Frame flex energy waste?

[[email protected]]

Hi All,

I finished putting together my TT bike and have been testing it out
the last few days. One problem is that the cheap aluminum frame flexes
a lot when I am mashing big gears with a low cadence. The crankarms
clip the chainstays and I can see the seat tube flex like an S as I
push through the powerstroke. The crankarms with no load have about
4-5mm clearance to the chainstays.

This is obviously due to the cheap frame, which I may replace at some
point, but probably not. The point of this bike was to be cheap and
this was a frame I had lying around, so buying a new non-cheap frame
doesn't really fit the brief...

So how much energy am I actually wasting with this flex? Is there some
way to quantify it mathematically with a couple of well placed WAG's?

Joseph

 

[_]

On 25 May 2007 05:31:34 -0700, [email protected] wrote:

> Hi All,
>
> I finished putting together my TT bike and have been testing it out
> the last few days. One problem is that the cheap aluminum frame flexes
> a lot when I am mashing big gears with a low cadence. The crankarms
> clip the chainstays and I can see the seat tube flex like an S as I
> push through the powerstroke. The crankarms with no load have about
> 4-5mm clearance to the chainstays.
>
> This is obviously due to the cheap frame, which I may replace at some
> point, but probably not. The point of this bike was to be cheap and
> this was a frame I had lying around, so buying a new non-cheap frame
> doesn't really fit the brief...
>
> So how much energy am I actually wasting with this flex? Is there some
> way to quantify it mathematically with a couple of well placed WAG's?
>

Yes.

Well, a qualified yes. The loss of energy in the frame is, absent
permanent deformation, hysteresis loss. For metals operating in the elstic
region, this is quite small, with the phisiological onesbut can be
measured by an increase in temperature. The instrumentation and labratory
conditions required to do so are likely to be several orders of magnitude
more costly than your frame.

Ther also exists the possible loss from drive-train changes in efficiency -
starting with physiological ones, then bearing losses dues to
mis-alignment, chain and transmission losses due to same, tyre scrub,
etcetera; but all of these

a) are also likely to be small;

b) would be *extremely* difficult to measure with any degreee of certainty;
and

c) may also show as a gain of efficiency, depending on circumstances.

 

[[email protected]]

On May 25, 4:39 pm, _ <[email protected]>
wrote:
> On 25 May 2007 05:31:34 -0700, [email protected] wrote:
>
>
>
> > Hi All,
>
> > I finished putting together my TT bike and have been testing it out
> > the last few days. One problem is that the cheap aluminum frame flexes
> > a lot when I am mashing big gears with a low cadence. The crankarms
> > clip the chainstays and I can see the seat tube flex like an S as I
> > push through the powerstroke. The crankarms with no load have about
> > 4-5mm clearance to the chainstays.
>
> > This is obviously due to the cheap frame, which I may replace at some
> > point, but probably not. The point of this bike was to be cheap and
> > this was a frame I had lying around, so buying a new non-cheap frame
> > doesn't really fit the brief...
>
> > So how much energy am I actually wasting with this flex? Is there some
> > way to quantify it mathematically with a couple of well placed WAG's?
>
> Yes.
>
> Well, a qualified yes. The loss of energy in the frame is, absent
> permanent deformation, hysteresis loss. For metals operating in the elstic
> region, this is quite small, with the phisiological onesbut can be
> measured by an increase in temperature. The instrumentation and labratory
> conditions required to do so are likely to be several orders of magnitude
> more costly than your frame.
>
> Ther also exists the possible loss from drive-train changes in efficiency -
> starting with physiological ones, then bearing losses dues to
> mis-alignment, chain and transmission losses due to same, tyre scrub,
> etcetera; but all of these
>
> a) are also likely to be small;
>
> b) would be *extremely* difficult to measure with any degreee of certainty;
> and
>
> c) may also show as a gain of efficiency, depending on circumstances.

So you think the actual waste is very small, orders of magnitude less
than things like low tire pressure, dry chain, etc?

Joseph

 

[Leo Lichtman]

<[email protected]> wrote: So you think the actual waste is
very small, orders of magnitude less than things like low tire pressure, dry
chain, etc?
^^^^^^^^^^^^^^^^^^^
It is true that energy wasted in the frame (flexing) appears as
heat--somewhere--and that measuring or calculating it would be extremely
difficult. A non-trivial waste of energy occurs when you are straining to
turn the pedals, as you describe, and this has very little to do with the
bike. Your body consumes energy to produce muscle force, even when you are
barely moving, or not moving at all. Imagine holding yourself up on a
chinning bar for a fraction of a minute. Even though you are producing zero
output, your heart and breathing rates will go up, and your muscles will
ache, indicating that your body is burning energy.

My main concern about an aluminum frame flexing that much would be the
possibility of fatigue cracks developing.

 

[jim beam]

[email protected] wrote:
> Hi All,
>
> I finished putting together my TT bike and have been testing it out
> the last few days. One problem is that the cheap aluminum frame flexes
> a lot when I am mashing big gears with a low cadence. The crankarms
> clip the chainstays and I can see the seat tube flex like an S as I
> push through the powerstroke. The crankarms with no load have about
> 4-5mm clearance to the chainstays.
>
> This is obviously due to the cheap frame,

not necessarily "cheap". indeed, some "cheap" frames [nashbar] are very
rigid. it's just that your frame, whatever the cost, is flexible.

> which I may replace at some
> point, but probably not. The point of this bike was to be cheap and
> this was a frame I had lying around, so buying a new non-cheap frame
> doesn't really fit the brief...
>
> So how much energy am I actually wasting with this flex? Is there some
> way to quantify it mathematically with a couple of well placed WAG's?

you're not losing energy in the frame - hysteresis is effectively zero.
but you can get losses doe to flex and subsequent [mis]alignment -
elastic temporary - and therefore tire scrub.

 

[[email protected]]

Joseph Santaniello writes:

> I finished putting together my TT bike and have been testing it out
> the last few days. One problem is that the cheap aluminum frame
> flexes a lot when I am mashing big gears with a low cadence. The
> cranks clip the chainstays and I can see the seat tube flex like an
> S as I push through the power stroke. The cranks with no load have
> about 4-5mm clearance to the chainstays.

> This is obviously due to the cheap frame, which I may replace at
> some point, but probably not. The point of this bike was to be
> cheap and this was a frame I had lying around, so buying a new
> non-cheap frame doesn't really fit the brief.

> So how much energy am I actually wasting with this flex? Is there
> some way to quantify it mathematically with a couple of well placed
> WAG's?

You aren't losing any energy but frame elasticity interferes with your
pedaling and possibly in riding a straight line. Have you coasted
no-hands down a smooth road to see at what speed the bicycle naturally
shimmies. Try this with pedals horizontal and vertical, the two
positions having different inertial effects.

The only energy loss would appear as heat in frame tubes, and there
isn't any being generated in the elastic excursions of the frame.
These are mostly torsional and are minuscule in physical deformation.

Jobst Brandt

 

[Leo Lichtman]

<[email protected]> wrote in: You aren't losing any energy but
frame elasticity interferes with your pedaling and possibly in riding a
straight line. (clip)
^^^^^^^^^^^^^^^^
I think Tom Ace makes a very good point. Let's do this thought experiment:
Put a pair of coil springs between your feet and the pedals. On every
downstroke, you have to compress the springs, and on the return stroke, you
absorb the stored energy back into your legs. You have a very spongy power
train, but the energy is not going into the springs as heat. It is going
into your legs (as heat.). On a much smaller scale, but the same principle
applies, frame flexure energy would be returned to the rider's legs.

But, there is this other possibility. It is conceivable that the frame
flexure is relaxed at the end of each power stroke in a direction which
contributes to bicycle travel. That energy would not be converted to heat,
and would not be wasted.

How much energy goes backward and how much goes forward probably depends on
the particular frame/rider combination.

 

[Ron Ruff]

On May 25, 6:31 am, "[email protected]"
<[email protected]> wrote:

> One problem is that the cheap aluminum frame flexes
> a lot when I am mashing big gears with a low cadence. The crankarms
> clip the chainstays and I can see the seat tube flex like an S as I
> push through the powerstroke. The crankarms with no load have about
> 4-5mm clearance to the chainstays.

Cheap aluminum frames are usually heavy and *stiff*. What kind of
frame is this?

 

[Ron Ruff]

On May 25, 9:42 am, Tom Ace <[email protected]> wrote:
> In the case of a flexible bike frame, the frame flexes during
> a valuable portion of the pedal stroke and springs back near
> the end of the stroke--where the crank is not in a position
> to effectively convert downward force on the pedal into drive
> torque.

The studies I've seen of force applied show a bell curve with the
maximum at around 90 degrees, tapering to zero at 0 and 180 degrees.
So the flex that occurs as the force increases (0 to 90) would be
returned as the force is decreasing (90 to 180). Also consider that
the frame will flex back and forth in equal amounts, and it makes
sense that it would pass through "neutral" at the 0 and 180 degree
positions. IMO a flexy frame acts like an oval chainring, and very
little of the energy is lost.

 

[[email protected]]

On May 25, 4:18 pm, Ron Ruff <[email protected]> wrote:
> On May 25, 9:42 am, Tom Ace <[email protected]> wrote:
>
> > In the case of a flexible bike frame, the frame flexes during
> > a valuable portion of the pedal stroke and springs back near
> > the end of the stroke--where the crank is not in a position
> > to effectively convert downward force on the pedal into drive
> > torque.
>
> The studies I've seen of force applied show a bell curve with the
> maximum at around 90 degrees, tapering to zero at 0 and 180 degrees.
> So the flex that occurs as the force increases (0 to 90) would be
> returned as the force is decreasing (90 to 180). Also consider that
> the frame will flex back and forth in equal amounts, and it makes
> sense that it would pass through "neutral" at the 0 and 180 degree
> positions. IMO a flexy frame acts like an oval chainring, and very
> little of the energy is lost.

http://www.microsporttech.com/faq.php shows a clock diagram. The force
at 180 isn't really zero. So maybe there is some loss?

 

[Marcus Coles]

[email protected] wrote:
>
> http://www.microsporttech.com/faq.php shows a clock diagram. The force
> at 180 isn't really zero. So maybe there is some loss?

I'm not buying their story based on their supporting documentation. So
why should I buy their diagram?

From the referenced document: The influence of cadence and power output
on the biomechanics of force application during steady-rate cycling in
competitive and recreational cyclists. Sanderson DJ

"It was concluded that the lack of difference between the groups was a
combined consequence of the limited degrees of freedom associated with
the bicycle and that the relatively low power output for the competitive
riders was insufficient to discriminate or highlight superior riding
technique."


Marcus Coles

 

[wizardB]

Ron Ruff wrote:
> On May 25, 6:31 am, "[email protected]"
> <[email protected]> wrote:
>
>> One problem is that the cheap aluminum frame flexes
>> a lot when I am mashing big gears with a low cadence. The crankarms
>> clip the chainstays and I can see the seat tube flex like an S as I
>> push through the powerstroke. The crankarms with no load have about
>> 4-5mm clearance to the chainstays.
>
> Cheap aluminum frames are usually heavy and *stiff*. What kind of
> frame is this?
>
You can have
cheap
light
or Strong
Pick any two
If you opt for cheap and light it will be weak this is a rule of bike buying

 

[lightninglad]

.....snip.......
> So how much energy am I actually wasting with this flex? Is there some
> way to quantify it mathematically with a couple of well placed WAG's?
>
> Joseph

I owned a very light aluminium lowracer recument trike. It was slower
up hills than my previous heavier steel trike and there was about 25mm
of flex in the BB under static load (brakes on, pedal at 12, push
hard.)
The manufacturer fitted a brace and reduced the flex to approx. 3mm.

My average speed increased approx. 10% with the more rigid frame and
virtually all of that was due to the increase in hill climbing speed.

With a perfect circular power stroke pedal action and a high cadence,
the power loss from frame flex would be small, because the flex under
those conditions is small. Your frame is not suitable for mashers!

 

[[email protected]]

On May 25, 11:09 pm, Ron Ruff <[email protected]> wrote:
> On May 25, 6:31 am, "[email protected]"
>
> <[email protected]> wrote:
> > One problem is that the cheap aluminum frame flexes
> > a lot when I am mashing big gears with a low cadence. The crankarms
> > clip the chainstays and I can see the seat tube flex like an S as I
> > push through the powerstroke. The crankarms with no load have about
> > 4-5mm clearance to the chainstays.
>
> Cheap aluminum frames are usually heavy and *stiff*. What kind of
> frame is this?

I don't want to name names, because my use isn't really typical for
this frame so it wouldn't be fair to rag on it. But it's a $69 frame
from one of the bigger eBay frame sellers. 6061 Al 32mm seat-tube, and
a wide teardrop shaped down-tube.

Joseph

 

[Lou Holtman]

[email protected] wrote:
> On May 25, 11:09 pm, Ron Ruff <[email protected]> wrote:
>> On May 25, 6:31 am, "[email protected]"
>>
>> <[email protected]> wrote:
>>> One problem is that the cheap aluminum frame flexes
>>> a lot when I am mashing big gears with a low cadence. The crankarms
>>> clip the chainstays and I can see the seat tube flex like an S as I
>>> push through the powerstroke. The crankarms with no load have about
>>> 4-5mm clearance to the chainstays.
>> Cheap aluminum frames are usually heavy and *stiff*. What kind of
>> frame is this?
>
> I don't want to name names, because my use isn't really typical for
> this frame so it wouldn't be fair to rag on it. But it's a $69 frame
> from one of the bigger eBay frame sellers. 6061 Al 32mm seat-tube, and
> a wide teardrop shaped down-tube.


If your crankarms hit the chainstays that's riduculous. Get rid of the
frame. Not because of the power loss, but because it's unnecessary and
annoying. Take your 69 dollar loss.

Lou
--
Posted by news://news.nb.nu (http://www.nb.nu)

 

[Tim McNamara]

In article <[email protected]>,
Lou Holtman <[email protected]> wrote:

> [email protected] wrote:
> > On May 25, 11:09 pm, Ron Ruff <[email protected]> wrote:
> >> On May 25, 6:31 am, "[email protected]"
> >>
> >> <[email protected]> wrote:
> >>> One problem is that the cheap aluminum frame flexes a lot when I
> >>> am mashing big gears with a low cadence. The crankarms clip the
> >>> chainstays and I can see the seat tube flex like an S as I push
> >>> through the powerstroke. The crankarms with no load have about
> >>> 4-5mm clearance to the chainstays.
> >> Cheap aluminum frames are usually heavy and *stiff*. What kind of
> >> frame is this?
> >
> > I don't want to name names, because my use isn't really typical for
> > this frame so it wouldn't be fair to rag on it. But it's a $69
> > frame from one of the bigger eBay frame sellers. 6061 Al 32mm
> > seat-tube, and a wide teardrop shaped down-tube.
>
>
> If your crankarms hit the chainstays that's riduculous. Get rid of
> the frame. Not because of the power loss, but because it's
> unnecessary and annoying. Take your 69 dollar loss.

Or get a longer BB axle.

 

[[email protected]]

On May 26, 4:23 pm, Tim McNamara <[email protected]> wrote:
> In article <[email protected]>,
> Lou Holtman <[email protected]> wrote:
>
>
>
> > [email protected] wrote:
> > > On May 25, 11:09 pm, Ron Ruff <[email protected]> wrote:
> > >> On May 25, 6:31 am, "[email protected]"
>
> > >> <[email protected]> wrote:
> > >>> One problem is that the cheap aluminum frame flexes a lot when I
> > >>> am mashing big gears with a low cadence. The crankarms clip the
> > >>> chainstays and I can see the seat tube flex like an S as I push
> > >>> through the powerstroke. The crankarms with no load have about
> > >>> 4-5mm clearance to the chainstays.
> > >> Cheap aluminum frames are usually heavy and *stiff*. What kind of
> > >> frame is this?
>
> > > I don't want to name names, because my use isn't really typical for
> > > this frame so it wouldn't be fair to rag on it. But it's a $69
> > > frame from one of the bigger eBay frame sellers. 6061 Al 32mm
> > > seat-tube, and a wide teardrop shaped down-tube.
>
> > If your crankarms hit the chainstays that's riduculous. Get rid of
> > the frame. Not because of the power loss, but because it's
> > unnecessary and annoying. Take your 69 dollar loss.
>
> Or get a longer BB axle.

That's the plan. It's 113 now, a 118 or 122 might do the trick. It'll
make cross chaining more problematic, which is unfortunate because I'd
like to just leave it in the big ring all the time as I'd like to have
use of the 53x21. But that would be less of an irritant than the clunk-
clunk-clunk and 53x21 doesn't really work due to the auto-shift
feature anyway!

Joseph

 

[Kerry Montgomery]

<[email protected]> wrote in message
news:[email protected]...
> On May 26, 4:23 pm, Tim McNamara <[email protected]> wrote:
>> In article <[email protected]>,
>> Lou Holtman <[email protected]> wrote:
>>
>>
>>
>> > [email protected] wrote:
>> > > On May 25, 11:09 pm, Ron Ruff <[email protected]> wrote:
>> > >> On May 25, 6:31 am, "[email protected]"
>>
>> > >> <[email protected]> wrote:
>> > >>> One problem is that the cheap aluminum frame flexes a lot when I
>> > >>> am mashing big gears with a low cadence. The crankarms clip the
>> > >>> chainstays and I can see the seat tube flex like an S as I push
>> > >>> through the powerstroke. The crankarms with no load have about
>> > >>> 4-5mm clearance to the chainstays.
>> > >> Cheap aluminum frames are usually heavy and *stiff*. What kind of
>> > >> frame is this?
>>
>> > > I don't want to name names, because my use isn't really typical for
>> > > this frame so it wouldn't be fair to rag on it. But it's a $69
>> > > frame from one of the bigger eBay frame sellers. 6061 Al 32mm
>> > > seat-tube, and a wide teardrop shaped down-tube.
>>
>> > If your crankarms hit the chainstays that's riduculous. Get rid of
>> > the frame. Not because of the power loss, but because it's
>> > unnecessary and annoying. Take your 69 dollar loss.
>>
>> Or get a longer BB axle.
>
> That's the plan. It's 113 now, a 118 or 122 might do the trick. It'll
> make cross chaining more problematic, which is unfortunate because I'd
> like to just leave it in the big ring all the time as I'd like to have
> use of the 53x21. But that would be less of an irritant than the clunk-
> clunk-clunk and 53x21 doesn't really work due to the auto-shift
> feature anyway!
>
> Joseph
>
Joseph,
If you don't plan to use the small ring, you might try putting the big ring
on the inside of the crank spider and see if the chain line is better.
Kerry

 

[Michael Press]

In article
<[email protected]>
,
"[email protected]"
<[email protected]> wrote:

> On May 26, 4:23 pm, Tim McNamara <[email protected]> wrote:
> > In article <[email protected]>,
> > Lou Holtman <[email protected]> wrote:
> >
> >
> >
> > > [email protected] wrote:
> > > > On May 25, 11:09 pm, Ron Ruff <[email protected]> wrote:
> > > >> On May 25, 6:31 am, "[email protected]"
> >
> > > >> <[email protected]> wrote:
> > > >>> One problem is that the cheap aluminum frame flexes a lot when I
> > > >>> am mashing big gears with a low cadence. The crankarms clip the
> > > >>> chainstays and I can see the seat tube flex like an S as I push
> > > >>> through the powerstroke. The crankarms with no load have about
> > > >>> 4-5mm clearance to the chainstays.
> > > >> Cheap aluminum frames are usually heavy and *stiff*. What kind of
> > > >> frame is this?
> >
> > > > I don't want to name names, because my use isn't really typical for
> > > > this frame so it wouldn't be fair to rag on it. But it's a $69
> > > > frame from one of the bigger eBay frame sellers. 6061 Al 32mm
> > > > seat-tube, and a wide teardrop shaped down-tube.
> >
> > > If your crankarms hit the chainstays that's riduculous. Get rid of
> > > the frame. Not because of the power loss, but because it's
> > > unnecessary and annoying. Take your 69 dollar loss.
> >
> > Or get a longer BB axle.
>
> That's the plan. It's 113 now, a 118 or 122 might do the trick. It'll
> make cross chaining more problematic, which is unfortunate because I'd
> like to just leave it in the big ring all the time as I'd like to have
> use of the 53x21. But that would be less of an irritant than the clunk-
> clunk-clunk and 53x21 doesn't really work due to the auto-shift
> feature anyway!

With a longer crank spindle you will get more auto-shifts.

--
Michael Press

 

[dvt]

Tom Ace wrote:
> This has been discussed (in this ng, and elsewhere) at length.
>
> Even if the frame is elastic (i.e. exhibits no hysteretic loss),
> I am not convinced that there is no waste. The question that
> interests me is, where does the energy returned by the frame go;
> does it drive the bike, or is it dissipated in the rider's legs?

Tom, you make a good point. There is an engineering term for this:
coupling coefficient [1]. If a device stores energy rather than turning
it into work, the device has a poor coupling coefficient. So a spongy
bike (energy stored in the spring) would have a worse coupling
coefficient than a stiff bike.

When energy is stored, it is subject to a few loss mechanisms. One
mechanism is the loss in the storage itself (i.e. a lossy spring).
Another mechanism is the load to which the energy is delivered. In your
system, one load is propulsion (useful energy) and another load is the
rider's legs (wasted energy). So energy storage in a flexible frame
opens the doors for a few loss mechanisms. Quantifying those mechanisms
is difficult.

Even if there are no additional losses, poor coupling generally results
in poor system performance. For example, the latter half of [1] deals
with piezoelectric devices. All else being equal, a piezoelectric sensor
with higher coupling will have better performance. I've proved this to
myself on several occasions. I haven't thought through the implications
for a bicycle and rider system, but I think there may be a connection.

[1] A good, brief summary of the meaning of coupling coefficient can be
found at http://www.morganelectroceramics.com/pdfs/tp224.pdf.

--
Dave
dvt at psu dot edu

Everyone confesses that exertion which brings out all the powers of body
and mind is the best thing for us; but most people do all they can to
get rid of it, and as a general rule nobody does much more than
circumstances drive them to do. -Harriet Beecher Stowe, abolitionist and
novelist (1811-1896)