it. After all - it's just my 2 cent ;-)

    EDIT:

    The whole ride:

    - NEO 2.......... = 251 W avg cad 71, max 440 W
    - V3 ............... = 238 W avg cad 69, max 439 W

    = 13 W's less on avg. or 5,5 % off.
  • So I bought a Quarq Dfour91 (it's Quarq's Dzero for Shimano 9100 series). I'm so glad I did - here's why...

    On todays ride, the first one with Quarq, the first part (50 min) of it was an ERG-mode workout with 4x10s 1kW surges.

    - NEO....... = avg. 213 max 1085
    - Quarq..... = avg. 211 max 1034

    The remaining part (55 min) was free-ride in Zwift with cadence drills:

    - NEO....... = avg. 178 max 362
    - Quarq..... = avg. 182 max 369

    This within +/- 2% error margin (I don't pay much attention to the 10s surges - since it's afterall only 10s intervals)

    In addition, I had a 10s rolling avg. power on my Edge 1030. The Quarq was always within +/- 2-4 W when riding steady. This is SO MUCH more in line with what I'd expect.

    So are my Vector 3 wrong? I did a solid static weight test with great result. So did Assioma. But then what is wrong here? I don't know. But I do think this has something to do with this video. Pedals are just more prone parasitics in power measurement. Some may see it more than others. I think this is what I did.

    https://www.youtube.com/watch?v=zZ9fh6srmnk

    So where does this leave me. Well as I see it; I learned a lesson. Pedals are just no good placement for precise measurement of power. At least not if you want super reliable, super trustable power each and everytime. For this, you need a crankbased powermeter. This is my take on this.

    With this, fairwell Vector 3. I will never buy a pedalbased powermeter again. I'm just too sensible to the marginal errors that I think I see.
  • Surely the thing to do is to ride with the Quarq and Vector 3's on the Neo...You then get 3 readouts to compare....You can't knock the Vectors until you have done the Threesome test....

    Looking forward to seeing the results of that test.

    I admire your tenacity....I couldn't have done the work you have.
  • But that's also the thing Merchantofcool I just want to ride my bike, I just want enjoy my ridning. In fact I don't like spending hours on hours testing and writing about. So unfortunately, I will not do the test you wish for. Most particulaly because I changed my setup slightly to fit the Dura Ace pedals. But also because I just want ride my bike.

    So take it for what it is.

    I tiny note on the Quarq. I do see small differences in shorter sprints, but I suspect that to be due to the short time. Looking at the average on the whole; Neo and Quarq agree very well. On today's last hour (without interval) Neo was 186 and Quarq was 187... Or was it versa? Well nevermind... ;-)
  • Ultimately it doesn't seem to matter what anyone else says. You decided that you didn't like what the pedal based meters were telling you and so you just kept searching until you got the answer you wanted. The pedal based meters are the most precise way to see what you are actually putting out power-wise. The crank based and NEO meters agree with each other and make you feel better, but the pedal based meters agree with each other AND a separate, independent measurement using a known weight. I think all this proves is that the NEO and Quarq are using the same calculation to estimate power, you still have not independently verified either one. Ultimately if you feel better about it, good for you.
  • Ultimately it doesn't seem to matter what anyone else says. You decided that you didn't like what the pedal based meters were telling you and so you just kept searching until you got the answer you wanted. The pedal based meters are the most precise way to see what you are actually putting out power-wise. The crank based and NEO meters agree with each other and make you feel better, but the pedal based meters agree with each other AND a separate, independent measurement using a known weight. I think all this proves is that the NEO and Quarq are using the same calculation to estimate power, you still have not independently verified either one. Ultimately if you feel better about it, good for you.


    This.