Cleat-based power metering

velonews.competitor.comquick look garmin vector power meter 188454Open ↗

Spring 2012. $1500. I'll stay with my wired powertap, thank you very much...

bikeradar.comtesting the power2max and look keo power pedals 32678Open ↗

Interesting article on a couple of the pedal-based power systems coming on-line.

Bottom-line for me: sticking with the wired Powertap, thank you very much.

edit: the Power2Max is a crank-based system ala SRM and Quark.

I keep 2 wired PowerTaps aorund. One with a 16-30 for riding on the road. One with an 11-23 for riding on the trainer. Since they are on 2 different bikes, I would need to get 2 sets of pedals and replace my 2 rear wheels.

Selling existing PowerTaps wheels and related components would net about $1200. Buying similar quality wheels would cost about $300. Add in 2 sets of pedals $3000. $2100 is well within my parts budget but ..

My colleague JJ is off to do a retest of that BikeRadar power comparison tomorrow, using a better 'control' environment. I for one am looking forward to the results.

Its a really interesting time for power monitoring systems, and I can't help but think we are not far away from a real price drop as more and more players are entering the market.

I'm writing this from Koolstof Towers whilst waiting (impatiently) for delivery of our first set of Look Power Pedals for stock (due in today), and being honest, I am more excited by this than I thought I would be.

Last week we heard that Garmin's Vector is not expecting to hit the shelves next spring due to ongoing challenges with the crystal based strain guages, so we have decided to ignore the missing ANT+ compatibility of the Look/Polar system and back them moving forward. They look (can you see what I did there?) a lot more snazzy anyway, plus the technology seems far more proven and ready to go.

I like the idea of the Look pedals, I like the idea of measuring leg power independently, I like the weight, transferrability and I also like the fact that the technology is sound, so I am sort of convinced.

That said, when talking to one of my athletes this morning, we ultimately think the answer will be a shoe based system, based on pressure monitoring technology, a simple accelerometer and transmitter that will be moulded directly into the sole of a shoe. You would then have one system for all your bikes, transferrability would not be an issue, and it should be crash resistant. Just need to work out how to transfer the fancy stuff when you get new shoes.

Oh, and yeah, if anyone is after a set of Look Power Pedals drop me a message... As you can imagine I need to offer our athletes first dibs, but we will have one of the only pairs in the UK before the end of January. Those are all allocated to customer apparently, so it'll be late February now before any other free stock is available.

Quoted post said:

Originally Posted by Koolstof [IMG]/img/forum/go_quote.gif[/IMG]

My colleague JJ is off to do a retest of that BikeRadar power comparison tomorrow, using a better 'control' environment. I for one am looking forward to the results.

Its a really interesting time for power monitoring systems, and I can't help but think we are not far away from a real price drop as more and more players are entering the market.

I'm writing this from Koolstof Towers whilst waiting (impatiently) for delivery of our first set of Look Power Pedals for stock (due in today), and being honest, I am more excited by this than I thought I would be.

Last week we heard that Garmin's Vector is not expecting to hit the shelves next spring due to ongoing challenges with the crystal based strain guages, so we have decided to ignore the missing ANT+ compatibility of the Look/Polar system and back them moving forward. They look (can you see what I did there?) a lot more snazzy anyway, plus the technology seems far more proven and ready to go.

I like the idea of the Look pedals, I like the idea of measuring leg power independently, I like the weight, transferrability and I also like the fact that the technology is sound, so I am sort of convinced.

That said, when talking to one of my athletes this morning, we ultimately think the answer will be a shoe based system, based on pressure monitoring technology, a simple accelerometer and transmitter that will be moulded directly into the sole of a shoe. You would then have one system for all your bikes, transferrability would not be an issue, and it should be crash resistant. Just need to work out how to transfer the fancy stuff when you get new shoes.

Oh, and yeah, if anyone is after a set of Look Power Pedals drop me a message... As you can imagine I need to offer our athletes first dibs, but we will have one of the only pairs in the UK before the end of January. Those are all allocated to customer apparently, so it'll be late February now before any other free stock is available.

Thanks for the spam!

Oh, what's not developed about piezo strain gauges? Obviously, you know next to nothing about piezo electric devices. That's ok, so long as you can spam away.

Quoted post said:

Originally Posted by alienator [IMG]/img/forum/go_quote.gif[/IMG]

Thanks for the spam!

Oh, what's not developed about piezo strain gauges? Obviously, you know next to nothing about piezo electric devices. That's ok, so long as you can spam away.

Maybe the 'crystal' based piezoelectric strain gauges are just too passe for him. Maybe he's holding out of the ceramic derivative - just because it's... well... ceramic.

Accelerometer just sounds sooooo much better than strain gauge for expensive bike parts. I reckon you could charge a good $500 more by just using it in the product description.

Hey Tony... just to make it clear, you're sticking with your wired power tap, right?

Quoted post said:

Originally Posted by Koolstof [IMG]/img/forum/go_quote.gif[/IMG]

I like the idea of the Look pedals, I like the idea of measuring leg power independently, I like the weight, transferrability and I also like the fact that the technology is sound, so I am sort of convinced.

1. Beyond basic curiosity, what use is knowing the power of each leg? What actionable intelligence will it provide? Just doubles the number of things that can go wrong.

2. Transferability is trickier and takes longer than for a crank, given the need to set the pedals and sensors up very carefully and no good if you use different pedals on different bikes (e.g. track/MTB/road)

3. The technology has not been tested in the marketplace, so I'll wait and see. It concerns me that in Jeff's test, one pedal unit failed and a Polar specialist is needed to validate proper installation.

4. I ain't changing my pedal/cleats to another brand, and I suspect lots of others won't either

5. These units are hardly cheap and are not compatible with anything other than Polar's limited functionality head units

As for shoe based meters, well that's not much good for me. 😄

Quoted post said:

Originally Posted by Alex Simmons [IMG]/img/forum/go_quote.gif[/IMG]

1. Beyond basic curiosity, what use is knowing the power of each leg? What actionable intelligence will it provide? Just doubles the number of things that can go wrong.

2. Transferability is trickier and takes longer than for a crank, given the need to set the pedals and sensors up very carefully and no good if you use different pedals on different bikes (e.g. track/MTB/road)

3. The technology has not been tested in the marketplace, so I'll wait and see. It concerns me that in Jeff's test, one pedal unit failed and a Polar specialist is needed to validate proper installation.

4. I ain't changing my pedal/cleats to another brand, and I suspect lots of others won't either

5. These units are hardly cheap and are not compatible with anything other than Polar's limited functionality head units

As for shoe based meters, well that's not much good for me. 😄

We'll have to wait and see if people find application for power meters that provide independent measurement of each side's power output. Of what benefit is the SRM scientific? Are there athletes that really need the precision that instrument provides?

Only the Look/Polar pedals require a specific orientation. The Garmin pedals aren't going to require any special setup.

Doubling the number things can go wrong doesn't mean that those things will go wrong. Computers, for example, have increased massively in complexity over their counterparts 10-15 years ago; yet they are arguably more reliable and robust now.

We're all waiting to see how these perform in the real world.

While some might not pay for different cleats, others won't pay to have a hub based power meter or won't pay to get a power meter that has to be sent back to the manufacturer when the battery dies. All power meters come with some compromise.

I think Polar's refusal to work with ANT+ is a mistake, one that Garmin at least isn't making. I've never cared for Polar's exclusive signal and software. Plus, Polar isn't Mac friendly.

The units are hardly cheap, but we haven't seen yet where the price will end up. Let's face it: there's no power meter that's really cheap. The benefit should be that there will be more competition among power meter manufacturers, and that will hopefully drive prices down a bit. If SRM, Power Tap, and Quarq were what everyone wanted in terms of price and function, then everyone that wanted a power meter would have one. Obviously that's not the case, so there is room for more players in the market.

Quoted post said:

Originally Posted by alienator [IMG]/img/forum/go_quote.gif[/IMG]

Of what benefit is the SRM scientific? Are there athletes that really need the precision that instrument provides?

There are some applications where this would be useful, such as track aero field testing.

Quoted post said:

Originally Posted by alienator [IMG]/img/forum/go_quote.gif[/IMG]

I think Polar's refusal to work with ANT+ is a mistake, one that Garmin at least isn't making. I've never cared for Polar's exclusive signal and software. Plus, Polar isn't Mac friendly.

Can't comment on Mac. Not sure how much the fact that Garmin owns ANT+ is in their equation.

Exclusive isn't so bad (original Powertaps and SRM only worked with their own head units), provided the system has the fundamental functionality required. Polar still dip out on things like memory capacity and switching off power recording in preference to heart rate.

Quoted post said:

However, I can see the day when PMs are built directly into the sole of the shoe.

Technology certainly exists, I saw it on tv the other day. This techology could benefit speed skaters, among others.

Quoted post said:

Originally Posted by Koolstof [IMG]/img/forum/go_quote.gif[/IMG] the technology is ...

The reason why alternate power measuring technology is slow to come to bicycles is that the technology is not that easy to implement.

Strain gages work best on simple geometries (rear hubs) under near constant power (not a bicyclist).

---

With a pedal based system, changes in pedaling style will induce all sorts of sytemic errors. A clete based system will do even worse.

Quoted post said:

Originally Posted by An old Guy [IMG]/img/forum/go_quote.gif[/IMG]

The reason why alternate power measuring technology is slow to come to bicycles is that the technology is not that easy to implement.

Strain gages work best on simple geometries (rear hubs) under near constant power (not a bicyclist).

---

With a pedal based system, changes in pedaling style will induce all sorts of sytemic errors. A clete based system will do even worse.

Please specify the systemic errors. Strain gauges work perfectly well on bikes and bike components, and there is no evidence to indicate otherwise. Certainly, a pedal axle has a "simple geometry". There's not much at all that's complex about it, certainly not so complex to make strain measurement difficult. Stain gauges have been well vetted for use as power meters for cycling, and the measurement error for the devices shows as much.

As for the lack of "near constant power" from a cyclist, the only time that is going to be an issue is when the period of the power change is on the order of or less than the relaxation time of the flexed part, or less than or on the order of the integration time of the sensor. Given that those periods or times are so short as to be meaningless to a cyclist, there's no reason to really worry about it.

You've made a lot of claims, but you've yet to offer a scintilla of evidence of any sort. This would be a good time for you to bring up facts about what power modulation frequency or response time a person can exhibit and then relate that info to component relaxation time and sensor integration time.

Quoted post said:

Originally Posted by alienator [IMG]/img/forum/go_quote.gif[/IMG]

Please specify the systemic errors. Strain gauges work perfectly well on bikes and bike components, and there is no evidence to indicate otherwise. Certainly, a pedal axle has a "simple geometry". There's not much at all that's complex about it, certainly not so complex to make strain measurement difficult. Stain gauges have been well vetted for use as power meters for cycling, and the measurement error for the devices shows as much.

As for the lack of "near constant power" from a cyclist, the only time that is going to be an issue is when the period of the power change is on the order of or less than the relaxation time of the flexed part, or less than or on the order of the integration time of the sensor. Given that those periods or times are so short as to be meaningless to a cyclist, there's no reason to really worry about it.

You've made a lot of claims, but you've yet to offer a scintilla of evidence of any sort. This would be a good time for you to bring up facts about what power modulation frequency or response time a person can exhibit and then relate that info to component relaxation time and sensor integration time.

If you have an engineering degree, you should be able to understand the issues I brought up. If you lack an engineering degree, you will not understand the complexity of resolving the issues I mentioned.

Even a person who casually uses a Powertap will understand that hub besed systems have systematic issues. Those issues are documented in part at the Powertap web site. Other brands also have documented issues. The internet search is a much better resource for answering your questions than I am going to be.

Quoted post said:

Originally Posted by An old Guy [IMG]/img/forum/go_quote.gif[/IMG]

If you have an engineering degree, you should be able to understand the issues I brought up. If you lack an engineering degree, you will not understand the complexity of resolving the issues I mentioned.

Even a person who casually uses a Powertap will understand that hub besed systems have systematic issues. Those issues are documented in part at the Powertap web site. Other brands also have documented issues. The internet search is a much better resource for answering your questions than I am going to be.

You haven't answered the question(s). You simply tried to dodge. Further, "systemic issues" can include lots of things, none of which you have pointed out and a lot of which may have nothing at all to do with strain gauges or their application in cycling. If you can't support your questionable assertions, maybe you shouldn't put them out there.

Why don't you just let the readers decide if something has complexity beyond comprehension?

Quoted post said:

Originally Posted by alienator [IMG]/img/forum/go_quote.gif[/IMG]

Why don't you just let the readers decide if something has complexity beyond comprehension?

If you find the topic is within your comprehension, why did you ask the questions? (That was intended to answer your question. It was retorical. I am not expecting an answer.)

If you want a tutor in the subject, I can suggest some. They are expensive. But they are more informative than I am. But they don't resond well to people with the attitude you are displaying. So walk softly around them.

Quoted post said:

Originally Posted by An old Guy [IMG]/img/forum/go_quote.gif[/IMG]

If you find the topic is within your comprehension, why did you ask the questions? (That was intended to answer your question. It was retorical. I am not expecting an answer.)

If you want a tutor in the subject, I can suggest some. They are expensive. But they are more informative than I am. But they don't resond well to people with the attitude you are displaying. So walk softly around them.

Again, you've failed to give concrete reasons for your assertions. You should be able to lay them out with some specificity, assuming you've had the engineering training that you mentioned. Please, give the technical reasons why power meters for cycling are a bad application, why strain gauges used to indirectly measure power are not suitable for said application, and why pedal based power meters will have so many more "systemic errors" than non-pedal based power meters.

How is asking you to be specific about what you are claiming with generalized statements show any attitude? As for why I asked the questions: it's simple. I am curious from a technical point of view as to what specifically is wrong with strain gauge based power meters for cycling such as you claim. It's common for engineers and scientists to have to respond specifically to such questions. It is, after all, how science, and thus engineering, progresses and is inherent in the scientific method.

For a strain gauge, a pedal axle is a perfectly acceptable structure for the mounting and use of a strain gauge, or as in the case of pedal axles and the two current systems, peizoelectric strain gauge. After all, a pedal axle operates in the elastic portion of a stress-strain curve and therefore exhibits linear response. Strain gauges provide linear signal response for a differential element of strain. Foil strain gauges can show drift over a lifespan, but pzts display virtually no drift over a lifetime, a lifetime which is longer than that of foil strain gauges. Pedal axles are easily modeled, and their stress-strain curves are easily calculated. No show stoppers so far. That pretty much just leaves manage the current into a foil strain gauge or managing the current out of a piezoelectric gauge and possibly needing to amplify that current from the pzt. There's nothing difficult about doing that either. So how do bike power meters stack up? Well, strain gauge based power meters typically have an error on the order of ± 2% or so. The scientific SRM has an error of ± 0.5%. For the record, the precision is the sum of all errors in the system. That's not bad at all for something that isn't suitable, in your estimation, for measuring power on a bike. I believe both the Look/Polar power meter and the Garmin Vector power meter both claim ± 2% precision, but I'm not completely sure that is the case. The Garmin, like the SRM and possibly other power meters, can be calibrated using a static torque measurement (The firmware update for all of the compatible Garmin head units now has a procedure for this calibration). I don't know if the Look/Polar unit allows such a calibration. Neither the Look/Polar nor the Garmin have been out long enough for testing to get an accurate read of their real-world precision, and for those interested, absolute accuracy.

Quoted post said:

Originally Posted by alienator [IMG]/img/forum/go_quote.gif[/IMG]

Again, you've failed to give concrete reasons for your assertions. You should be able to lay them out with some specificity, assuming you've had the engineering training that you mentioned. Please, give the technical reasons why power meters for cycling are a bad application, why strain gauges used to indirectly measure power are not suitable for said application, and why pedal based power meters will have so many more "systemic errors" than non-pedal based power meters.

How is asking you to be specific about what you are claiming with generalized statements show any attitude? As for why I asked the questions: it's simple. I am curious from a technical point of view as to what specifically is wrong with strain gauge based power meters for cycling such as you claim. It's common for engineers and scientists to have to respond specifically to such questions. It is, after all, how science, and thus engineering, progresses and is inherent in the scientific method.

1) I never made the claims "power meters for cycling are a bad application, why strain gauges used to indirectly measure power are not suitable for said application."

2) I said I was not about to explain to you what "systemic errors" mean. You are free to consult with the manufcturers or with professionals in the field.

3) This is not a professional engineering/science site. It is a bicycling site. If you want engineering/scientific answers, please go to an engineering/scientific site and bother the people there. You might find someone who is willing to explain my words. You might not.

---

While you use a lot of engineering words, you seem to misunderstand the problem. Perhaps you would be well served by implementing a shoe ot pedal based system. (That must be the 4th or 5th recommendation I have made to help you answer your questions.)

These are your words, old guy:

Quoted post said:

Originally Posted by An old Guy [IMG]/img/forum/go_quote.gif[/IMG]

The reason why alternate power measuring technology is slow to come to bicycles is that the technology is not that easy to implement.

Strain gages work best on simple geometries (rear hubs) under near constant power (not a bicyclist).

---

With a pedal based system, changes in pedaling style will induce all sorts of sytemic errors. A clete based system will do even worse.

Apparently you are unable and lack the knowledge to back up your claims. Ok. You should have just that in the beginning. You can't be faulted for having an insufficient education or lacking the skills to understand the topic. You do dance well, though.

Are you the guy in the center, wearing the gold hat?