Garmin vector 3s accuracy and crank length

Discuss light weight issues concerning road bikes & parts.
RocketRacing
Posts: 947
Joined: Thu May 10, 2018 2:43 am

by RocketRacing

So i was struggling to sort why i am making more power on my new tri bike vs my road bike.

I swap a single sided vector 3s pedal between bikes for power.

The road bike has 170mm cranks, and the tri bike 165mm. I looked at my garmin today... and oops... it had defaulted to 172.5mm. Apparently correct crank length is key for these power based pedals.

So i presume the shorter cranks on the tri bike resulted in overestimation of power. Does anyone know which way power readings go if you have a crank that is shorter/longer than what the power meter thinks the crank is?

by Weenie


TobinHatesYou
Posts: 4695
Joined: Mon Jul 24, 2017 12:02 pm

by TobinHatesYou

RocketRacing wrote:
Wed Sep 18, 2019 10:02 pm
So i was struggling to sort why i am making more power on my new tri bike vs my road bike.

I swap a single sided vector 3s pedal between bikes for power.

The road bike has 170mm cranks, and the tri bike 165mm. I looked at my garmin today... and oops... it had defaulted to 172.5mm. Apparently correct crank length is key for these power based pedals.

So i presume the shorter cranks on the tri bike resulted in overestimation of power. Does anyone know which way power readings go if you have a crank that is shorter/longer than what the power meter thinks the crank is?

If your cranks are shorter than what the pedals are configured for, then it will assume you have a greater mechanical advantage. Another way to think about it is your pedals think your feet are traveling a greater distance than they actually are. Your reported power will be too high proportional to the difference in crank length. This happened to me once when I updated my Assioma firmware. It got reset to 172.5mm when I was actually running 165mm cranks. I knew immediately something was wrong because my 20min power went from 288W to 301W.

jfranci3
Posts: 794
Joined: Tue Jul 26, 2016 5:21 pm

by jfranci3

Does it say that somewhere? I'd assume this would be used for fact-checking the cadence. The torque measure at the pedal is basically a digital scale. The mechanical advantage shouldn't matter. Maybe it's used to cancel out some vibration or oppisitie side crossplay.

TobinHatesYou
Posts: 4695
Joined: Mon Jul 24, 2017 12:02 pm

by TobinHatesYou

jfranci3 wrote:
Thu Sep 19, 2019 1:58 am
Does it say that somewhere? I'd assume this would be used for fact-checking the cadence. The torque measure at the pedal is basically a digital scale. The mechanical advantage shouldn't matter. Maybe it's used to cancel out some vibration or oppisitie side crossplay.

I guess what needs to be calculated is linear strain -> torque and angular velocity -> linear velocity.
Last edited by TobinHatesYou on Thu Sep 19, 2019 2:34 am, edited 1 time in total.

joejack951
Posts: 719
Joined: Fri Feb 18, 2005 6:50 pm
Location: Wilmington, DE

by joejack951

jfranci3 wrote:
Thu Sep 19, 2019 1:58 am
Does it say that somewhere? I'd assume this would be used for fact-checking the cadence. The torque measure at the pedal is basically a digital scale. The mechanical advantage shouldn't matter. Maybe it's used to cancel out some vibration or oppisitie side crossplay.
Unless I’m severely mistaken, pedal-based powermeters can only measure force via strain in the pedal spindle. To calculate power, the head unit needs to know torque and cadence. And the only way to get the former is to know the crank length.

RocketRacing
Posts: 947
Joined: Thu May 10, 2018 2:43 am

by RocketRacing

doing the math, the errors I am seeing pretty closely correlate with the % error in crank length from what the garmin assumed vs reality.

Bad news, I am not getting as strong as quickly as I had thought in the last month or so. Also bad news... I am not sure how long my garmin had assumed the wrong crank length... so I kind of need to scrap my power data. Garmin seems to do a good job at changing things unannounced with updates. Like how my garmin stopped trackign because GPS was shut off. I never shut my gps off, and it took a couple short rides before I figured it out.

jfranci3
Posts: 794
Joined: Tue Jul 26, 2016 5:21 pm

by jfranci3

TobinHatesYou wrote:
Thu Sep 19, 2019 2:22 am
jfranci3 wrote:
Thu Sep 19, 2019 1:58 am
Does it say that somewhere? I'd assume this would be used for fact-checking the cadence. The torque measure at the pedal is basically a digital scale. The mechanical advantage shouldn't matter. Maybe it's used to cancel out some vibration or oppisitie side crossplay.

I guess what needs to be calculated is linear strain -> torque and angular velocity -> linear velocity.
That doesn’t sound right. Torque is strain. Cadence just gives you the frequency. Actual travel doesn’t matter. Now a pedal needs to know strain at multiple axis because a lot of input won’t make it to the crank and you can apply tq off axis that will make it to the crank. Maybe the cadence is more important to know the exact crank angle detail to know what’s making it way to the chain.

GothicCastle
Posts: 246
Joined: Mon Jul 04, 2016 1:52 am

by GothicCastle

jfranci3 wrote:
Thu Sep 19, 2019 3:34 am

That doesn’t sound right. Torque is strain. Cadence just gives you the frequency. Actual travel doesn’t matter. Now a pedal needs to know strain at multiple axis because a lot of input won’t make it to the crank and you can apply tq off axis that will make it to the crank. Maybe the cadence is more important to know the exact crank angle detail to know what’s making it way to the chain.
Unless I'm not understanding you, torque (strain gauge output) is not power. The pedals want to output a power measurement, which is a measure of work. Work is a calculation of torque and some kind of rotation, like cadence or forward velocity.

If the vectors just gave you the output of the strain gauges, you'd be seeing something line newton meters, rather than Watts.

As you mentioned, knowing the pedal's position in space lets Garmin do interesting things like show pedaling smoothness.

kevinw
Posts: 123
Joined: Fri Jan 25, 2019 3:49 pm

by kevinw

The pedals actually measure just force but it needs to know torque (and cadence) to calculate power.

Torque = Force x Distance

In this case Force is measured by the pedals using strain gauges and distance is the crank length. So crank lenth is important with regards to power output.

Where I think the confusion above comes from is that it does seem counter-intiutive that you can exert the same force on the pedals for the same lenth of time (same cadence) but be doing less work with the shorter crank - but this is actually the case.

User avatar
cyclespeed
Posts: 644
Joined: Mon Jun 06, 2016 8:45 am

by cyclespeed

Power = torque x cadence

The Garmin 3S measures force at the pedal (and cadence), not torque.

To go from force to torque, you need a length unit (in metres), therefore crank length is essential.

User avatar
cyclespeed
Posts: 644
Joined: Mon Jun 06, 2016 8:45 am

by cyclespeed

Thanks, Kevin, we just posted pretty much the same answer.

If Force is 50 Newtons and cadence is 100, then a 175mm crank makes more power than a 170mm.

This is because the Force is being exerted over a bigger circle. Your foot/leg is doing more work, and therefore making more power.

The danger though is to simplistically think ´longer crank = more torque, more power = better', but it's not quite that simple.....

jfranci3
Posts: 794
Joined: Tue Jul 26, 2016 5:21 pm

by jfranci3

You’re right, distance not rate/rpm/cadence. My mistake.

TobinHatesYou
Posts: 4695
Joined: Mon Jul 24, 2017 12:02 pm

by TobinHatesYou

So we pretty much went around in a circle back to my initial reply. :p

joejack951
Posts: 719
Joined: Fri Feb 18, 2005 6:50 pm
Location: Wilmington, DE

by joejack951

jfranci3 wrote:
Thu Sep 19, 2019 12:52 pm
You’re right, distance not rate/rpm/cadence. My mistake.
You're both right.

Image

Source: https://en.wikipedia.org/wiki/Torque

RocketRacing
Posts: 947
Joined: Thu May 10, 2018 2:43 am

by RocketRacing

TobinHatesYou wrote:
Thu Sep 19, 2019 4:43 pm
So we pretty much went around in a circle back to my initial reply. :p
Yup.

For every 2.5mm of crank that your crank is shorter than your pedal/garmin thinks it is, there is about a 1.5% over reporting of power. And vice versa.

by Weenie


Post Reply
  • Similar Topics
    Replies
    Views
    Last post