woodcraft

Recently I did some one legged drills on the rollers and discovered that my pedal stroke has gone to crap.

Possible causes:
- not enough miles/ too much sitting
- too much spin class
- too much reading on BF how stomping is best way to pedal

Anyway, this led to thinking how an inefficient pedal stroke e.g. forward leg is pushing the trailing leg up, would show up as higher power output with pedal or crank based power measure.

So would comparing that to spider or hub based measure show relative pedaling efficiency?

HTupolev

Not necessarily. I would think that the strain gauge would be able to read the weighting of the foot on the upstroke as negative power.

It could contribute to single-sided inaccuracy, like if one leg is resisting the cranks far more than the other. But that's not really any different from how single-sided measurement inaccuracy is in general.

Seattle Forrest

A specific type of pedal stroke would lose more power between the pedals and rear hub than others? That doesn't seem right to me.

I've always assumed pedal meters don't see you pulling up on the pedal, but I don't know.

79pmooney

I've always wondered about that, A true measurement of useful power would only count that part of the force on the pedal spindle that is perpendicular to the crank and add it in the direction of travel and deduct that which opposes. How would you do that? Now it would be useful to also measure forces in line with the crank and in the opposing direction and document that, but as wasted energy.

That is one reason I like that crude analog device that has been used by racers, especially for early season training for well over a century - the fix gear. It very effectively punishes the user for pedal strokes with wasted effort (to a radical extent when you try to go down a fast hill!) I was told many years ago as a first year racer to set my second bike up fix gear to learn to pedal more smoothly, I:m still thanking that vet. (Racing veteran, not military to my knowledge.)

79pmooney

Hub based power is the actual power the rear wheel sees, almost exactly what you are applying to the road, Very simple. Chain tension times cog diameter times RPM. Pedal based depend on what is measured and how. The full force times distance that you apply with your feet? That would be calories expended but if those forces oppose themselves, the hub never sees it. If the forces are in line with the crank, ie radial, then it isn't even work since that radius is fixed. Just wasted force, like standing holding a barbell. Now a pedal power meter could take the total force on the pedal, multiply it by the crank arm length and RPM and call that the work you have done, but that is not correct, (As I said int e post above, I always wondered how pedals "see" the force direction and account for it since there is only a loose association with pedal orientation and crank orientation, (Heel up, heel down, "ankling" etc.)

Edit: In the post above I talked of the club vets telling me to ride fix gear. They also stressed pedaling circles, ie consciously putting all our energy into the circle the pedals make and no more. Yes, on race day that was far from out thoughts. But concentrating on those circles over many miles of fix gear in March paid of big time later in the season. (I also used to do a long day mid-week during the season. If it was raining, I checked the weather, picked a town 50 miles upwind and rode the fix gear there for lunch. Turned around and spun home downwind on tired legs. (This also taught my legs to relax and recover at 100 RPM+. Very useful racing skill.

Ben

HTupolev

I'd be surprised if the exact angle doesn't involve some guesswork, but pushing down versus pulling up at any particular point in the stroke should be detectable as long as the power meter correctly identifies the overall sequencing of the pedal stroke (which should be possible in pretty much any situation where the rider isn't deliberately trying to confuse the power meter).

Also, weighting a pedal on the upstroke is not necessarily any more calories expended. If your leg is actually using its muscles to push down during the upstroke, sure, but that's bizarre behavior and certainly not implied. Weighting the pedal merely implies that it hasn't been fully unweighted! And there's hardly strong evidence that it's more energy-efficient to overcome this fully with the hip flexors versus having the downstroke of the opposite leg contribute.

Carbonfiberboy 

Did DC Rainmaker ever experiment with this? Just need both hub and pedal meters. Anyone? Wouldn't take but a few minutes to see what happens when one changes pedal stroke in different ways. There are several hypotheses here. Only experiment can sort them.

rubiksoval

Why do you think it's gone to crap and why do you think it's inefficient?

Are you bouncing out of the saddle ?
Are you hurting yourself when pedaling?
Are you putting out less power at a given effort/hr despite being having better fitness?

If not, I assert this is a non-issue and your time would be better spent on simply using both legs to produce more power. Because frankly, that's how you'll ride the bike 99.99% of the time, barring a heinous cramp or injury or a pedal falling off or something. 1 legged riding on rollers has to be one of the most useless things I've heard of when it comes to actually riding a bike faster.

RChung

Pedal power meters need to be calibrated in a certain position (either with the crank arm straight down, or the crank arm exactly horizontal), with no pressure on the pedal so they can figure out which direction is down. There are acclerometers and strain gages oriented on the pedal axle so that once it knows what direction is down it can figure out whether applied force vectors are in a "useful" direction; that, btw, is why the Garmin Vector pedals are called "vectors."


The original Vectors, and Power Tap P1 pedals, could measure elapsed time until the pedal made a complete revolution, so they could get cadence. They also could measure the force vectors so could get the average force. Then they calculated power by getting the product of average force and average pedal rotational speed.


There's a slight problem, however: there are certain products that alter the pedal rotational speed. Chief among them is elliptical chain rings, like Osymmetric. It turns out that if you assume constant rotational speed at the average speed, and you multiply it by average pedal force, then when you use an elliptical chain ring that purposely alters the rotational speed, you can bias the power calculation. --> This wasn't specific to pedal meters <--, the SRM power meter also was susceptible to this. How much was the bias? It depends on the eccentricity of the ring but we found that it explained the "additional power" that Osymmetric and Rotor and other elliptical ring manufacturers claimed. (The puzzle was that they claimed 2 or 3% more power, but we didn't see a corresponding increase in speed).


This is a long explanation, but it turns out that when we put elliptical rings on a bike that was simultaneously fitted with an SRM and a Power Tap rear hub, you could see a difference in reported power. The PT hub was rotating at a faster rate than the bottom bracket (as long as we were in a gear ratio higher than 1) so there was less variation in rotational speed in the hub. As an aside, the SRM-PT comparison let us estimate drive train losses. The elliptical ring thing isn't actually a loss -- it was that the SRM's power calculation was being fooled.

Seattle Forrest

​​​​​​The most important thing to know is pedal based power meters are as accurate (1-2%) as anything else. With circular chain rings.

I've only used Vectors, any time you touch them with anything but a cleat, you have to go through a procedure to set/calibrate the angle they're installed.

Anyway, back to the original question. Say I'm applying 300 watts to the pedals, in a weird stroke, we're asking if only 200 (or whatever) watts reach the hub and road. Where do the rest go - as a result of pedal stroke? There will be drive train losses, that's why pedals read slightly higher than hubs, but we're taking about whether a pedal stroke can increase the loss. And I don't see how it could?

Trakhak

Back when I raced, I once overheard a teammate complimenting a guy on another team on how smoothly he rode. After the race, I asked Greg what that was about. He said, "Some guys take it to heart and start concentrating on pedaling even more smoothly. I like psyching people into wasting energy that way."

DaveSSS 

I do a lot of climbing in Colorado. I've been neglecting my brief pull on the up stroke. Today I did 25 miles of climbing on my 53 mile route and noticed an increase in cadence, going up a 16% stretch, by using the upstroke pull. A higher cadence under a heavy load means more power. I no longer use a power meter. I'm too old for it to matter.

HTupolev

Imagine if, instead of unweighting the foot on the upstroke, you continued to push downward with your leg. This would increase the resistance for the other leg's downstroke, and cause your body to have to output a total increased power, but it would not result in more power reaching the rear wheel. Woodcraft is wondering whether a pedal-based or crank-based power meter would see this sort of thing as a higher power delivery number.
(I think his underlying concern is that the not-pulling-up-on-the-upstroke advocates are artificially inflating the power numbers of not-pulling-up form.)

The point I was making in my first post was that a properly-designed power meter should recognize downforce during the upstroke as subtracting from the power delivery to the drivetrain, so this isn't necessarily an issue.

Quote:

The most important thing to know is pedal based power meters are as accurate (1-2%) as anything else. With circular chain rings.

I wasn't questioning that, I was trying to guess at potential ways in which a power meter could be fooled in lieu of Woodcraft's original question.

Although if they have good orientation sensors within them as RChung notes, the issue is moot anyhow.

woodcraft

My question is first about pedaling mechanics and less about accuracy of power meters.

If you are still on the power stroke past bottom center, or (in my case) needing extra force to get the trailing leg up and over the top of the stroke,

that is power being produced, but it is not moving the bike forward. Compare to brake dragging or incorrect timing in a car.

So two cyclists could produce the same amount of force but the one with efficient mechanics would go further/faster.

And second, would the power meters show some or all of the difference? Maybe it's true that pedal/crank units account for all of that.

woodcraft

No, I'm putting out less power and having worse fitness.

Not a regular thing, but I've done one legged drill various times in the past and could continue for an extended time,

but this time it was like starting the power stroke before TDC, & stalling out.

HTupolev

I brought up the hypothetic example of pushing downward on the downstroke to help explain the concern regarding power meter behavior. I wasn't envisioning any kind of normal pedaling form.

It takes power to lift the leg on the upstroke, regardless of whether this is fully accomplished by the rising leg's hip flexors or whether the opposing leg's downstroke also contributes. There's no obvious reason to expect that a form in which the opposing leg helps lift the rising leg is necessarily less efficient.

Seattle Forrest

Sorry if I'm digressing too much. Point was that we can trust pedal power meters. We know they agree very closely with other kinds.

Here are Garmin's notes on their Vector pedals:

Understanding Torque Effectiveness and How Is It Calculated by Vector Power Meter Pedals

The Torque Effectiveness (TE) metric is calculated on dual sided Vector Power Meter Pedals by analyzing the positive (foward) and negative (backward) torque applied to the crank over each revolution.

The following diagram shows a typical power curve for one crank arm, where P+ represents the positive power applied to the crank and is the sum of the instantaneous power measurements. Similarly, P- is the sum of the negative instantaneous power measurements. P- is typically produced when the backstroke of one leg is opposing the downstroke of the other leg.



https://support.garmin.com/en-US/?fa...lu1PNvHK6eTm98

Carbonfiberboy 

Yet oddly enough, it works. I think it helps a cyclist in three ways:
1) OLP forces one to focus on the tangential application of force, all the way around the circle.
2) OLP strengthens the hip flexors. Studies have shown that unweighting the backstroke pedal saves energy by decreasing the downstroke pedal effort. Strong hip flexors also increase OOS power.
3) During the exercise, it forces one to make an even application of force to the pedal so as to maintain pedal speed going into the backstroke. That's the secret to OLP, not so much pulling up as it is pedal speed and unweighting. One only needs to apply enough up-force on the backstroke to keep the chain taut, allowing the wheel to slow a bit. Instead of the usual absolutely steady whiirr on the rollers, one hears a definite change of note but only on the backstroke. Note the similarity to the results in the Pezcycling study below.

As you say, there's a good argument to be made that all one has to do is simply focus on power output. That said, there is also evidence the finesse during the pedal stroke matters. It is to me intuitive that lowering peak forces during the pedal stroke prolongs one's high power efforts. And in fact, pro cyclists do just this The whole "no one pulls up" thing is a red herring. Of course no one pulls up. The question is rather "how hard do you push down?"

See what the pros do: https://pezcyclingnews.com/amp/toolb...our-technique/
There's also the study to which you allude involving counterweighted pedaling: https://www.researchgate.net/publica...keletal_muscle

If one googles "cyclist one legged pedaling," there are many suggestions of doing OLP out on the road with both feet clipped in. thus allowing one to imitate the couterweighted pedaling in the above study.

I limit my OLP to 1 session/week, which is on a recovery day. As has been pointed out, OLP produces less than half the cardio load of equivalent force production during double legged pedaling. Thus I put out about 1/2 my FTP power during OLP, and only for 2 minutes per leg-interval before my hip flexor gives out.. This is a feature, not a bug because it gives us a method to maximally exercise our less-used leg muscles without adding much to TSS.

A comparatively even force application to the pedals is has been one of my little tricks which has enabled me to ride with more talented and younger riders.

Carbonfiberboy 

I think the Pez study I posted answers this question.

rubiksoval

Mind games.

rubiksoval

Continue training and as fitness improves, so will power. Pretty cut and dry.

woodcraft

And pretty much refutes the "stomping" concept.

rubiksoval

No, it doesn't.

"Stomping" simply means more forceful power on the downstroke. It doesn't (and isn't) a jerky, static motion. It's simply how you pedal.

It's an idea to counter the erroneous notion of "equally applying power throughout the pedal stroke", which is impossible.

woodcraft

They said that the pros had lower peak force and power applied at wider range of the pedal circle, relative to the non-pro elites and the amateurs,

so the highest performers "stomped" less.

"Professionals also had tendencies towards a lower maximal torque than elite and club cycles at all three power outputs of 200, 250, and 300 W. combined with the above finding of greater positive force proportion, this suggests that the professional cyclists had a smoother and more even pedal stroke, with less need for a single massive peak burst of torque."

Nobody anywhere near here is talking about equal power throughout the pedal stroke.

Carbonfiberboy 

It doesn't really. Note that the max watts in that pro study was 300. We don't know how they pedal over 300. But from my own experience, to stay with the group seated at double my FTP, I increase my ankle flexion, start the power stroke earlier and hold it past the bottom. Which could be because I'm just not that muscularly strong. My max ATG squat is only 1.2 X body weight, so I can't hit 1000 watts by stomping. My max 30" watts is only ~500w at 74. Used to be a lot more 10 years ago. etc. That said, for sure as an endurance cyclist, what the pros did in that study is, uh, duh, correct in my experience.