Go Back  Bike Forums > Bike Forums > Road Cycling
Reload this Page >

Advice on Power Meter

Notices
Road Cycling “It is by riding a bicycle that you learn the contours of a country best, since you have to sweat up the hills and coast down them. Thus you remember them as they actually are, while in a motor car only a high hill impresses you, and you have no such accurate remembrance of country you have driven through as you gain by riding a bicycle.” -- Ernest Hemingway

Advice on Power Meter

Old 05-25-20, 09:31 PM
  #51  
woodcraft
Senior Member
 
woodcraft's Avatar
 
Join Date: Apr 2012
Location: Nor Cal
Posts: 5,063
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 1385 Post(s)
Likes: 0
Liked 397 Times in 241 Posts
I'm getting a used (1 yr) powertap hub.

I have a rim that I think will work, & can use it on 2nd bike and a check on primary one,

which has an older Quarg crank spider. I could even loan it out to the PM curious.
woodcraft is offline  
Likes For woodcraft:
Old 05-26-20, 08:28 AM
  #52  
topflightpro
Senior Member
 
topflightpro's Avatar
 
Join Date: Jul 2007
Posts: 6,837
Mentioned: 54 Post(s)
Tagged: 0 Thread(s)
Quoted: 1508 Post(s)
Liked 234 Times in 156 Posts
If you're not familiar with him, RChung is a legit source of information and more than likely knows more about PMs and power data than anyone else on this forum.

Last edited by topflightpro; 05-26-20 at 11:16 AM.
topflightpro is offline  
Likes For topflightpro:
Old 05-26-20, 09:38 AM
  #53  
OUGrad05
Senior Member
 
Join Date: Jun 2017
Posts: 206
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 110 Post(s)
Liked 5 Times in 5 Posts
I've got the Assiomas and they're good. The downside is if you get a second bike later and want to go with crank based, you'll have to have two different shoes since they'll have different pedals. Or you could get another set of Assiomas. So far I've been pleased.
OUGrad05 is offline  
Old 05-26-20, 04:05 PM
  #54  
PoorInRichfield
Senior Member
 
Join Date: Nov 2013
Location: Richfield, WI
Posts: 330

Bikes: Trek Domane SL7 Disc, Trek Boone 9, Cannondale F29

Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 152 Post(s)
Liked 103 Times in 72 Posts
Originally Posted by RChung View Post
Sigh.

1. It's possible to check the accuracy of a bathroom scale.
2. Whether consistency is all you need depends on what you're doing with the information.
3. My step dad was a butcher so I used to work in the shop. I can guarantee that if the scales in the shop were consistent but inaccurate, we would have been in a lot of trouble.
4. Whether consistency is all you need depends on what you're doing with the information.
A butcher or any other company that sells items to the public would be in trouble with the U.S. Department of Weights and Measures (and maybe other governmental organizations) if such company was found to be cheating customers out of product due to inaccurately configured scales. With that in mind, if I do a search for "scale calibration services", I find a lot of them. Heck, all you'd need to do is buy a calibration weight and set it on a scale and at least know if your scale was in the ballpark.

My bathroom scale example isn't a perfect analogy to power meters. For one, power meters are more complicated (at least they are to me). Do a search for "bicycle power meter calibration services" and see what you get. I got nothing. Even on the web site Power Meter City where all they do is sell power meters, I see no calibration service. So how can I verify that this power meter I just spent $1k+ on isn't just making @#$@ up? I could buy a couple other power meters and compare results, but how would I know they aren't all inaccurate?

I did find this article, "Why Calibrate Your Bikes Power Meter" and hoped it would contain a link to a company that does calibration, but no such link exists. Not only that, I'm pretty sure this site isn't talking about calibration at all. It's referring to zeroing-out (the equivalent of the tare feature on a scale) one's own power meter so that it's accurate with itself which is the point I've been trying to make. That's not calibration. From the article:

4. WHY SHOULD I CALIBRATE MY POWER METER?
If you’re tracking training data, then zeroing regularly is important to get data that is comparable week-to-week. A power meter can easily drift by 2% in a few days if it is not being zeroed regularly, so to track long term training improvements zeroing often is important.
Do the strain gauges in a power meter wear-out or get inaccurate over time? Does fatigue of the bicycle's materials, like aluminum crank arms, play into the power readings over time? In a perfect world, a power meter would stay accurate forever, but I'm pretty sure that it's only a matter of time before a power meter should really be recalibrated for accuracy... if such service existed.

So my original concern stands... How does anyone know if the expensive power meter on their bike is remotely accurate when compared to other power meters and not just itself? If I ride with a friend and he's on a steel mountain bike with aluminum cranks and an off-brand power meter hub that's 5 years old and I'm on a carbon road bike with power pedals that I bought yesterday, how would we know if both power meters are accurate so we could compare each other's fitness levels? Hopefully both meters would be "close enough" to give ourselves something to talk about, but I doubt they'd both be dead-on accurate.
PoorInRichfield is offline  
Old 05-26-20, 05:11 PM
  #55  
RChung
Senior Member
 
Join Date: Sep 2009
Posts: 1,599
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 367 Post(s)
Liked 175 Times in 74 Posts
Originally Posted by PoorInRichfield View Post
So my original concern stands... How does anyone know if the expensive power meter on their bike is remotely accurate when compared to other power meters and not just itself?
Sometimes people seem to think that if they don't know how to do something, or if they can't find the way to do something via a quick google search, then either that something is impossible to do or, at least, no one else can do it either.

For power meters based on strain gages (or gauges), there are three places where errors can occur: in force measurement, in rotational speed, and in the manufacturer's firmware that takes observations about force and speed and combines them into a calculation of power.

To measure force, the exact procedure varies from power meter but the typical method is exactly analogous to checking the accuracy of a scale in a bathroom or a butcher shop: you weigh a known mass and see if they agree. Here's how Garmin suggests checking the force on their Vectors: https://support.garmin.com/en-US/?fa...vi3oBYZUUZdPV6 . Similar methods exist for most strain gage power meters. That's how I check the accuracy of my Power Tap: I hang some known weights from the pedal, I combine the crank length with the gear ratio, and calculate what torque the hub ought to report. Then I see if it does. Lather, rinse, repeat for different gear ratios.

To measure rotational speed, you have to check the accuracy of the rotating sensor: the pedal spindle, the BB, or, for the Power Tap, the rear hub. As it happens, this used to be the most reliable part of a power meter but with the current pedal power meters that measure rotational speed with accelerometers rather than once per rev with a reed switch, sometimes this part of the calculation can be off, too. You can check cadence if you know rear wheel speed and the gear ratio, so I typically check cadence that way. That's how I found out that accelerometer-based cadence is noisier than reed-switch-based cadence.

The last piece is in the manufacturer's firmware. SRM's used to sample at something like 200Hz, PT's used so sample around 60Hz, I think Stages sample at 64Hz, so even if you're pedaling at 120 rpm, there are a couple dozen measurements that are being combined each second. That said, that's how I identified some firmware oddities for certain manufacturers.

The foregoing describe what are called "static" checks. There are ways to do dynamic checks too, though they're more of a hassle. I only suggest doing them if you have a type of power meter that doesn't use strain gages (there are a couple, but they're not very common). If you're in that situation, I can describe that protocol too, but that's really kind of a threat. Most normal people would never need to do a dynamic check; most normal people would've stopped reading several paragraphs back.

And, although Power Meter City may not offer calibration services there are power meter dealers who do, so if you find hanging a weight off your pedal too daunting you can pay someone to do it for you.
RChung is offline  
Old 05-26-20, 06:49 PM
  #56  
PoorInRichfield
Senior Member
 
Join Date: Nov 2013
Location: Richfield, WI
Posts: 330

Bikes: Trek Domane SL7 Disc, Trek Boone 9, Cannondale F29

Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 152 Post(s)
Liked 103 Times in 72 Posts
Thanks for the detailed response. Links like the one above are what I am looking for. I didn't state that no one in the world could possibly calibrate a power meter on a bicycle, but it clearly isn't common knowledge and most people I know who ride aren't home bicycle mechanics, so they'd want to have someone calibrate their power meters for them (of which I have yet to find mention of such a service via an online search). I'll remember to ask the mechanics at my local bike shops if they've ever calibrated a power meter... to see if it's common knowledge.

In reading the directions, I find the first step a little humorous in light of my bathroom scale example from earlier ...
1. Accurately weigh and record weight of entire mass in kilograms (weight, hook, rope, etc.)
Since a power meter doesn't come with a 25 pound weight (mass), everyone who wants to make sure their Garmin power pedals are calibrated needs to go buy a 25 pound weight/mass. Then according to Step 1, they must accurately weigh the mass. LOL! With an inaccurate el-cheapo scale purchased on Amazon? Unless one has also purchased a reference weight to ensure that the scale is indeed accurate or has paid a company to come and calibrate the scale, who knows if the scale is accurate? No one I know has the fancy scales that doctor's offices use that actually have weights on them. If the scale is not accurate, don't proceed past Step 1 as the data will be meaningless.

[Rhetorical Question] Of all the people who buy power meters, what percentage actually perform the steps prescribed in the article on a consistent basis (or ever)? [/Rhetorical Question]

I'm willing to bet that the majority of power meter users buy the meters and assume accuracy from day 1, then blissfully use the meter with no consideration for calibrating it and assume it to be "right" for as long as they use it, just like they do with their cycle computers. If that's true, that group of people does not need $1k+ left/right mega top-end power meters. They'd benefit just as much with an inexpensive power meter that's consistent with itself. Racers and highly competitive triathletes are a different story, but I bet that most athletes that don't depend on cycling to pay the bills would do just fine training with a low-end power meter as long as it gives consistent readings with itself over time.

PoorInRichfield is offline  
Old 05-26-20, 07:47 PM
  #57  
RChung
Senior Member
 
Join Date: Sep 2009
Posts: 1,599
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 367 Post(s)
Liked 175 Times in 74 Posts
Originally Posted by PoorInRichfield View Post
Thanks for the detailed response. Links like the one above are what I am looking for.
Since you're looking for links, in addition to the Garmin Vector instructions, here are two more videos for how to do a static check for the Power Tap and the Quarq.

I got a friend who works at the university to weigh my barbell weights in the physics lab and write their mass on them with a sharpie. I also could've taken them to a FedEx or UPS or USPS office to get to within an ounce or so. Actually, if my step dad were still working, I could've brought them to the butcher shop. You don't really need NIST weights (all you need is to weigh barbell weights on a scale that's been certified).

20 years ago when power meters were unusual and expensive (and sometimes finicky) we all used to check their accuracy. It was sort of like needing to know how to adjust your own derailleurs, or true your own wheels. We used to call it the "Stomp test" or "the Coggan Stomp" and Coggan described it here: https://forum.slowtwitch.com/Slowtwi...test_P2327677/ That so few do it today, or know that it can be done, doesn't mean that it's hard to do. As I said in my first few posts in this thread, most riders who use a power meter do only one thing: they use it to train their FTP, and training FTP is one of the least demanding things you can do with power data. I didn't say this here but elsewhere I've said that if all you need is consistency (and for many riders that's all they need) then a much cheaper and very consistent device is a wristwatch. Just time yourself over a regular course, check the weather service for windspeed and direction, and write it all down in a notebook. Over time you'll figure out whether you're getting faster or not. That's how riders have trained pretty effectively for more than a century.
RChung is offline  
Old 05-26-20, 09:15 PM
  #58  
colnago62
Senior Member
 
Join Date: Aug 2014
Posts: 1,896
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 476 Post(s)
Liked 146 Times in 93 Posts
Originally Posted by RChung View Post



And, although Power Meter City may not offer calibration services there are power meter dealers who do, so if you find hanging a weight off your pedal too daunting you can pay someone to do it for you.
I sent mine back to the manufacturer and they did a diagnostic, calibration, firmware update and cleaning.
colnago62 is offline  
Old 05-26-20, 09:52 PM
  #59  
ridethecliche
Village Idiot
 
Join Date: Jan 2007
Location: Dirty Jerz
Posts: 20,422
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 66 Post(s)
Liked 57 Times in 42 Posts
Originally Posted by colnago62 View Post
I have never heard or talked to a sprinter who used a powermeter to even out leg strength. My guess would be that they spend so much time in the gym that an issue like that would dealt with there. What is cool is how sprinters and their coaches use power to determine what their optimum gear will be for a particular track or situation. A good gear not only gives quick acceleration, but also slower deceleration.
Yeah. I was being slightly superfluous.

Originally Posted by RChung View Post
Oh, I think I've said that too, since I think the majority of riders use their power meters in pretty simple ways. That said, some don't, and we can't know ahead of time who is going to be in which camp. It's better to say that whether you need more than consistent inaccuracy depends on what you're going to do with the data than it is to declare that no one ever needs anything more than that.

Actually, I know Ray Maker. He's consulted with me (and occasionally still does) on especially complicated power meter issues. I helped him set up his test protocols. And, although I've never met Shane in person, if you look up his observations on dual-sided Stages PMs you'll see what he thinks of their accuracy.
Horses for courses. We're on a road cycling forum, not an elite athlete or even the racing sub here. It's kind of like the 'technically correct is the best correct' axiom. Ends up being largely semantics.
Not sure if you're insinuating that I said that 'no one needs anything more than that' but that was never my point. In the event that it was directed at me... plz drop the strawman. He's on fire.

Glad to know that you're able to help them out.

Originally Posted by topflightpro View Post
If you're not familiar with him, RChung is a legit source of information and more than likely knows more about PMs and power data than anyone else on this forum.
Yeah, I remember him from posting here ages ago. He knows his ish. Doesn't mean he can't be pedantic though. He knows he's technically correct!
The comment below kinda exemplifies what I'm talking about.

Sometimes people seem to think that if they don't know how to do something, or if they can't find the way to do something via a quick google search, then either that something is impossible to do or, at least, no one else can do it either.

Last edited by ridethecliche; 05-26-20 at 10:48 PM.
ridethecliche is offline  
Old 05-26-20, 10:34 PM
  #60  
RChung
Senior Member
 
Join Date: Sep 2009
Posts: 1,599
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 367 Post(s)
Liked 175 Times in 74 Posts
Originally Posted by ridethecliche View Post

Horses for courses. We're on a road cycling forum, not an elite athlete or even the racing sub here.
People aren't constants. Their interests change, and it's hard to know who's going to end up interested in something they aren't interested in right now. There do appear to be some riders here who race, or raced; there are probably some riders here who eventually will race. Road cycling is a pretty big tent--nonetheless, the majority of road riders don't really want a power meter and of those who do, most have pretty simple needs.

That said, there's a common misconception that only pros need accurate power data. I've had a chance to talk with people from some pretty big pro teams. The irony is that many pro teams actually don't get very high quality data, or don't have time to do much with the data they do get. Sometimes their sponsors dictate what bikes they'll ride, what tires they'll use, what clothes they'll wear, and what power meters they'll use. When Sky/Ineos went to the velodrome for aerotesting, they mounted a different brand of power meter on the bikes even though in races they were sponsored by Stages. Conversely, some amateurs absolutely need high quality data because they don't have all of the support systems that pros have, so they have to rely on the data they collect. It's both kinda impressive and kinda amusing to look at the things that amateurs figured out first that were subsequently adopted by pro teams.
RChung is offline  
Likes For RChung:
Old 05-26-20, 10:54 PM
  #61  
ridethecliche
Village Idiot
 
Join Date: Jan 2007
Location: Dirty Jerz
Posts: 20,422
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 66 Post(s)
Liked 57 Times in 42 Posts
Originally Posted by RChung View Post
People aren't constants. Their interests change, and it's hard to know who's going to end up interested in something they aren't interested in right now. There do appear to be some riders here who race, or raced; there are probably some riders here who eventually will race. Road cycling is a pretty big tent--nonetheless, the majority of road riders don't really want a power meter and of those who do, most have pretty simple needs.

That said, there's a common misconception that only pros need accurate power data. I've had a chance to talk with people from some pretty big pro teams. The irony is that many pro teams actually don't get very high quality data, or don't have time to do much with the data they do get. Sometimes their sponsors dictate what bikes they'll ride, what tires they'll use, what clothes they'll wear, and what power meters they'll use. When Sky/Ineos went to the velodrome for aerotesting, they mounted a different brand of power meter on the bikes even though in races they were sponsored by Stages. Conversely, some amateurs absolutely need high quality data because they don't have all of the support systems that pros have, so they have to rely on the data they collect. It's both kinda impressive and kinda amusing to look at the things that amateurs figured out first that were subsequently adopted by pro teams.
That's not at all surprising to me though. I always figured that pros just rode and amateurs and enthusiasts had more time to tinker.

I mean... I was perfectly content to send my SRM back once a year for them to make sure things were all nice and dandy. I just wish it didn't cost as much as it did back in the day.

In any event. I'm glad you're still here. Looking forward to learning how far things have come in the last, oh I don't know... decade!

As a side note, wasn't the crank based issue only on the newer shimano cranks? I know stages had a helluva time getting things dialed in. A lot of the folks that jumped into the water afterwards had to keep pushing back dates for releases because there was variation between crank arms that they just couldn't account for. AFAIK, folks found it easier to do crank based on carbon cranks.

FWIW, I ended up going with the specialized branded 4iiii left sided PM because things looked decent for my needs in this snapshot in time. Subject to change of course!
ridethecliche is offline  
Old 05-26-20, 10:56 PM
  #62  
Seattle Forrest
Senior Member
 
Seattle Forrest's Avatar
 
Join Date: Mar 2010
Location: Seattle, WA
Posts: 21,117
Mentioned: 66 Post(s)
Tagged: 0 Thread(s)
Quoted: 11817 Post(s)
Liked 3,129 Times in 1,764 Posts
Originally Posted by RChung View Post
That's how I found out that accelerometer-based cadence is noisier than reed-switch-based cadence.
​​​​​​I could have told you that. Looking at the stats from a ride where the min and max are, sometimes my Vectors say I did 200+ rpm for a second. The rest of the file is normal. Last set I had did it too.
Seattle Forrest is offline  
Old 05-26-20, 11:15 PM
  #63  
Seattle Forrest
Senior Member
 
Seattle Forrest's Avatar
 
Join Date: Mar 2010
Location: Seattle, WA
Posts: 21,117
Mentioned: 66 Post(s)
Tagged: 0 Thread(s)
Quoted: 11817 Post(s)
Liked 3,129 Times in 1,764 Posts
Originally Posted by PoorInRichfield View Post
In reading the directions, I find the first step a little humorous in light of my bathroom scale example from earlier ...

Since a power meter doesn't come with a 25 pound weight (mass), everyone who wants to make sure their Garmin power pedals are calibrated needs to go buy a 25 pound weight/mass. [snip for brevity]

If that's true, that group of people does not need $1k+ left/right mega top-end power meters. They'd benefit just as much with an inexpensive power meter that's consistent with itself. Racers and highly competitive triathletes are a different story, but I bet that most athletes that don't depend on cycling to pay the bills would do just fine training with a low-end power meter as long as it gives consistent readings with itself over time.
Your gym will lend you a weight overnight. Depending on the gym you go to, maybe over the weekend.

Why do triathletes need more accuracy than anyone else? A typical (full not one sided) meter today has a maximum error of 1.5%, that's close enough for training purposes. Am I missing something?

Strava power is free. Since that's available, I think if you're going to pay $400 to $1,000 for a watt meter, you deserve and should get something that can accurately measure your power. That's just on principal.

You can use a power meter like a wind tunnel to refine your position on the bike, or figure out what clothing is most slippery. Coincidentally, @RChung came up with a method to do just that and it's in an excellent, free product called Golden Cheetah.

Finally, there are people who use power meters for diet purposes. Some are hefty folks losing weight, I know a health nut that does it too. Being off by say 20% would lead to really bad results. He's worried about eating too little and hurting his fitness.
Seattle Forrest is offline  
Old 05-26-20, 11:38 PM
  #64  
RChung
Senior Member
 
Join Date: Sep 2009
Posts: 1,599
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 367 Post(s)
Liked 175 Times in 74 Posts
Originally Posted by ridethecliche View Post
As a side note, wasn't the crank based issue only on the newer shimano cranks?
On the newer 4-arm asymmetric right hand cranks, yes. But that means Ultegra and Dura-Ace (and possibly the newest 105??). It turns out, ironically for this part of the thread, that the RHS cranks are not only inaccurate but inconsistent.
RChung is offline  
Old 05-27-20, 12:01 AM
  #65  
RChung
Senior Member
 
Join Date: Sep 2009
Posts: 1,599
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 367 Post(s)
Liked 175 Times in 74 Posts
Originally Posted by Seattle Forrest View Post
​​​​​​I could have told you that. Looking at the stats from a ride where the min and max are, sometimes my Vectors say I did 200+ rpm for a second. The rest of the file is normal. Last set I had did it too.
Yeah. It's easy to spot and handle anomalies like a second of 200+rpm. What's harder is spotting errors that are *almost but not quite* right. Elsewhere there's a discussion going on about pedaling technique. In order to measure that, you need pretty precise measurements of crank rotational speed and position.


Originally Posted by Seattle Forrest View Post
A typical (full not one sided) meter today has a maximum error of 1.5%, that's close enough for training purposes. Am I missing something?
That 1.5% error only refers to "design" errors, not usage errors. If you forget to tare the torque, you can absolutely end up more than 1.5% away from the true value. But even ignoring that, FTP is a single value. Sometimes what you may be interested in is how different values compare with each other -- for example, you may want to know how your 5 second or 1 minute power compares with your 20 min power. Riders who have measured it often say that they notice L/R asymmetry that varies with power level. If your L/R asymmetry varies with power, then the error around your 1 minute power may be different than the error around 20 minute power or 1 hour power.

In addition, my way of estimating rolling and aero drag depends on the relationship between speed and power at different levels. Aero drag varies with the square of air speed while rolling drag is constant with speed. So inaccuracy even though you have consistency doesn't make estimating CdA easy; it throws off the relationship between forces that're constant with increasing speed and those that are quadratic with increasing speed. That's why I don't (can't) recommend a one-sided system if you ever intend to use it for measuring aero drag. (But not many riders are interested in doing that anyway).

Last edited by RChung; 05-27-20 at 07:50 AM.
RChung is offline  
Likes For RChung:
Old 05-27-20, 09:51 AM
  #66  
colnago62
Senior Member
 
Join Date: Aug 2014
Posts: 1,896
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 476 Post(s)
Liked 146 Times in 93 Posts
Originally Posted by Seattle Forrest View Post
Your gym will lend you a weight overnight. Depending on the gym you go to, maybe over the weekend.

Why do triathletes need more accuracy than anyone else?
I know why they need very accurate numbers, at least according to the coaches at TrainerRoad. It is because working too hard in the ride section has a dramatic effect on the run section.
colnago62 is offline  
Likes For colnago62:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off


Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service - Do Not Sell My Personal Information

Copyright © 2018 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.