Frame stiffness

William <[email hidden]> wrote in message
news:<[email hidden]>...

Quoted message said:

There seem to be several sacred cows on this newsgroup, and today I would like to question the
frame stiffness one. If I understand it rightly, it says that frames can flex a bit and this is
okay because the energy that goes into sideways flexing is "returned" on the upstroke.

People may be thinking more along the lines of, if there is loss - it is small. I've not seen a
good explanation of how the energy is returned or how it is wasted. "The frame doesn't heat up"
doesn't cover it.

IIRC people have calculated significant flex loss for sprinters on the assumption that the energy
used to flex the frame is then dissipated.

Quoted message said:

Flexing is only bad if it make the chainstays rub on the wheel or the chain fall off.

But how is the energy returned? Is it returned in a way that makes my bike go forward faster? I
imagined a thought experiment. Imagine I sit still on my beautifully lugged bicycle and then push
backwards and forwards on the pedals. The hand-painted bottom bracket moves from side to side and
the frame doesn't heat up or distort so no energy is being wasted. But my legs are getting tired
and I'm not going forwards.

It seems to me that the energy used to push the bottom bracket sideways is returned on the
upstroke by pushing my leg back the other way, and not in any way by making the bike go forwards
and that any design that minimises this sideways movement will make me go faster.

This is what a lot of people believe. British Cycling have moved into frame manufacture at some
expense and the result is that David Millar pummels the opposition in the world tt. What more proof
do you need?

If there is loss, the energy must be going somewhere. Perhaps rider damping is a factor (the frame
wouldn't heat up).

For conservative pedalling, the "unwind" of the spring would have to go to the drivetrain by pushing
a pedal towards an appropriately positioned foot or help in moving leg mass around or go into
flexing the frame the other way on the next stroke or something...

Andrew Bradley

William said:

There seem to be several sacred cows on this newsgroup, and today I would like to question the
frame stiffness one. If I understand it rightly, it says that frames can flex a bit and this is
okay because the energy that goes into sideways flexing is "returned" on the upstroke. Flexing is
only bad if it make the chainstays rub on the wheel or the chain fall off.

But how is the energy returned? Is it returned in a way that makes my bike go forward faster? I
imagined a thought experiment. Imagine I sit still on my beautifully lugged bicycle and then push
backwards and forwards on the pedals. The hand-painted bottom bracket moves from side to side and
the frame doesn't heat up or distort so no energy is being wasted. But my legs are getting tired
and I'm not going forwards.

It seems to me that the energy used to push the bottom bracket sideways is returned on the
upstroke by pushing my leg back the other way, and not in any way by making the bike go forwards
and that any design that minimises this sideways movement will make me go faster.

It helps push the crank back up the other side. There is a small energy loss depending on the
material, because frame materials aren't perfect springs. Also, to some extent the tyres will scrub,
and you'll never be able to get that energy back.

However, if your gears don't shift themselves and your brakes don't rub, it's nothing to worry
about. Look at how many TdF stages Sean Kelly won on his Vitus alu bike - that was about 1/3 as
stiff as steel because they were using "normal" tube diameters. None of this Cannondale
my-tubes-are-so-fat-I-don't-need-a-front-fender stuff. I don't think the Vitus frames would last
long in hard use though.

Andrew Bradley said:

This is what a lot of people believe. British Cycling have moved into frame manufacture at some
expense and the result is that David Millar pummels the opposition in the world tt. What more
proof do you need?

Has anyone told you that you would make a great research scientist? Didn't think so.

Tim McTeague

"Tim McTeague" <[email hidden]> wrote in message "]news:[email hidden]...

Quoted message said:

Andrew Bradley said:

This is what a lot of people believe. British Cycling have moved

into

Quoted message said:

Quoted message said:

frame manufacture at some expense and the result is that David

Millar

Quoted message said:

Quoted message said:

pummels the opposition in the world tt. What more proof do you need?

Has anyone told you that you would make a great research scientist?

Didn't

Quoted message said:

think so.

Andrew knows better. I took his statement to be tongue in cheek. We all know the real reason is
because he has a superior power to drag ratio. Now if only he could work on his power to weight
ratio........and his endurance......and his bike handling skills .........and his equipment
......his teamwork .......his people skills ..... and .....

Phil Holman

Zog The Undeniable said:

William wrote:

<snip> However, if your gears don't shift themselves and your brakes don't rub, it's nothing to
worry about. Look at how many TdF stages Sean Kelly won on his Vitus alu bike - that was about 1/3
as stiff as steel because they were using "normal" tube diameters. None of this Cannondale
my-tubes-are-so-fat-I-don't-need-a-front-fender stuff. I don't think the Vitus frames would last
long in hard use though.

Those Vitus frames were lovely, Light as well for the time. I used to work for a LBS that sold them
but was warned off them because I am not light and because after a few years of club use they would
go excessively 'whippy' and the owner would have to buy another frame. I would have thaough that
Sean would have been lucky to get a years use out of a frame.

Stan Cox

--
Brighton Excelsior CC Member Cat 3 Fat Bloke Brighton, UK

Zog The Undeniable <[email hidden]> wrote in message news:<[email hidden]>...

[snip lateral frame flex question]

Quoted message said:

It helps push the crank back up the other side.

[snip]

Dear Zog,

I'm not denying your well-worded explanation, just struggling to follow what seems obvious to
everyone else. (I've been enjoying your posts all morning.)

I don't really follow how a lateral flexing of the bicycle helps the upstroke or even results from
the downstroke. I keep thinking that a sideways flexing must result from the wasted side-to-side
motion of the rider's--well, let's say his hips. I want to believe, but I need help here.

But even if I get past how an up-and-down pedal motion produces a sideways wiggle in a stiff metal
frame (maybe the offset of the pedals from the center of the frame?), I'm still vague about when the
downstroke flexing of the frame is returned during the upstroke.

If I shove down on my left pedal and my frame flexes laterally, storing energy--

(Come to think of it, which way does my rear axle flex with regard to my crank--left or right?)

--my right pedal is coming up at the same time and supposed to be restoring energy, isn't it?

How does my laterally flexing frame simultaneously store the energy from my left pedal going down
while restoring it to my right pedal coming up?

That is, what's the detailed sequence of store and restore for both pedals during a full cycle and
where is my stupid frame at each point? I don't even know which way it's flexing.

I hate being outwitted by these metal tubes.

Your befuddled but potential worshipper,

Carl Fogel

Andrew Bradley said:

If there is loss, the energy must be going somewhere. Perhaps rider damping is a factor (the frame
wouldn't heat up).

The loss is the frame going sideways. Energy is being spent flexing the frame instead of driving
the wheel.

~PB

"Carl Fogel" <[email hidden]> wrote in message
"]news:[email hidden]...

Quoted message said:

Zog The Undeniable <[email hidden]> wrote in message

news:<[email hidden]>...

Quoted message said:

[snip lateral frame flex question]

Quoted message said:

It helps push the crank back up the other side.

[snip]

Dear Zog,

I'm not denying your well-worded explanation, just struggling to follow what seems obvious to
everyone else. (I've been enjoying your posts all morning.)

I don't really follow how a lateral flexing of the bicycle helps the upstroke or even results from
the downstroke. I keep thinking that a sideways flexing must result from the wasted side-to-side
motion of the rider's--well, let's say his hips. I want to believe, but I need help here.

But even if I get past how an up-and-down pedal motion produces a sideways wiggle in a stiff metal
frame (maybe the offset of the pedals from the center of the frame?), I'm still vague about when
the downstroke flexing of the frame is returned during the upstroke.

If I shove down on my left pedal and my frame flexes laterally, storing energy--

(Come to think of it, which way does my rear axle flex with regard to my crank--left or right?)

--my right pedal is coming up at the same time and supposed to be restoring energy, isn't it?

How does my laterally flexing frame simultaneously store the energy from my left pedal going down
while restoring it to my right pedal coming up?

That is, what's the detailed sequence of store and restore for both pedals during a full cycle and
where is my stupid frame at each point? I don't even know which way it's flexing.

I hate being outwitted by these metal tubes.

Your befuddled but potential worshipper,

BB lateral flex is caused by the moment offset of the pedal from the neutral axis when applying a
force normal to the direction of flexing. It is reversible. It is at a maximum when the down stroke
pedal is horizontal and is a minimum when that pedal is at bottom dead center
i.e. the return is in the bottom half of the power stroke. The frame material's restitution
properties will govern the loss which will be in the region of 10%.

If we use a power output of 300 watts at 90rpm with a BB flex of 5mm, a crude estimate of the losses
will be *energy input into the frame with each power stroke times the number of power strokes per
second times the percentage loss*.

Using Hooke's Law (1/2kx^2), this will be .5k*x^2*3*.10 = .5*360*200*.005^2*3*.10 = .27 watt. IMO
this loss is tiny compared to the psychological loss of riding a noodly frame that handles poorly.

Phil Holman

In article <[email hidden]>, Stan Cox

Quoted message said:

Zog The Undeniable said:

William wrote:

[Vitus]

Quoted message said:

Quoted message said:

Those Vitus frames were lovely, Light as well for the time. I used to work for a LBS that sold them
but was warned off them because I am not light and because after a few years of club use they would
go excessively 'whippy' and the owner would have to buy another frame.

That's a popular myth, albeit usually applied to steel.

Some of them did suffer from the joints seperating. Mostly the carbon-fiber Vituses (Vitii?), not
the aluminum ones.

I rode mine for ~7 years, training and racing 5-8k miles/year and never noticed it getting any more
'whippy' than it was when I started.

Eric

"Stan Cox" <[email hidden]> wrote in message
"]news:[email hidden]...

Quoted message said:

{snip stuff about small dia. Al frames]

Quoted message said:

Those Vitus frames were lovely, Light as well for the time. I used to work for a LBS that sold
them but was warned off them because I am not light and because after a few years of club use they
would go excessively 'whippy'

Shouldn't this be stored in the same round metal filing cabinet on the floor as the old canard
regarding steel frames going "dead"?

Unless you're *really* suggesting that the elastic modulus of the frame material changed?

Quoted message said:

and the owner would have to buy another frame. I

Of course they would. Just like the owner of a steel frame that had gone "dead"...

Quoted message said:

would have thaough that Sean would have been lucky to get a years use out of a frame.

Stan Cox

Tim.

---
Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (grisoft.comgrisoft.comOpen ↗).
Version: 6.0.528 / Virus Database: 324 - Release Date: 16/10/03

"William" <[email hidden]> wrote in message
"]news:[email hidden]...

Quoted message said:

There seem to be several sacred cows on this newsgroup, and today I would like to question the
frame stiffness one. If I understand it rightly, it says that frames can flex a bit and this is
okay because the energy that goes into sideways flexing is "returned" on the upstroke. Flexing is
only bad if it make the chainstays rub on the wheel or the chain fall off.

I found this article by Keith Bontrager, entitled "Stiffness," to be somewhat informative.
bontrager.comrants.aspOpen ↗

"Phil Holman" <[email hidden]> wrote in message
news:<[email hidden]>...

Quoted message said:

"Carl Fogel" <[email hidden]> wrote in message
"]news:[email hidden]...

Quoted message said:

Quoted message said:

Zog The Undeniable <[email hidden]> wrote in message

news:<[email hidden]>...

Quoted message said:

[snip lateral frame flex question]

Quoted message said:

It helps push the crank back up the other side.

[snip]

[snip all of Fogel's confused question]

Quoted message said:

BB lateral flex is caused by the moment offset of the pedal from the neutral axis when applying a
force normal to the direction of flexing. It is reversible. It is at a maximum when the down
stroke pedal is horizontal and is a minimum when that pedal is at bottom dead center
i.e. the return is in the bottom half of the power stroke.

[snip]

Quoted message said:

Phil Holman

Dear Phil,

If I follow your description, my left pedal stroke flexes the frame sideways for the first
quarter-turn, reaching maximum with the crank arms horizontal, and then somehow returns the stored
energy for the second-quarter of the left-hand pedal's downstroke?

This is exactly what I was hoping for--someone to connect the dots for me and work out what kind of
energy is actually involved.

I'm afraid that this is a dumb question, but is the result during the downward quarter of the
left-pedal stroke:

a) an increased downward force on the left pedal (seems unlikely)

b) a reduced resistance on the pedal
(what you may be implying)

c) an upward force through the other pedal (maybe a bad interpretation of Zog's brief line: "It
helps push the crank back up the other side."😉

d) all or some of the above (it's a solid crank from pedal to pedal)

I take all your calculations on faith and am pleased that this frame-flex works out to very little
wasted energy. I've asked this before, but either no one was interested, or else no one interested
understood it as well as you and Zog do.

Thanks,

Carl Fogel

Phil Holman said:

Andrew knows better. I took his statement to be tongue in cheek. We all know the real reason is
because he has a superior power to drag ratio. Now if only he could work on his power to weight
ratio........and his endurance......and his bike handling skills .........and his equipment
......his teamwork .......his people skills ..... and .....

... add a front derailleur to his time trial bike.

Mark Hickey Habanero Cycles habcycles.comhabcycles.comOpen ↗ Home of the $695 ti frame

"Carl Fogel" <[email hidden]> wrote in message
"]news:[email hidden]...

Quoted message said:

"Phil Holman" <[email hidden]> wrote in message

news:<[email hidden]>...

Quoted message said:

Quoted message said:

"Carl Fogel" <[email hidden]> wrote in message
"]news:[email hidden]...

Quoted message said:

Quoted message said:

Zog The Undeniable <[email hidden]> wrote in message

news:<[email hidden]>...

Quoted message said:

[snip lateral frame flex question]

> It helps push the crank back up the other side.

[snip]

[snip all of Fogel's confused question]

Quoted message said:

BB lateral flex is caused by the moment offset of the pedal from the neutral axis when applying
a force normal to the direction of

flexing.

Quoted message said:

Quoted message said:

It is reversible. It is at a maximum when the down stroke pedal is horizontal and is a minimum
when that pedal is at bottom dead center
i.e. the return is in the bottom half of the power stroke.

[snip]

Quoted message said:

Phil Holman

Dear Phil,

If I follow your description, my left pedal stroke flexes the frame sideways for the first
quarter-turn, reaching maximum with the crank arms horizontal, and then somehow returns the stored
energy for the second-quarter of the left-hand pedal's downstroke?

This is exactly what I was hoping for--someone to connect the dots for me and work out what kind
of energy is actually involved.

I'm afraid that this is a dumb question, but is the result during the downward quarter of the
left-pedal stroke:

a) an increased downward force on the left pedal (seems unlikely)

b) a reduced resistance on the pedal
(what you may be implying)

No, but what happens is, there is more pedal travel on the top quarter part of the stroke relative
to rear wheel travel (~1%) according to my numbers. Energy is force times distance so you're storing
some energy in flexing the frame and not in turning the back wheel up to the half way point of the
downstroke.

There is one percent less travel on the lower quarter of the pedal stroke as the deflection
unwinds and you've exerted a little less energy for the same amount of work on the back wheel as
the top quarter.

Quoted message said:

c) an upward force through the other pedal (maybe a bad interpretation of Zog's brief line: "It
helps push the crank back up the other side."😉

It all ends up pushing the cranks around to provide power to the back wheel. Did you ever have a
clockwork trainset as a kid. If you didn't hold the wheels stationary they would turn as you wound
it up. When you stopped winding, the wheels would still turn until the spring ran down. This is a
very similar situation but with much smaller deflections and energy storage. For a bicycle there is
also deflection in every part of the drive train and not just the BB. Crank arms and pedal axles
twist and bend, chains stretch, back wheels wind up torsionally. You can apply the brakes and stand
on a forward pedal to see how much you can make the pedal move without the back wheel turning. It's
about 1/4 inch on my regular bike and 1/2 inch on the front crank of my tandem. Even if all this
energy was lost it would still be small (perhaps 2%). Most of it is returned and the actual loss is
more like .2%.

Quoted message said:

d) all or some of the above (it's a solid crank from pedal to pedal)

I think I've covered it.

Quoted message said:

I take all your calculations on faith and am pleased that this frame-flex works out to very little
wasted energy.

My calculations are just good enough to show that the losses are small. In an all out sprint
deflections are larger and losses greater but the stiffness consideration is probably prevention of
wheels rubbing the brakes, chains rubbing the derailleur cage and general handling through tight
bends at speed.

Phil Holman

Phil Holman writes:

about frame flex:

Quoted message said:

It all ends up pushing the cranks around to provide power to the back wheel. Did you ever have a
clockwork trainset as a kid. If you didn't hold the wheels stationary they would turn as you wound
it up. When you stopped winding, the wheels would still turn until the spring ran down. This is a
very similar situation but with much smaller deflections and energy storage.

I don't think that is a good parallel to cyclic frame deflection. Nothing happens with the frame
after the pedal stroke where the bicycle is back to its neutral position.

It's often better to consider the entire stroke of the pedals for a typical revolution. The bicycle
moves at a reasonably constant speed through several crank revolutions so any variability in any
part of the stroke is not rejected as a rebound or unused work. The pedals turn at a rate that is as
constant as the wheel rotation. Side sway of the BB is also not a rebound and absorbs no energy
since it is working against a metal spring (the frame) where no energy is dissipated.

Quoted message said:

For a bicycle there is also deflection in every part of the drive train and not just the BB. Crank
arms and pedal axles twist and bend, chains stretch, back wheels wind up torsionally. You can
apply the brakes and stand on a forward pedal to see how much you can make the pedal move without
the back wheel turning. It's about
1/4 inch on my regular bike and 1/2 inch on the front crank of my tandem. Even if all this energy
was lost it would still be small (perhaps 2%). Most of it is returned and the actual loss is
more like .2%.

I'm not sure where such numbers arise, but I suspect they depend on the rider and many other factors
but I doubt that 0.2% is a measurable value.

Quoted message said:

Quoted message said:

I take all your calculations on faith and am pleased that this frame-flex works out to very
little wasted energy.

Quoted message said:

My calculations are just good enough to show that the losses are small. In an all out sprint
deflections are larger and losses greater but the stiffness consideration is probably prevention
of wheels rubbing the brakes, chains rubbing the derailleur cage and general handling through
tight bends at speed.

I don't see how you can calculate this, considering that the losses are with the rider, if any, not
in the machinery of the bicycle.

Jobst Brandt [email hidden]

[email hidden] wrote in message news:<[email hidden]>...

Quoted message said:

Phil Holman writes:

about frame flex:

Quoted message said:

It all ends up pushing the cranks around to provide power to the back wheel. Did you ever have a
clockwork trainset as a kid. If you didn't hold the wheels stationary they would turn as you
wound it up. When you stopped winding, the wheels would still turn until the spring ran down.
This is a very similar situation but with much smaller deflections and energy storage.

I don't think that is a good parallel to cyclic frame deflection. Nothing happens with the frame
after the pedal stroke where the bicycle is back to its neutral position.

It's a stretch but I use it to illustrate that torsional and bending deflections incurred when
powering a bicycle store energy and just because the drivetrain isn't infinitely stiff doesn't mean
losses are incurred. The clockwork motor being the antithesis of infinitely stiff.

Quoted message said:

It's often better to consider the entire stroke of the pedals for a typical revolution. The
bicycle moves at a reasonably constant speed through several crank revolutions so any variability
in any part of the stroke is not rejected as a rebound or unused work. The pedals turn at a rate
that is as constant as the wheel rotation. Side sway of the BB is also not a rebound and absorbs
no energy since it is working against a metal spring (the frame) where no energy is dissipated.

Except for a small hysteresis loss.

Quoted message said:

Quoted message said:

For a bicycle there is also deflection in every part of the drive train and not just the BB.
Crank arms and pedal axles twist and bend, chains stretch, back wheels wind up torsionally. You
can apply the brakes and stand on a forward pedal to see how much you can make the pedal move
without the back wheel turning. It's about
1/4 inch on my regular bike and 1/2 inch on the front crank of my tandem. Even if all this
energy was lost it would still be small (perhaps 2%). Most of it is returned and the actual
loss is more like .2%.

I'm not sure where such numbers arise, but I suspect they depend on the rider and many other
factors but I doubt that 0.2% is a measurable value.

Exactly, it's so small as to be negligible. The question being answered here is how much more
usuable power will be available with a stiffer frame. My answer is a negligible amount.

Quoted message said:

Quoted message said:

Quoted message said:

I take all your calculations on faith and am pleased that this frame-flex works out to very
little wasted energy.

Quoted message said:

My calculations are just good enough to show that the losses are small. In an all out sprint
deflections are larger and losses greater but the stiffness consideration is probably prevention
of wheels rubbing the brakes, chains rubbing the derailleur cage and general handling through
tight bends at speed.

I don't see how you can calculate this, considering that the losses are with the rider, if any,
not in the machinery of the bicycle.

By what mechanism are you saying the losses are with the rider? I believe my calculations, crude as
they are, agree that the losses are not significant in the machinery. My estimate is probably fairly
conservative but good enough to determine a rough order of magnitude. I don't think we need to run a
finite element model to conclude that frame flex within the limit of inducing chain or wheel rub or
poor handling will not result in any significant losses of power to the rear wheel.

Phil Holman

Phil Holman said:

Quoted message said:

I don't see how you can calculate this, considering that the losses are with the rider, if any,
not in the machinery of the bicycle.

By what mechanism are you saying the losses are with the rider?

Phil was responding to Jobst, but to offer an answer--

The losses can be in your legs. The frame may be close to 100% elastic (no significant hysteretic
loss) but the energy doesn't necessarily go into driving the bicycle.

Imagine you're tightening a bolt with a ratchet and socket, and you have a long extension that's not
stiff torsionally and winds up with each stroke. You won't turn the bolt as many degrees with each
stroke as you would with a stiff extension. Where does the energy go that was stored in winding up
the extension?

Tom Ace

Tom Ace said:

The losses can be in your legs. The frame may be close to 100% elastic (no significant hysteretic
loss) but the energy doesn't necessarily go into driving the bicycle.

Quoted message said:

Imagine you're tightening a bolt with a ratchet and socket, and you have a long extension that's
not stiff torsionally and winds up with each stroke. You won't turn the bolt as many degrees with
each stroke as you would with a stiff extension. Where does the energy go that was stored in
winding up the extension?

Oops! There is a rebound with that example and one that doesn't advance the bolt. In a bicycle,
there is no rebound, the foot never retreating from its most advanced position of the pedal
revolution. This is the point I made earlier.

Jobst Brandt [email hidden]

Quoted message said:

Quoted message said:

Imagine you're tightening a bolt with a ratchet and socket, and you have a long extension that's
not stiff torsionally and winds up with each stroke. You won't turn the bolt as many degrees
with each stroke as you would with a stiff extension. Where does the energy go that was stored
in winding up the extension?

Oops! There is a rebound with that example and one that doesn't advance the bolt. In a bicycle,
there is no rebound, the foot never retreating from its most advanced position of the pedal
revolution. This is the point I made earlier.

Bear with me please, I missed any earlier post where this was mentioned.

I'm still not convinced. There isn't the same kind of rebound as in the ratchet example, but there
is a stroke, and downward force on the pedal has different effects at different parts of the stroke.
You can't apply much drive torque near the bottom of the stroke. If the frame springs back at that
point, is the energy returned really going primarily into driving the bicycle?

Tom Ace