Planing

[Too Tall]

The presumed benefit is a bunch of cr@p. The energy which goes into and consequently flexes a frame THAT (stored and soon to be released) energy is not in dispute...like duh insult me. That stored energy we are being lead to believe is returned to make you and the bike fast/more efficent blah blah blah.

Well....stored energy don't pedal the bike. It does not make the chain pull harder against the freehub nor does it emmit a forceful cloud of magic rainbows dust rearward. Kidding about the last one.

The energy (returned) is not in dispute. The premise that it moves the bike is BS.

Comfort on the bike, physical ability to be powerful on a particular bike is the crux of the biscuit. But bikes won't sell themselves I guess.

Hey plane that.

PS Mark Kelly is right, he usually is.

[waterlaz]

Well....stored energy don't pedal the bike. It does not make the chain pull harder against the freehub nor does it emmit a forceful cloud of magic rainbows dust rearward. Kidding about the last one.

The energy (returned) is not in dispute. The premise that it moves the bike is BS.

You are wrong!

It very much depends on the form of the pedal stroke.
If you could pedal by always applying the force exactly in tangential direction (perpendicular to the crank arms) then 100% of the energy returned by the frame would go into moving the bike forward. I guarantee it!
Now in real life it may be that some of the returned energy goes into flexing your legs and hence is effectively lost. But it is kinda hard to estimate the percentage without analyzing the cyclists pedal stroke.

[one60]

fascinating subject...

1. Does the 'spring hypothesis' apply to wheels?

from above..."As is obvious from footage of Sean Kelly sprinting on a Vitus 979, some of the strain in the frame serves to twist the frame and thus move the wheels away from co-planarity, a lateral perturbation in the model referenced above."

2. has Sean Kelly ever weighed in on this topic? Surely he has an opinion about which frameset was fastest?

Grant Risdon

[Andrew R Stewart]

fascinating subject...

1. Does the 'spring hypothesis' apply to wheels?

from above..."As is obvious from footage of Sean Kelly sprinting on a Vitus 979, some of the strain in the frame serves to twist the frame and thus move the wheels away from co-planarity, a lateral perturbation in the model referenced above."

2. has Sean Kelly ever weighed in on this topic? Surely he has an opinion about which frameset was fastest?

Grant Risdon

And if Sean Kelly had an opinion of which frame was the fastest it would still be just that, an opinion. It has been shown over and over that our perception is not always reality or even based in fact. Andy

[David Tollefson]

Pretty neat demonstration of frame strain energy being returned to the rear wheel that Jan sent to folks who subscribe to his emai newsletter. I wish I'd thought of it!

Interesting "demonstration" of a situation that doesn't even come close to mimicking the real-world release of that stored flex.

[false_aesthetic]

Hey-o.

They're riding Boyds.

[Matthew Strongin]

Interesting "demonstration" of a situation that doesn't even come close to mimicking the real-world release of that stored flex.

So you don't ride your bike by holding the brake, stomping on the pedal and then releasing the brake to see how far you roll?

[jclay]

So you don't ride your bike by holding the brake, stomping on the pedal and then releasing the brake to see how far you roll?

No. You wisely substitute friction, aerodynamic drag and climbing hills for the brake! And yes, there's some fore/aft strain energy stored in the SSs & brake arch which contributed to wheel rotation and which wouldn't exist on the road.

And the strain energy isn't increasing the tension on the chain...that's kinda the point. For the same total energy delivered in a given time, the strain energy flow reduces peak chain tension values in exchange for broadening the shape of the tension v pedal position curve. Smoother power delivery and reduced max muscle fiber tension. It's all in the hand drawn curve that some of you guys liked so much a while back! ( Flickr )

[Matthew Strongin]

240h5t.jpg

[David Tollefson]

So you don't ride your bike by holding the brake, stomping on the pedal and then releasing the brake to see how far you roll?

My point is that the deflection is not returned at any point during the pedal stroke that created it, as the experiment suggests. In fact, it's returned when the opposite pedal stroke begins to produce an opposing lateral moment, so the proposed benefit is in actuality a detriment to the opposite pedal stroke.

[jclay]

My point is that the deflection is not returned at any point during the pedal stroke that created it, as the experiment suggests. In fact, it's returned when the opposite pedal stroke begins to produce an opposing lateral moment, so the proposed benefit is in actuality a detriment to the opposite pedal stroke.

And how did you determine this?

[David Tollefson]

And how did you determine this?

At what point is the force that causes the deflection removed?

[e-RICHIE]

Can I get the Cliffs Notes to this issue one more time. Is someone suggesting that a maker can produce a bicycle a certain way such that, pedaling it (a certain) way can actually produce energy that will (conceptually or in reality) make it go forward with some degree of "free speed" (to borrow from downhill skiing and wind tunnel jargon.)? Is that what this is about?

PS I'll add this: no one ever said stiffer is better unless they were of the Mad Men ilk, and few if any with experience in the sport or trade ever believed them. Further, no one ever defines "stiff" to any degree so that leaves the wider conversation brought with subjectivity. The same goes for "flexy". And furthermore, to borrow from an at least decade old or older reference, folks who claim to feel this marginal (sic) gain must be the cycling equivalent of a human mamograph machine.

[Too Tall]

You are wrong!

It very much depends on the form of the pedal stroke.
If you could pedal by always applying the force exactly in tangential direction (perpendicular to the crank arms) then 100% of the energy returned by the frame would go into moving the bike forward. I guarantee it!
Now in real life it may be that some of the returned energy goes into flexing your legs and hence is effectively lost. But it is kinda hard to estimate the percentage without analyzing the cyclists pedal stroke.

Very well said. You get it. Now we psychoanalyze every cyclist and tell them to modify their pedal stroke in order to not get beat up by this returned energy. Surviving the experience will include: efficiency on the bike and learning to rest while going fast. There is no free lunch.

[jclay]

It seems to me that any bicycle structural element which deflects as a direct result of applying tension to the chain by pedalling would be able (required actually) to return strain energy back into the system. That would happen as pedalling torque declines from maximums and I'd expect some phase lag. I see no reason to think the energy flow is unidirectional or that it goes off into the ether.

That behavior would, I think, result in smoother power delivery by reducing the instantaneous chain tension maxima in exchange for applying the energy driving the tension over a slightly longer period. That's what I attempted to illustrate with my cardboard sketch. Same amount of energy input per crank rotation but the delivery modified to reduce the magnitude of maximum impulse while broadening the chain tension v. crank position curve. Smoothing that curve would increase propulsion efficiency (reducing the alternating acceleration/deceleration pairs per power stroke) and reduce the maximum instantaneous muscle fiber tension in the rider and therefore fatigue.

( Flickr )
L
Springs are frequently used to smooth power delivery in industrial machinery.

Nobody is suggesting that more energy is harvested from the system than is added to it. Getting free speed has not been suggested, at least not by those who think there is a case for this characteristic.

Decades ago Cannondale and Klein sold bikes on the basis of asserting that flexible frames “wasted” energy. Their stiffer frames wasted less energy because they were stiff, they said. That notion was widely accepted. Fast forward a few decades and in more recent times when some would suggest that frame flex might be beneficial to propulsion, the commonly voiced opposing view was that frame flex could not contribute to propulsion. This conversation is the prima facia evidence.

Is someone suggesting that a maker....: That's not really part of the assertion. Jan and others (including me) are interested in knowing whether or not some amount of frame flex (and for context less than the stiffer end of the bicycle flexibility spectrum) might be beneficial to propulsion. Obviously everything about the rider and his/her power level bear on the issue. I believe that Jan has made it clear that he finds frames at the more flexible end of this spectrum (thinner wall tubes) improve the results of his higher output efforts. Sorry if I misspoke Jan but I think that's a fair representation of the context. Given enough testing it seems reasonable that correlations could be made between rider (mass, power capability, event, whatever else) and bicycle frame flex ( within the current spectrum of most flexible to least) that would yield improved performance under the circumstances at issue (say match sprint v PBP). I don't see how that should be a controversial idea. There are too many analogies in the sports world to mention but I'll just say tennis racket stringing tension and golf club shaft flexibilities. Performance improvements related to optimizing string tension and shaft flexibility are not questionable. I think that the main difficulty is envisioning how bicycle frame strain energy can be directed into the chain in beneficial ways. It's not as obvious as the racket, club or diving board.

As a qualitative working definition of stiff let's say that the spectrum ranges from whichever bikes are stiffest to those that are most flexible. For the time being we're not discussing frames outside of that range, whatever the actual values happen to be.

While not perfect, I don't think that performing a large number of timed tests over the same courses, with the same people spanning many different events, all massaged by phD statisticians who specialize in knowing how to extract useful information from somewhat noisy data, constitutes a human mammography machine.

[e-RICHIE]

Decades ago Cannondale and Klein sold bikes on the basis of asserting that flexible frames “wasted” energy. Their stiffer frames wasted less energy because they were stiff, they said. That notion was widely accepted.

That's the fallacy here (and there). Who accepted it? Only the uneducated. Or the under-experienced. Or folks who read ad copy and believed it to be true. That the counterpoint to all this (the Mad Ave speak about stiff and about flexy) is simply the Y2K version of it for the same demographic.

And from my perspective (alone) comparing or even contrasting any of this to a diving board or tennis racquet is misplaced. A diving board is only a diving board, as is a tennis racquet. A bicycle is a frame, with wheels, and tires, and tire pressure, and many components, and all of these parts as a whole also depend on the individual quality level of each, as well as how each (and every) part is installed and maintained. And all of these parts flex (or not) too.

[Mark Kelly]

Here's a thought: "planing" could be a thing even if no energy at all was ever delivered from frame recoil to the chain.

I asked myself "How much power does it take to lift the leg in the rising phase of the pedal stroke?". Self, I said, we can calculate that.

As a rough approximation, the lower part of the leg rises by 350mm per stroke while the top part stays level but the weight isn't evenly distributed so lets use 1/4 as the weighting factor.

It takes 12.5 Joules to lift a 15 kg weight* 85mm in an average gravitational field.

At 90 RPM there are 3 "lifts" per second, 12.5 J x 3 s^-1 is 37.5 watts.

Of course there are also 3 falls per second, so the net energy required is zip BUT this is because on each stroke the descending leg gains as much energy as the rising leg loses.

The frame strain energy accumulates in the first phase of the downward stroke as the force applied reaches its peak, then dissipates as the applied force reduces.

It would take a very small phase lag for the rebound strain energy to contribute to accelerating the leg upwards on the return stroke. This phase lag could be supplied by storing the energy in the achilles tendon**.

This would reduce the amount of energy being supplied by the descending leg. If the power being applied to the descending leg by the cyclist remains constant, more of this power would be available to propel the bike via the chain.

Planing, then, could be a matter of harmony: if the strain recovery curve of the frame matches the strain recovery curve of the achilles tendon, it could contribute to forward propulsion simply by reducing the amount of the cyclist's power that is used to lift the ascending leg.

* for the average person, each leg is about 17% of body mass. For the average cyclist it's a bit higher. 15 kg is a reasonable approximaton of my legs so that's what I used.

** This is a well known effect. See, for instance, kangaroos.

[thollandpe]

Mark, aren't the dynamics of a relatively stiff bike frame going to have a much more rapid response than the musculoskeletal system of a kangaroo?

I'd guess that the return of a frame's stored strain energy would be almost immediate as the rider's power stroke ebbs, and that there would be no way it would make it past 180 degrees and help lift the ascending leg.

[jclay]

That's the fallacy here (and there). Who accepted it? Only the uneducated. Or the under-experienced. Or folks who read ad copy and believed it to be true. That the counterpoint to all this (the Mad Ave speak about stiff and about flexy) is simply the Y2K version of it for the same demographic.

Then we all believe that some measure of flexibility, between the stiffest and least stiffest available, contributes to bicycle propulsion? If that were the case a review of the individual postings within this and allied threads would read quite differently than they do.

And from my perspective (alone) comparing or even contrasting any of this to a diving board or tennis racquet is misplaced. A diving board is only a diving board, as is a tennis racquet. A bicycle is a frame, with wheels, and tires, and tire pressure, and many components, and all of these parts as a whole also depend on the individual quality level of each, as well as how each (and every) part is installed and maintained. And all of these parts flex (or not) too.

Its not misplaced at all. Those are clear demonstrations of strain energy (via some measure of flexibility between stiffest possible and way too flexible) employed to enhance performance. That a bicycle has many other components and details of components only serves to increase the complexity of the situation and any attempt at analysis. That the other elements you mention may (or may not) affect performance doesn't depreciate the notion that frame flex can be effective in aiding performance; or in either extreme diminish it.

A quick read of Mark Kelly's thoughts are interesting. I'll need to re-read and digest them though.

[Mark Kelly]

Yes, there's a mismatch but it's not as big as you might think. As near as I can work out the spring constant in my achilles tendon is about 20 kN/m*.

A 570mm frame with the pedals at 125mm from centreline and a frame stiffness of 100 Nm/ degree will flex about 10mm laterally and 2 mm vertically in response to a vertical force of 800N on the pedal. The spring constants are thus 400 kN/m vertically and 80 kN/m horizontally. Since rebound velocity is proportional to the square root of the spring constant, these are mismatches of about 6:1 and 2:1 respectively.

Interestingly, the mismatch is smaller for a less stiff frame.