Planing

This really deserves its' own thread. For anyone interested, the articles being referred to he

https://www.renehersecycles.com/what-is-planing/

and he

https://www.renehersecycles.com/the-...cs-of-planing/

On Saturday, February 13, 2021 at 6:51:30 AM UTC-5, Axel Reichert wrote:
Frank Krygowski writes:

I'm pretty skeptical of his concept of "planing" regarding bikes -
i.e. that there's an undefined property in certain frames that makes
them significantly faster than somewhat similar frames of identical
weight with identical components. His justification for that seems
entirely anecdotal.

According the the second link above, the conclusions were drawn from a double-blind study.

"We know that riders can put out 12% more power on bikes with optimized flex characteristics. Our observations during our double-blind test – where we rode bikes that were identical except one was stiffer than the others – are consistent with the idea that the best frames allow us to input more energy with less fatigue. "

Not anecdotal, but empirical.

Yes. When I (trained as a Finite Element Analyst) first read about this,
my bull**** detector was heavily triggered. While it is clear (contrary
to folk wisdom) that the energy put into a flexible frame does not
vanish (damping in metal frames is extremely low), I strongly doubt that
a frame stiffness "tuned" to the cadence of the rider will make a bike
"perform better". Here is why I think so:

First, the cadence is not fixed but varies (we are talking about
climbing here) between, say, 40/min and 80/min, depending on slope,
(whimsical) gear choice, mood of the day, etc.

Second, the stiffness (or first relevant eigenfrequency) even of "soft"
frames is so high that upon unloading they return to the undeformed
configuration almost immediately, definitely NOT within a time scale of
a second (which would match a cadence of 60/min). The pedalling is slow
in comparison to the fast spring-back, so there is no point in "matching"
here.

While in fact it was shown both with FEA and experiments that a deformed
frame will return the stored energy by helping turn the crank during
unloading, this result kind of trivially follows from first
principles. Unfortunately, the dynamics of the process was ignored, and
this is the crucial point if you want to achieve some "matching" between
riding style and frame stiffness.

I don't think they're ignoring the system dynamics. What you seem to be describing is the momentary response of the spring characteristic, i.e. measuring the response from a single input at the bottom bracket (feel free to correct me if I have that wrong). The thing that needs to be understood is that the frame will respond to an input at the bottom bracket very differently under the weight of the rider (quantifiable, but extremely variable), and that the pedaling dynamic is cyclic, which you allude to with

"deformed frame will return the stored energy by helping turn
the crank during unloading"

but taking the entire system into account, as the returned energy from the flex is added to the energy input from the rider in the same vector, the 'period' of the returned energy transient changes dramatically. Given the damping of the rider weight plus the contribution of the rider input, I don't think it would be unreasonable to see a resonance sympathetic to _that_ given riders' pedaling dynamic which results in more efficient power transfer. We aren't taking about a perpetual motion machine, just energy feedback that overcomes some of the losses.

I think the most valid point you make is:

the cadence is not fixed but varies (we are talking about
climbing here) between, say, 40/min and 80/min, depending on slope,
(whimsical) gear choice, mood of the day, etc.

For most applications, trying to _design_ a bike that matches the dynamics of a certain rider is an exercise in futility, and Heine even says as much. In a fixed situation, such as a flat time trial, certainly a bike designed that induces sympathetic frame flex _might_ help. However, the situational variables are why racing favors a very stiff very responsive bike, whereas touring or ultra-marathoning favor something more compliant.

My own anecdote is that I raced a Merlin Road (Tom Kellog design) for many years, it was a step up from the Basso Gap I started with. I love the bike, it rides fantastic. When I ride it now, I feel like I've 'come home'. However, some time ago a bought a Scott CR1 Saunier Duval edition. It very clearly upped my game, to the point where my teammates would complain when I rode it during our training rides: "****, he rode he fast bike today". In races I went from struggling to follow an attack, to making successful attacks.. I ride the merlin for my endurance training rides, but no way in hell am I going to go with anything but the Scott in a race. Yes, it makes that much difference. To be clear, I'm agreeing with your point (as I understand it) that there is too much variability in cycling conditions to really be concerned about Planing.


So in my opinion the effect in theory is there, but a frame that would
properly be tuned to the (typical) cadence of the rider (Jan likes to
employ a trampoline analogy, which I am sceptical of, because here the
time scales match better) would be so soft that even the most
retro-grouchy steel advocates would not dare to speed downhill on such a
wobbly piece of metal.

I think the analogy to the trampoline wasn't very appropriate, but you have to remember the target audience of those articles. I think the real danger here is the possibility of negative feedback - once you start designing frames that show a sympathetic resonance, it's not unreasonable to show that a certain rider with a certain pedaling dynamic might experience negative feedback - the returned energy coming back at the _wrong_ time and damping the input from the rider. The stiffer the bike, the less likely this is to happen.


Here are a couple more resources that may help understand the issues for those not well-versed.

https://web.archive.org/web/20060214.../Frameflex.htm

https://thebicycleacademy.org/blogs/...lex-gcn-tech-1

On 2/13/2021 9:10 AM, wrote:
This really deserves its' own thread. For anyone interested, the articles being referred to he

https://www.renehersecycles.com/what-is-planing/

and he

https://www.renehersecycles.com/the-...cs-of-planing/

On Saturday, February 13, 2021 at 6:51:30 AM UTC-5, Axel Reichert wrote:
Frank Krygowski writes:

I'm pretty skeptical of his concept of "planing" regarding bikes -
i.e. that there's an undefined property in certain frames that makes
them significantly faster than somewhat similar frames of identical
weight with identical components. His justification for that seems
entirely anecdotal.

According the the second link above, the conclusions were drawn from a double-blind study.

"We know that riders can put out 12% more power on bikes with optimized flex characteristics. Our observations during our double-blind test – where we rode bikes that were identical except one was stiffer than the others – are consistent with the idea that the best frames allow us to input more energy with less fatigue."

Not anecdotal, but empirical.

Yes. When I (trained as a Finite Element Analyst) first read about this,
my bull**** detector was heavily triggered. While it is clear (contrary
to folk wisdom) that the energy put into a flexible frame does not
vanish (damping in metal frames is extremely low), I strongly doubt that
a frame stiffness "tuned" to the cadence of the rider will make a bike
"perform better". Here is why I think so:

First, the cadence is not fixed but varies (we are talking about
climbing here) between, say, 40/min and 80/min, depending on slope,
(whimsical) gear choice, mood of the day, etc.

Second, the stiffness (or first relevant eigenfrequency) even of "soft"
frames is so high that upon unloading they return to the undeformed
configuration almost immediately, definitely NOT within a time scale of
a second (which would match a cadence of 60/min). The pedalling is slow
in comparison to the fast spring-back, so there is no point in "matching"
here.

While in fact it was shown both with FEA and experiments that a deformed
frame will return the stored energy by helping turn the crank during
unloading, this result kind of trivially follows from first
principles. Unfortunately, the dynamics of the process was ignored, and
this is the crucial point if you want to achieve some "matching" between
riding style and frame stiffness.

I don't think they're ignoring the system dynamics. What you seem to be describing is the momentary response of the spring characteristic, i.e. measuring the response from a single input at the bottom bracket (feel free to correct me if I have that wrong). The thing that needs to be understood is that the frame will respond to an input at the bottom bracket very differently under the weight of the rider (quantifiable, but extremely variable), and that the pedaling dynamic is cyclic, which you allude to with

"deformed frame will return the stored energy by helping turn
the crank during unloading"

but taking the entire system into account, as the returned energy from the flex is added to the energy input from the rider in the same vector, the 'period' of the returned energy transient changes dramatically. Given the damping of the rider weight plus the contribution of the rider input, I don't think it would be unreasonable to see a resonance sympathetic to _that_ given riders' pedaling dynamic which results in more efficient power transfer. We aren't taking about a perpetual motion machine, just energy feedback that overcomes some of the losses.

I think the most valid point you make is:

the cadence is not fixed but varies (we are talking about
climbing here) between, say, 40/min and 80/min, depending on slope,
(whimsical) gear choice, mood of the day, etc.

For most applications, trying to _design_ a bike that matches the dynamics of a certain rider is an exercise in futility, and Heine even says as much. In a fixed situation, such as a flat time trial, certainly a bike designed that induces sympathetic frame flex _might_ help. However, the situational variables are why racing favors a very stiff very responsive bike, whereas touring or ultra-marathoning favor something more compliant.

My own anecdote is that I raced a Merlin Road (Tom Kellog design) for many years, it was a step up from the Basso Gap I started with. I love the bike, it rides fantastic. When I ride it now, I feel like I've 'come home'. However, some time ago a bought a Scott CR1 Saunier Duval edition. It very clearly upped my game, to the point where my teammates would complain when I rode it during our training rides: "****, he rode he fast bike today". In races I went from struggling to follow an attack, to making successful attacks. I ride the merlin for my endurance training rides, but no way in hell am I going to go with anything but the Scott in a race. Yes, it makes that much difference. To be clear, I'm agreeing with your point (as I understand it) that there is too much variability in cycling conditions to really be concerned about Planing.


So in my opinion the effect in theory is there, but a frame that would
properly be tuned to the (typical) cadence of the rider (Jan likes to
employ a trampoline analogy, which I am sceptical of, because here the
time scales match better) would be so soft that even the most
retro-grouchy steel advocates would not dare to speed downhill on such a
wobbly piece of metal.

I think the analogy to the trampoline wasn't very appropriate, but you have to remember the target audience of those articles. I think the real danger here is the possibility of negative feedback - once you start designing frames that show a sympathetic resonance, it's not unreasonable to show that a certain rider with a certain pedaling dynamic might experience negative feedback - the returned energy coming back at the _wrong_ time and damping the input from the rider. The stiffer the bike, the less likely this is to happen.


Here are a couple more resources that may help understand the issues for those not well-versed.

https://web.archive.org/web/20060214.../Frameflex.htm

https://thebicycleacademy.org/blogs/...lex-gcn-tech-1


And rider position is subtly different bike to bike[1]
despite attempts to duplicate it. You don't know why you
ride faster on your Scott, nor do I, but another guy on the
same bike will experience it differently.

[1] Albert Eisentraut postulated a relationship between the
rider's inner ear along the centerline of the steering axis
as critical to the experienced 'ride feel' of a bicycle. One
of many possible avenues for inquiry.

--
Andrew Muzi
www.yellowjersey.org/
Open every day since 1 April, 1971

" writes:

https://www.renehersecycles.com/what-is-planing/
https://www.renehersecycles.com/the-...cs-of-planing/

Thanks for digging up the links for a fresh re-read!

What you seem to be describing is the momentary response of the spring
characteristic, i.e. measuring the response from a single input at the
bottom bracket

Yes, you got that right. Of course the force/moment at the bottom
bracket is not an on/off thing, but rather shows some time
variation. Nevertheless the spring-back will occur on a much smaller time
scale than pedalling, almost instantaneously.

Think of a 20 % slope, too big a gear, you are pedalling out of the
saddle and trying to force the crank around at 40/min. This is clearly
no easy spin and one can imagine the system of bike plus rider being
propelled in bursts, the bike almost coming to a stand-still at the
bottom dead centre. At that point, your power output has almost
completely stopped. The frame (whether stiff or soft) will spring back
within a blink of an eye compared with the time it took for the
down-stroke.

taking the entire system into account, as the returned energy from the
flex is added to the energy input from the rider in the same vector,
the 'period' of the returned energy transient changes dramatically.

Why? The release time should be very short, on the order of 0.2 s, I
would guestimate.

Given the damping of the rider weight

Mass does not dampen. You probably mean that inertia is slowing things
down and thus spreads out the energy release over time? As long as there
is still load on the pedal the spring-back will not be completed?

We aren't taking about a perpetual motion machine, just energy
feedback that overcomes some of the losses.

Sure. And Jan Heine claims that this can occur at the "right point in
time" with the right bike for a rider.

For most applications, trying to _design_ a bike that matches the
dynamics of a certain rider is an exercise in futility

Maybe.

My own anecdote

I guess the ranking according to increasing stiffness was Basso, Merlin,
Scott?

real danger here is the possibility of negative feedback - once you
start designing frames that show a sympathetic resonance, it's not
unreasonable to show that a certain rider with a certain pedaling
dynamic might experience negative feedback - the returned energy
coming back at the _wrong_ time and damping the input from the
rider. The stiffer the bike, the less likely this is to happen.

On re-reading my post before submitting it, I am thinking that the time
scales might no be that different after all: 40/min means 1.5 s per
revolution, or 0.75 s per down-stroke. With my estimate of 0.2 s for the
spring-back, this will cover roughly 45 degrees around the bottom dead
centre. For 80/min still around 25 degrees. This might be quite useful
and maybe I should revise my opinion. (-;

It is a shame I do not have the opportunity to participate in a
double-blind test (or ride nice, different bikes every other week).

Best regards

Axel

On Saturday, February 13, 2021 at 10:29:15 AM UTC-5, AMuzi wrote:

And rider position is subtly different bike to bike[1]
despite attempts to duplicate it. You don't know why you
ride faster on your Scott, nor do I, but another guy on the
same bike will experience it differently.

You would be correct that I also modified my position slightly when I bought the CR1 by dropping and extending the stem, making me lower in the 'cockpit' (based on the input from the bikefitter I met with when setting up the bike). However, I subsequently modified the Merlin as close as I could (the geometry is very different), and still feel better/faster in race situations on the Scott. My guess is that the geometry of the Merlin is more 'slack' - set up like a classic european road racing geometry whereas the Scott is much more steep - so quicker and more responsive handling. When I do an out-of-the-saddle sprint, the difference between the two is dramatic. I feel like there's a delay in the Merlin, whereas the Scott just snaps to attention. All my road bikes now are set up within a few millimeters of each other, but the riding differences are just amazing. I have an older Cube aluminum frame that feels almost as quick as the Scott, but doesn't like the corners as much. I have a Ross Signature 508 (another Kellog design) which rides almost as nice as the Merlin, and I have a Giant Team TCR1 aluminum which I also raced for a while, but is simply too twitchy, especially when the road gets rough (new england winters)

[1] Albert Eisentraut postulated a relationship between the
rider's inner ear along the centerline of the steering axis
as critical to the experienced 'ride feel' of a bicycle. One
of many possible avenues for inquiry.

--
Andrew Muzi
www.yellowjersey.org/
Open every day since 1 April, 1971

On Saturday, February 13, 2021 at 11:09:11 AM UTC-5, Axel Reichert wrote:
" writes:

https://www.renehersecycles.com/what-is-planing/
https://www.renehersecycles.com/the-...cs-of-planing/

Thanks for digging up the links for a fresh re-read!
What you seem to be describing is the momentary response of the spring
characteristic, i.e. measuring the response from a single input at the
bottom bracket
Yes, you got that right. Of course the force/moment at the bottom
bracket is not an on/off thing, but rather shows some time
variation. Nevertheless the spring-back will occur on a much smaller time
scale than pedalling, almost instantaneously.

Hmm....see below


Think of a 20 % slope, too big a gear, you are pedalling out of the
saddle and trying to force the crank around at 40/min. This is clearly
no easy spin and one can imagine the system of bike plus rider being
propelled in bursts, the bike almost coming to a stand-still at the
bottom dead centre. At that point, your power output has almost
completely stopped. The frame (whether stiff or soft) will spring back
within a blink of an eye compared with the time it took for the
down-stroke.
taking the entire system into account, as the returned energy from the
flex is added to the energy input from the rider in the same vector,
the 'period' of the returned energy transient changes dramatically.

Why? The release time should be very short, on the order of 0.2 s, I
would guestimate.

At a cadence of 100, one crank revolution takes ~.6 seconds, so each downstroke takes .3 seconds, and an extended power stroke over 135 degrees takes ..2 seconds. If a legit transient of spring back is ~.2 seconds, I think we're all playing within the same time frame, no?

Given the damping of the rider weight
Mass does not dampen. You probably mean that inertia is slowing things
down and thus spreads out the energy release over time? As long as there
is still load on the pedal the spring-back will not be completed?

Correct. I'm thinking of the similar dynamic of a fixed-gear, where the forward moment of the bike reduces the losses from the sprung and ratcheted drivetrain. If you don't follow though your pedal stroke, the bike will push your legs around until the losses exceed the momentum. As I see it frame flex has the same effect (albeit significantly smaller, unnoticeable in most cases). If you can get a frame that can kick back within the transient period that is similar to the period of your power stroke your can use that sympathetic motion and possible even set up a resonance.

We aren't taking about a perpetual motion machine, just energy
feedback that overcomes some of the losses.
Sure. And Jan Heine claims that this can occur at the "right point in
time" with the right bike for a rider.
For most applications, trying to _design_ a bike that matches the
dynamics of a certain rider is an exercise in futility
Maybe.

My own anecdote

I guess the ranking according to increasing stiffness was Basso, Merlin,
Scott?

Yes. In fact the reason I bought the Merlin was because I had crashed the Basso pretty hard, and took it to a local builder for a frame alignment. I asked him how bad it was after he fixed it and he said the lug brazing on those Bassos give way well before the tubes - he called them soft - so it was an easy fix to tweak it back into shape. He also said if I crashed it again it would likely have to be rebrazed, since they wouldn't take another tweak without fracturing. I remember he said "that's why it rides so smooth, it's a soft bike". But that's another entire discussion on competent frame building - why my Ross/Kellog rides nearly as nice as my Merlin/Kellog.

real danger here is the possibility of negative feedback - once you
start designing frames that show a sympathetic resonance, it's not
unreasonable to show that a certain rider with a certain pedaling
dynamic might experience negative feedback - the returned energy
coming back at the _wrong_ time and damping the input from the
rider. The stiffer the bike, the less likely this is to happen.
On re-reading my post before submitting it, I am thinking that the time
scales might no be that different after all: 40/min means 1.5 s per
revolution, or 0.75 s per down-stroke. With my estimate of 0.2 s for the
spring-back, this will cover roughly 45 degrees around the bottom dead
centre. For 80/min still around 25 degrees. This might be quite useful
and maybe I should revise my opinion. (-;

That's my opinion (as noted above), but I still think it's way too complicated of a characteristic to design for.


It is a shame I do not have the opportunity to participate in a
double-blind test (or ride nice, different bikes every other week).

Best regards

Axel

On 2/13/2021 11:22 AM, wrote:
On Saturday, February 13, 2021 at 11:09:11 AM UTC-5, Axel Reichert wrote:
" writes:

https://www.renehersecycles.com/what-is-planing/
https://www.renehersecycles.com/the-...cs-of-planing/

Thanks for digging up the links for a fresh re-read!
What you seem to be describing is the momentary response of the spring
characteristic, i.e. measuring the response from a single input at the
bottom bracket
Yes, you got that right. Of course the force/moment at the bottom
bracket is not an on/off thing, but rather shows some time
variation. Nevertheless the spring-back will occur on a much smaller time
scale than pedalling, almost instantaneously.

Hmm....see below


Think of a 20 % slope, too big a gear, you are pedalling out of the
saddle and trying to force the crank around at 40/min. This is clearly
no easy spin and one can imagine the system of bike plus rider being
propelled in bursts, the bike almost coming to a stand-still at the
bottom dead centre. At that point, your power output has almost
completely stopped. The frame (whether stiff or soft) will spring back
within a blink of an eye compared with the time it took for the
down-stroke.
taking the entire system into account, as the returned energy from the
flex is added to the energy input from the rider in the same vector,
the 'period' of the returned energy transient changes dramatically.

Why? The release time should be very short, on the order of 0.2 s, I
would guestimate.

At a cadence of 100, one crank revolution takes ~.6 seconds, so each downstroke takes .3 seconds, and an extended power stroke over 135 degrees takes .2 seconds. If a legit transient of spring back is ~.2 seconds, I think we're all playing within the same time frame, no?

Given the damping of the rider weight
Mass does not dampen. You probably mean that inertia is slowing things
down and thus spreads out the energy release over time? As long as there
is still load on the pedal the spring-back will not be completed?

Correct. I'm thinking of the similar dynamic of a fixed-gear, where the forward moment of the bike reduces the losses from the sprung and ratcheted drivetrain. If you don't follow though your pedal stroke, the bike will push your legs around until the losses exceed the momentum. As I see it frame flex has the same effect (albeit significantly smaller, unnoticeable in most cases). If you can get a frame that can kick back within the transient period that is similar to the period of your power stroke your can use that sympathetic motion and possible even set up a resonance.

We aren't taking about a perpetual motion machine, just energy
feedback that overcomes some of the losses.
Sure. And Jan Heine claims that this can occur at the "right point in
time" with the right bike for a rider.
For most applications, trying to _design_ a bike that matches the
dynamics of a certain rider is an exercise in futility
Maybe.

My own anecdote

I guess the ranking according to increasing stiffness was Basso, Merlin,
Scott?

Yes. In fact the reason I bought the Merlin was because I had crashed the Basso pretty hard, and took it to a local builder for a frame alignment. I asked him how bad it was after he fixed it and he said the lug brazing on those Bassos give way well before the tubes - he called them soft - so it was an easy fix to tweak it back into shape. He also said if I crashed it again it would likely have to be rebrazed, since they wouldn't take another tweak without fracturing. I remember he said "that's why it rides so smooth, it's a soft bike". But that's another entire discussion on competent frame building - why my Ross/Kellog rides nearly as nice as my Merlin/Kellog.

real danger here is the possibility of negative feedback - once you
start designing frames that show a sympathetic resonance, it's not
unreasonable to show that a certain rider with a certain pedaling
dynamic might experience negative feedback - the returned energy
coming back at the _wrong_ time and damping the input from the
rider. The stiffer the bike, the less likely this is to happen.
On re-reading my post before submitting it, I am thinking that the time
scales might no be that different after all: 40/min means 1.5 s per
revolution, or 0.75 s per down-stroke. With my estimate of 0.2 s for the
spring-back, this will cover roughly 45 degrees around the bottom dead
centre. For 80/min still around 25 degrees. This might be quite useful
and maybe I should revise my opinion. (-;

That's my opinion (as noted above), but I still think it's way too complicated of a characteristic to design for.


It is a shame I do not have the opportunity to participate in a
double-blind test (or ride nice, different bikes every other week).

Best regards

Axel


That Basso brazed joint analysis is not correct at all.
Either there was a communication problem or he is seriously
ill informed about lugged bicycle frames.

--
Andrew Muzi
www.yellowjersey.org/
Open every day since 1 April, 1971

On Saturday, February 13, 2021 at 12:30:54 PM UTC-5, AMuzi wrote:


Axel
That Basso brazed joint analysis is not correct at all.
Either there was a communication problem or he is seriously
ill informed about lugged bicycle frames.

Ok, edify me.

On 2/13/2021 10:10 AM, wrote:
This really deserves its' own thread. For anyone interested, the articles being referred to he

https://www.renehersecycles.com/what-is-planing/

and he

https://www.renehersecycles.com/the-...cs-of-planing/

On Saturday, February 13, 2021 at 6:51:30 AM UTC-5, Axel Reichert wrote:
Frank Krygowski writes:

I'm pretty skeptical of his concept of "planing" regarding bikes -
i.e. that there's an undefined property in certain frames that makes
them significantly faster than somewhat similar frames of identical
weight with identical components. His justification for that seems
entirely anecdotal.

According the the second link above, the conclusions were drawn from a double-blind study.

"We know that riders can put out 12% more power on bikes with optimized flex characteristics. Our observations during our double-blind test – where we rode bikes that were identical except one was stiffer than the others – are consistent with the idea that the best frames allow us to input more energy with less fatigue."

Not anecdotal, but empirical.

Sorry, I don't find those articles at all convincing. (I probably did
read them when they were first published.) For one thing, they seem to
be very limited tests with very few (only two?) riders, at least one of
whom were very invested in finding some corroboration of his claims.
There's no data table, just one graph for one (which?) rider. There are
no calculations to speak of, but there are plenty of soft,
marketing-style justifications for the theory, especially in the second
article.

Jan's claiming (with no calculation details) a 12% increase in power
output. That is huge! It would yield minutes of lead on a major race
climb like Mont Ventoux. But is there any race data showing teams are
aware of such a huge benefit, that bikes are built to generate that
advantage, that speeds have increased as a result of its discovery? Are
there technical papers explaining and evaluating this, and specifying
how bikes can be designed (using FEA) to generate the effect? AFAIK, all
those are absent.

The most rigid bike in my basement is my old Cannondale touring bike. I
think the most flexible is a light custom Reynolds 531 frame (not built
for me) that I built up into a three speed upright bar town bike. (BTW,
that bike is prone to shimmy when ridden no hands.)

I just did some crude measurements with brakes locked and the crank
horizontal, putting my weight (well, most of it) on the pedal of each. I
got roughly 1/2" vertical deflection with the "soft" bike and roughly
1/4" with the rigid one. That includes tire and fork deflection, so the
1/4" difference is probably not all attributable to the main frame.

I'd say the power stroke is about 5.8" which makes that only about 5%
difference in pedal travel. It's not clear to me how that could generate
a power gain as great as is claimed.

And it occurs to me, if a "flex now, get repaid later" effect really
helped bike speed, there are other more obvious ways to get it. In
keeping with the idea that everything has been tried with bikes, ISTR a
gizmo that put tangential springs somewhere in the drive train - perhaps
between freewheel cogs and the hub? - to "give" then spring forward.
Even if my memory is faulty, would that not be as effective?

Again, 12% power gain is huge. I'd think Bicycle Quarterly would be
carefully measuring the stiffness of each bike tested, correlating it
with measured power output, doing ergometer tests with varying
stiffnesses, helping others (Jim Papadopoulos?) tease out the details of
physics, and make the benefits of "planing" widely known.

None of that is happening. It all seemed to end with "Riding these bikes
fast was more fun, because our legs didn’t hurt" except for "a little
tingling."

I don't doubt that different bikes can feel better or worse, and can
inspire different performance. But I think this "planing" gets filed
with _Buycycling_ magazine's "rigid, yet flexible."

- Frank Krygowski

On Sat, 13 Feb 2021 15:42:39 -0500, Frank Krygowski
wrote:

On 2/13/2021 10:10 AM, wrote:
This really deserves its' own thread. For anyone interested, the articles being referred to he

https://www.renehersecycles.com/what-is-planing/

and he

https://www.renehersecycles.com/the-...cs-of-planing/

On Saturday, February 13, 2021 at 6:51:30 AM UTC-5, Axel Reichert wrote:
Frank Krygowski writes:

I'm pretty skeptical of his concept of "planing" regarding bikes -
i.e. that there's an undefined property in certain frames that makes
them significantly faster than somewhat similar frames of identical
weight with identical components. His justification for that seems
entirely anecdotal.

According the the second link above, the conclusions were drawn from a double-blind study.

"We know that riders can put out 12% more power on bikes with optimized flex characteristics. Our observations during our double-blind test – where we rode bikes that were identical except one was stiffer than the others – are consistent with the idea that the best frames allow us to input more energy with less fatigue."

Not anecdotal, but empirical.

Sorry, I don't find those articles at all convincing. (I probably did
read them when they were first published.) For one thing, they seem to
be very limited tests with very few (only two?) riders, at least one of
whom were very invested in finding some corroboration of his claims.
There's no data table, just one graph for one (which?) rider. There are
no calculations to speak of, but there are plenty of soft,
marketing-style justifications for the theory, especially in the second
article.

Jan's claiming (with no calculation details) a 12% increase in power
output. That is huge! It would yield minutes of lead on a major race
climb like Mont Ventoux. But is there any race data showing teams are
aware of such a huge benefit, that bikes are built to generate that
advantage, that speeds have increased as a result of its discovery? Are
there technical papers explaining and evaluating this, and specifying
how bikes can be designed (using FEA) to generate the effect? AFAIK, all
those are absent.

The most rigid bike in my basement is my old Cannondale touring bike. I
think the most flexible is a light custom Reynolds 531 frame (not built
for me) that I built up into a three speed upright bar town bike. (BTW,
that bike is prone to shimmy when ridden no hands.)

I just did some crude measurements with brakes locked and the crank
horizontal, putting my weight (well, most of it) on the pedal of each. I
got roughly 1/2" vertical deflection with the "soft" bike and roughly
1/4" with the rigid one. That includes tire and fork deflection, so the
1/4" difference is probably not all attributable to the main frame.

I'd say the power stroke is about 5.8" which makes that only about 5%
difference in pedal travel. It's not clear to me how that could generate
a power gain as great as is claimed.

And it occurs to me, if a "flex now, get repaid later" effect really
helped bike speed, there are other more obvious ways to get it. In
keeping with the idea that everything has been tried with bikes, ISTR a
gizmo that put tangential springs somewhere in the drive train - perhaps
between freewheel cogs and the hub? - to "give" then spring forward.
Even if my memory is faulty, would that not be as effective?

Again, 12% power gain is huge. I'd think Bicycle Quarterly would be
carefully measuring the stiffness of each bike tested, correlating it
with measured power output, doing ergometer tests with varying
stiffnesses, helping others (Jim Papadopoulos?) tease out the details of
physics, and make the benefits of "planing" widely known.

None of that is happening. It all seemed to end with "Riding these bikes
fast was more fun, because our legs didn’t hurt" except for "a little
tingling."

I don't doubt that different bikes can feel better or worse, and can
inspire different performance. But I think this "planing" gets filed
with _Buycycling_ magazine's "rigid, yet flexible."

- Frank Krygowski

As I understand this "theory" you are using some of the force applied
to the pedals to bend the frame so essentially this is force that is
not driving the rear wheel and you recover some of this power from a
"springy" frame.

Why not simply build a frame that will not deflect and then all the
power will be driving the rear wheel?
--
Cheers,

John B.

On Saturday, February 13, 2021 at 9:10:34 AM UTC-6, wrote:

"We know that riders can put out 12% more power on bikes with optimized flex characteristics. Our observations during our double-blind test – where we rode bikes that were identical except one was stiffer than the others – are consistent with the idea that the best frames allow us to input more energy with less fatigue. "


What intrigues me is that riders are different. Pedal differently. Some riders my size, height, weight, strength, pedal in a grinding style. Others spin. And others pedal medium. I'm probably in the medium category. But also sometimes I spin and sometimes I grind. Flats and hills I vary my riding style. For me a 58cm frame is perfect. So do I need three different 58cm frames/bikes that are all identical except for this planing characteristic.? I imagine Trek and Specialized would jump for joy if they could sell three identical bikes to each person with the only difference being the frame is hard, medium, soft planing. And if I'm medium riding on the flat and get to a hill that I am going to grind up, do I need a bike change at the bottom of the hill. And if I spin going down the hill do I need another bike change at the top to optimize the spin down. And then I ride medium on the flat so do I get another bike change at the bottom of the hill to optimize the medium riding on the flat until I get to the next hill. I'd need two or three support cars following me all the time to change bikes every few miles.