Vector 2S power variability/noise

Jan 11, 18 15:58

Post #1 of 15 (1905 views)

Anyone's data look similar to this? Can't seem to find a good answer on the webs, so I'm beginning to think this is problem. I'm experiencing what I believe is pretty dramatic variation in second-to-second readings (observed both via Zwift and 920xt) and I'm wondering what magnitude is within normal limits. As a test, I rode a very constant 150W at 85 RPM. That's the data shown here. +/- 20W (or ~15%) seems extraordinarily high to me. Published reports appear to focus on error as it relates to average power (which is logical), but I'm curious whether these second-to-second variations are an indication of a faulty device. Garmin replaced the pedal pod, which appeared to improve the complete power dropouts I was experiencing, but did nothing for the variability. Thanks.

---------------------------------------------------------------

https://connect.garmin.com/modern/profile/domingjm

Last edited by: domingjm: Jan 11, 18 15:59

Re: Vector 2S power variability/noise [domingjm] [ In reply to ]

Jan 11, 18 16:20

Post #2 of 15 (1894 views)

domingjm wrote:

As a test, I rode a very constant 150W at 85 RPM.

Measured by what?

Re: Vector 2S power variability/noise [aravilare] [ In reply to ]

Jan 11, 18 16:33

Post #3 of 15 (1885 views)

Perceived effort ~3/10, indoor. What would you recommend? It's really pretty easy to maintain cadence +/- 1 RPM. And a difference of 40W with a 40lb flywheel would be pretty noticeable, right?
Basically I'm just asking if anyone's data, for a constant effort, has this magnitude of second to second variation. Thanks.

---------------------------------------------------------------

https://connect.garmin.com/modern/profile/domingjm

Re: Vector 2S power variability/noise [domingjm] [ In reply to ]

Jan 11, 18 17:04

Post #4 of 15 (1877 views)

domingjm wrote:

Perceived effort ~3/10, indoor. What would you recommend? It's really pretty easy to maintain cadence +/- 1 RPM. And a difference of 40W with a 40lb flywheel would be pretty noticeable, right?
Basically I'm just asking if anyone's data, for a constant effort, has this magnitude of second to second variation. Thanks.

Nearly every dcrainmaker PM review has some multi-meter comparisons (https://www.dcrainmaker.com/...in-depth-review.html).

What are you comparing the Vector 2S power data to? Data from a Kickr? A Tacx? Zwift virtual power?

Re: Vector 2S power variability/noise [aravilare] [ In reply to ]

Jan 11, 18 17:13

Post #5 of 15 (1871 views)

I've read more than a handful of comparison articles, both peer reviewed and consumer, but they usually compare average power after a ride or after an interval. I'm interested in the acute variations I'm seeing.
Again, I have nothing to compare this observation to, other than the apparent absence of this variation elsewhere. I've not heard anyone discuss it, so I'm trying to get an idea of what others experience on an indoor trainer under ~constant effort.

---------------------------------------------------------------

https://connect.garmin.com/modern/profile/domingjm

Re: Vector 2S power variability/noise [domingjm] [ In reply to ]

Jan 11, 18 17:32

Post #6 of 15 (1862 views)

domingjm wrote:

I've read more than a handful of comparison articles, both peer reviewed and consumer, but they usually compare average power after a ride or after an interval. I'm interested in the acute variations I'm seeing.
Again, I have nothing to compare this observation to, other than the apparent absence of this variation elsewhere. I've not heard anyone discuss it, so I'm trying to get an idea of what others experience on an indoor trainer under ~constant effort.

lol.

https://media.dcrainmaker.com/images/2017/07/image6.png

Trainers are not very accurate or precise power meters. Then again, neither are single-sided PMs.

Re: Vector 2S power variability/noise [domingjm] [ In reply to ]

Jan 11, 18 19:45

Post #7 of 15 (1838 views)

domingjm wrote:

Anyone's data look similar to this?

That's completely normal, irrespective of what kind of power meter you use. It's the reason why many people use and recommend 3 to 10 second averaging of the power field on the display. Even when you're riding on the trainer it takes extreme diligence and practice to achieve a truly steady power output.
If you are riding on an erg-controlled trainer, you will still see power variations like this because the erg's control loop does not respond that quickly.

Out on the road you can expect even more variation.

Less is more.

Re: Vector 2S power variability/noise [Big Endian] [ In reply to ]

Jan 11, 18 19:53

Post #8 of 15 (1835 views)

Big Endian wrote:

domingjm wrote:

Anyone's data look similar to this?

That's completely normal, irrespective of what kind of power meter you use. It's the reason why many people use and recommend 3 to 10 second averaging of the power field on the display. Even when you're riding on the trainer it takes extreme diligence and practice to achieve a truly steady power output.
If you are riding on an erg-controlled trainer, you will still see power variations like this because the erg's control loop does not respond that quickly.

Out on the road you can expect even more variation.

Thanks for the input! I use 10sec averaging on the watch, but Connect doesn't appear to have a smoothing function. No big deal. More than anything, I just wanted some input on whether I need to complain more to Garmin. The complete power drops I was experiencing made me skeptical of the device in general.

---------------------------------------------------------------

https://connect.garmin.com/modern/profile/domingjm

Re: Vector 2S power variability/noise [domingjm] [ In reply to ]

Jan 11, 18 21:49

Post #9 of 15 (1822 views)

New 1sec PR.....

---------------------------------------------------------------

https://connect.garmin.com/modern/profile/domingjm

Re: Vector 2S power variability/noise [domingjm] [ In reply to ]

Jan 11, 18 21:58

Post #10 of 15 (1817 views)

domingjm wrote:

New 1sec PR.....

If you were interested in anything besides whining, you'd post the cadence from the Vector 2S since the headunit you're using is the one that converts torque into power values...

Last edited by: aravilare: Jan 11, 18 21:59

Re: Vector 2S power variability/noise [aravilare] [ In reply to ]

Jan 11, 18 22:02

Post #11 of 15 (1813 views)

You've provided lots of helpful posts. Thanks for your effort.

---------------------------------------------------------------

https://connect.garmin.com/modern/profile/domingjm

Re: Vector 2S power variability/noise [domingjm] [ In reply to ]

Jan 11, 18 22:09

Post #12 of 15 (1808 views)

If you weren’t too busy being sensitive, you’d be able to diagnose the issue from what I told you. But oh well!

Re: Vector 2S power variability/noise [aravilare] [ In reply to ]

Jan 11, 18 22:34

Post #13 of 15 (1798 views)

I bookmarked this thread because your advice will always be my favorite.
The PR screenshot was a total joke about a weird ass coincidence: I step off the bike only hours after posting about my pedals, then see the most unrealistic abnormality I've ever seen. How did that warrant any comment from you other than "lol" or "haha"?

---------------------------------------------------------------

https://connect.garmin.com/modern/profile/domingjm

Re: Vector 2S power variability/noise [domingjm] [ In reply to ]

Jan 11, 18 23:02

Post #14 of 15 (1792 views)

domingjm wrote:

I bookmarked this thread because your advice will always be my favorite.
The PR screenshot was a total joke about a weird ass coincidence: I step off the bike only hours after posting about my pedals, then see the most unrealistic abnormality I've ever seen. How did that warrant any comment from you other than "lol" or "haha"?

Nice retort, 10/10.

Re: Vector 2S power variability/noise [domingjm] [ In reply to ]

Jan 12, 18 4:33

Post #15 of 15 (1761 views)

domingjm wrote:

As a test, I rode a very constant 150W at 85 RPM. That's the data shown here. +/- 20W (or ~15%) seems extraordinarily high to me.

You may have seen plots of torque during a crank revolution but it’s not anything close to constant. Torque or force applied to the pedal peaks around 3 o’clock and is close to zero when the pedals are at 6 and 12 o’clock. Most powermeters simply average the power over one second and display that but in one second you don’t get an integral number of pedal strokes unless your cadence is exactly 60, 90 or 120 RPM. The result is the variability you see. If you change your cadence you’ll likely see the variability change.

The variability also looks much lower if you do your intervals at 400W instead of 150 :)

Our Partners

transition-DOT-area-AT-slowtwitch-DOT-com

Advertise with us

©1999-2024 Slowtwitch, Inc., and Slowtwitch.com
Reproduction of material from any Slowtwitch.com page
without written permission is strictly prohibited.