Torgue effects on spoke tension
Ron Ruff
calculate the effect of wheel torque on the spoke's tension... and I'm
surprised at how large it is.
First lets look at the force that a rider exerts on a pedal. This is
very straight forward if we know power output and cadence... except
for needing to guess the relationship between the peak and the average
force. I assumed a ratio of 2, based on measured data for seated
cycling. I used the same number for sprinting as well, though I
suspect it might be higher.
Rider power (W).......... 300....1500
Cadence.......................70........90
Crank length (mm).......175......175
Peak torque/ avg............2..........2
Peak torque (N-m).........82......318
Peak force (N).............468.....1819
Peak force (lb)............105.......409
The first column is a typical cyclist on a seated climb. The second
column is an elite cyclist starting a sprint. You might think that an
elite cyclist in a sprint would be the highest value possible, but I
think higher instantaneous forces can happen pretty easily. For
instance, it isn't difficult to stomp down on the pedal with double
your body weight... especially if it is only for an instant. So a
200lb rider could match the elite sprinters pedal force, and a 300lb
rider could generate a force of ~600lb with a good stomp. Plus, as far
as the spokes are concerned, being in a low gear while applying a high
force is the worst case.
Next I looked at the change in spoke tension for a given crank force
and gear ratio. This is a straight forward calculation as well, since
I assumed that the rim was infinitely stiff. Actually this assumption
will result in a slightly *lower* peak change in tension than reality,
because flexing of the rim will amplify the variation... but you will
still have the same *average* change in spoke tension regardless.
I chose a configuration that was fairly typical but on the "flexible"
side of things... 28h hub and rim, laced 2x on the drive side, and
radial on the NDS (ie the NDS does not transfer torque). The effective
flange radius where the spokes attach is 20mm.
Effective flange radius (mm)..........20.......20......20
Crank length (mm)......................175......175......175
Front sprocket teeth.....................34.......53........34
Rear sprocket teeth......................27........11.......27
Crossed spokes on drive side.......14.......14........14
Force exerted at crank (N)..........364.....1819....2668
Spoke tension change +-(N).......180.......236......1324
The first column is our seated climber putting out 300W in a 34/27
gear at 70rpm. Since a typical DS static spoke tension is over 1000N,
a 180N tension change shouldn't cause any trouble. For the elite
sprinter it is hardly any higher... only 236N in a 53/11 at 90rpm. But
for the 300lb rider applying double his body weight in a 34/27 gear,
it is 1324N! This means that the "pushing" spokes would go completely
slack, and the "pulling" spokes would probably exceed yield. A mere
400lb force would still result in a 883N... hardly trivial.
So... the moral of the story is: If you are the sort of rider who can
produce a good amount of force and occasionally like to do so in low
gear uphill situations, then using a reasonably fat rear hub with
plenty of tangential spokes on both sides would be a good idea. In any
case the effects of torque are high enough that they need to be
considered.
jim beam
b...@mambo.ucolick.org
>
> The first column is our seated climber putting out 300W in a 34/27
> gear at 70rpm. Since a typical DS static spoke tension is over 1000N,
> a 180N tension change shouldn't cause any trouble. For the elite
> sprinter it is hardly any higher... only 236N in a 53/11 at 90rpm. But
> for the 300lb rider applying double his body weight in a 34/27 gear,
> it is 1324N! This means that the "pushing" spokes would go completely
> slack, and the "pulling" spokes would probably exceed yield. A mere
> 400lb force would still result in a 883N... hardly trivial.
>
> So... the moral of the story is: If you are the sort of rider who can
> produce a good amount of force and occasionally like to do so in low
> gear uphill situations, then using a reasonably fat rear hub with
> plenty of tangential spokes on both sides would be a good idea. In any
> case the effects of torque are high enough that they need to be
> considered.
400 lb force on a pedal is 210 lb force at the tire/ground
contact (just the ratio of 175mm crank to 335mm radius tire).
Under many circumstances this will spin out the rear wheel,
so the full tension increase won't be realized. (For
a 200 lb rider with 70% weight on the back wheel, that's
140 lb of normal force and you need a coefficient of friction
of 1.5 to keep from spinning out at this force.)
If you're riding offroad and stomp the granny gear hard
rather than steadily, you'll almost always spin it.
Ben
jobst....@stanfordalumni.org
> Spoke tension change +-(N).......180.......236......1324
I disagree and suggest you look at the computed values and graphic
display of that data in "the Bicycle Wheel" in which the torque is
chosen to be the normal weight on the wheel for climbing a vertical
wall. The reason for this is that all the force vectors, load, brake,
torque and combination of these are the same so that comparing the
effects is visually simple.
You'll notice that the radial load of the rider on flat ground is the
largest stress change for a wheel and that torque is insignificantly
low in comparison. I suggest you try finite element modeling to get a
better result.
Jobst Brandt
Leo Lichtman
<b...@mambo.ucolick.org> wrote: (clip) 400 lb force on a pedal is 210 lb
^^^^^^^^^^^^^^^^^^^^^
This is calculation leaves out the ratio between chainwheel and rear
cog--true only for a child's tricycle or a big wheel bike.
Ron Ruff
> I disagree and suggest you look at the computed values and graphic
> display of that data in "the Bicycle Wheel" in which the torque is
> chosen to be the normal weight on the wheel for climbing a vertical
> wall.
If I understand it correctly, your analysis assumes an infinitely
rigid hub, 36 1.6mm dia spokes with a +-5 degree angle to the rim
(effective flange radius of 26.5mm), a wheel radius of 300mm, and a
tangential load of 50kg. Since you are applying the load to the rim,
the cranks and gear ratio don't matter. I've selected the following
values in order to get the tangential force at the tire to match your
50kg (490.5N).
Effective flange radius (mm).......26.5
Crank length (mm).....................300
Front sprocket teeth......................1
Rear sprocket teeth.......................1
Crossed spokes on drive side.........36
Force exerted at crank (N)............490.5
Spoke diameter (mm)..................1.6
Spoke length (mm)......................300
Spoke area (mm^2).....................2.01
Tangential force at tire (N)...........490.5
Spoke tension change +-(N).........154.2
Spoke length change +-(mm)........0.112
This isn't a very high change in spoke tension... but assuming that
there are 36 spokes in an infinitely stiff hub isn't reasonable. You
calculated elsewhere in your book that a typical narrow hub shaft had
only 12% of the stiffness of 18 crossed spokes, so you would
effectively have 20 crossed spokes at most. A hub radius of 26.5mm is
a bit high as well.
Your FEM of the above configuration gave a tangential deflection of
~1mm. The change in spoke tension is not given. Elsewhere, you showed
a test (2nd edition p131) where the wheel (small flange hub) had a
tangential deflection of 3.43mm with a 30kg load. This is .114mm/kg,
compared to .020mm/kg in your FEM... nearly a factor of 6 different. I
don't know for sure what is going on in your FEM, but your tested
value seems closer to reality.
I have built an FEM of a wheel with 14 crossed spokes on the drive
side. The tangential deflection is .12mm/kg (nearly the same as your
tested value). Though the spoking is probably not as stiff as the
wheel you tested (I'm guessing that yours was a 36h), my rim is likely
much stiffer (30mm deep) and this has a significant effect. As I
stated in the original post, calculating the change in spoke tension
is a trivial thing if all your crossed spokes are on one side, and you
assume that the rim is infinitely stiff. Any flexibility in the rim
will make the situation *worse* not better.
At any rate the torsional forces must balance, so the avg change in
tension times the effective flange radius times the number of spokes
must add up to the resisted torque... or in other words the tangential
load times the wheel radius.
dTs(avg) x Rf x Ns = F x Rw
Ron Ruff
b...@mambo.ucolick.org wrote:
> 400 lb force on a pedal is 210 lb force at the tire/ground
> contact (just the ratio of 175mm crank to 335mm radius tire).
Close... but you must include the gear ratio as well.
400lb x 27/34 x 175/335 = 166lb
> Under many circumstances this will spin out the rear wheel,
> so the full tension increase won't be realized. (For
> a 200 lb rider with 70% weight on the back wheel, that's
> 140 lb of normal force and you need a coefficient of friction
> of 1.5 to keep from spinning out at this force.)
If there is an instant of force of 400lb then that is what we need to
use. I'd guess a 50/50 distribution is more likely with the foot at 90
deg on the crank, giving us 200lb on the rear wheel. The coefficient
of friction would then need to be 166/200 = .83... which I think is
reasonable.
b...@mambo.ucolick.org
wrote:
You are correct and I missed that out, and that's why
it's easier to spin out in low gear. However, in the
example given of a 34/27, that ratio was fairly close to 1.
Suppose I step on the pedal with 1000 N force. I have
a 175mm crank, 34t ring, 27t cog, whose radii are 69mm
and 55mm; and a 335mm radius wheel.
The chain tension force is Ftens = 1000 * 175/69.
The torque on the rear wheel is
Trear = Ftens*Rcog = 1000 * 175/69 * 55
The force at the tire is
Trear/Rtire = 1000 * 175/69 * 55/335 = 416 N.
So I overestimated the torque; it's lower by 55/69
or 1/(34/27) = 0.79. However it's not so that one only
finds 1:1 on tricycles. MTBs often have 22/32;
even old ones got as low as 26/28. Admittedly you
rarely need that except steep pitches offroad,
and then you're spinning because dirt has less
traction than asphalt. But on asphalt I still bet
that wheelspin is far more common than loading
spokes to yield, especially since there are
of order 16 spokes to share the force.
Ben
b...@mambo.ucolick.org
Yes, I missed the gear ratio, see previous post.
You just assumed a 400lb rider though. If it's a 200 lb rider
stomping 2x body weight, then the coefficient of friction
required goes up to ~1.66.
I don't know if I've ever spun out a bit on a standing
start on clean pavement, but it's possible to do on
slightly dirty pavement, crosswalk stripes etc.
Haven't broken any spokes that way though.
Ben
jim beam
if you're spinning, you're not pedaling smoothly and your c.g. is
bouncing. when i first started mtb, i had those kinds of problems. it
comes down to technique [and good tires of course]. if you get your
technique down, mountain goats can't catch you.
Ron Ruff
wrote:
> Yes, I missed the gear ratio, see previous post.
> You just assumed a 400lb rider though. If it's a 200 lb rider
> stomping 2x body weight, then the coefficient of friction
> required goes up to ~1.66.
The point is that if there is an instant of downward force equal to
400lb, then *that* is the force that is supported by the tires... it
doesn't matter what the rider weighs. So if it is equal on both tires,
then the coefficient of friction needed is only 0.83.
I haven't managed to yield my spokes either, though I have "tried"...
there was a steep hill near my house that I was in the habit of
stomping up at full power on a regular basis with bad form
(thrashing)... but my rear wheel has a beefy hub with 32 crossed
spokes, so I wouldn't expect it to be possible. I'll soon be building
one with only 14 crossed spokes, so I may have more success with that.
Even if the spokes don't yield, it is still pretty severe service. It
is more likely that the bottom "pushing" spoke or two will go
completely slack, since they are seeing a compressive load as well...
plus there is the lateral force to consider.
carl...@comcast.net
wrote:
[snip]
> This means that the "pushing" spokes would go completely
>slack, and the "pulling" spokes would probably exceed yield.
[snip]
Dear Ron,
Forgive my skepticism, but . . .
If it were possible to yield spokes by stomping hard on the pedals,
wouldn't we see spokes routinely when we grab the brakes?
Any rider can (de)accelerate far faster than he can accelerate, for
any range of speed, time, or distance
Another way to look at things is that a single butted spoke with a 1.8
mm midsection will easily hold a 400-lb weight without yielding.
Jobst's tests showed that a 1.8 mm midsection spoke yields at a bit
over 500 pounds of tension.
It seems unlikely that a rider using just one leg can raise the
tension of all the pulling spokes (pre-tensioned to ~250 lbs) by
another 250 pounds each.
Even if Superman did stomp on the pedal, the tire would probably just
spin, with the rubber losing traction before the spokes could be
pulled hard enough to reach yield.
Cheers,
Carl Fogel
jobst....@stanfordalumni.org
>> This means that the "pushing" spokes would go completely slack, and
>> the "pulling" spokes would probably exceed yield.
> Forgive my skepticism, but...
> If it were possible to yield spokes by stomping hard on the pedals,
> wouldn't we see spokes routinely when we grab the brakes?
> Any rider can (de)accelerate far faster than he can accelerate, for
> any range of speed, time, or distance
> Another way to look at things is that a single butted spoke with a
> 1.8 mm midsection will easily hold a 400-lb weight without yielding.
> Jobst's tests showed that a 1.8 mm midsection spoke yields at a bit
> over 500 pounds of tension.
> It seems unlikely that a rider using just one leg can raise the
> tension of all the pulling spokes (pre-tensioned to ~250 lbs) by
> another 250 pounds each.
> Even if Superman did stomp on the pedal, the tire would probably
> just spin, with the rubber losing traction before the spokes could
> be pulled hard enough to reach yield.
I think there is a factor of two missing in the assessment. Unless
"pushing" spokes rattle and loosen in this experiment, they carry half
the torque load by losing tension equal to the increase of pulling
spokes. Beyond that, Ron pointing out that thin hub tubes transmit
less torque to the left side of the wheel is a dodge. Today many if
not most hubs have a large enough torque tube to load spokes of both
sides of the wheel nearly equally. Some are even spokes radially on
the right side so that all torque is transmitted by the left spokes.
We haven't seen yielding spokes even with these oddballs or the FSA
three flange hub in which only the few spoke in the center flange are
tangentially laced. This whole subject is so half baked and flawed
that it seems to be spreading rumors. This has all been analyzed and
published many years ago in "the Bicycle Wheel". I take it that Ron
Ruff didn't check the literature before presenting his hypothesis.
Jobst Brandt
Ron Ruff
> Forgive my skepticism, but . . .
>
> If it were possible to yield spokes by stomping hard on the pedals,
> wouldn't we see spokes routinely when we grab the brakes?
>
> Any rider can (de)accelerate far faster than he can accelerate, for
> any range of speed, time, or distance
This is true for a *sustained* time period, but I'm only looking at an
instant here. From my experience when braking, an endo will happen
before I exceed the traction limit of my front tire on dry pavement,
so I think it would be possible to apply more torque to the rear wheel
with a stomp... since the high downward force will aid traction. Plus
you would need disc brakes for the braking force to effect your
spokes. I don't have a bike with disc brakes but from what I hear,
they *do* stress the spokes a lot, and these hubs tend to have large
flanges and barrels and crossed spokes on both sides.
> Another way to look at things is that a single butted spoke with a 1.8
> mm midsection will easily hold a 400-lb weight without yielding.
> Jobst's tests showed that a 1.8 mm midsection spoke yields at a bit
> over 500 pounds of tension.
That sounds about right. In the example I gave above with a 600lb
force the avg spoke dT was 1324N (298lb). If the initial tension was
less than this then all the pushing spokes would be slack, and the
pulling spokes would have to make up the difference on their own (ie
they would see more than a 1324N increase). But even if the initial
tension was 1324N, the pulling spokes would be seeing a 2648N load
(596lb).
> Even if Superman did stomp on the pedal, the tire would probably just
> spin, with the rubber losing traction before the spokes could be
> pulled hard enough to reach yield.
As I mentioned earlier, a coefficient of friction of ~.83 is what
would be necessary, assuming equal weight distribution on the front
and rear wheels.
Ron Ruff
> I think there is a factor of two missing in the assessment. Unless
> "pushing" spokes rattle and loosen in this experiment, they carry half
> the torque load by losing tension equal to the increase of pulling
> spokes.
I haven't made that mistake... all the crossed spokes are sharing the
change in load equally.
> Beyond that, Ron pointing out that thin hub tubes transmit
> less torque to the left side of the wheel is a dodge. Today many if
> not most hubs have a large enough torque tube to load spokes of both
> sides of the wheel nearly equally. Some are even spokes radially on
> the right side so that all torque is transmitted by the left spokes.
That is true, but plenty of wheels are configured with crossed spokes
on only one side, and the radial spokes then contribute almost nothing
to the torsional loads, regardless of which side of the hub they are
on.
> We haven't seen yielding spokes even with these oddballs or the FSA
> three flange hub in which only the few spoke in the center flange are
> tangentially laced.
That FSA RD-600 sure looks like a silly design! The center flange *is*
very large though, so the spokes would transfer torque with a
relatively small tension change. The strangest part is the same
"torque" flange on the front wheel... where it is completely useless.
> I take it that Ron
> Ruff didn't check the literature before presenting his hypothesis.
Well, I did study your book :). If there is data available that shows
test measurements of instantaneous pedal, crank, or spoke loads under
extreme conditions, then I'd love to see it. I couldn't find any on
the internet, though...
Ben C
> On Feb 14, 6:35 pm, carlfo...@comcast.net wrote:
>> Forgive my skepticism, but . . .
>>
>> If it were possible to yield spokes by stomping hard on the pedals,
>> wouldn't we see spokes routinely when we grab the brakes?
>>
>> Any rider can (de)accelerate far faster than he can accelerate, for
>> any range of speed, time, or distance
>
> This is true for a *sustained* time period, but I'm only looking at an
> instant here. From my experience when braking, an endo will happen
> before I exceed the traction limit of my front tire on dry pavement,
> so I think it would be possible to apply more torque to the rear wheel
> with a stomp... since the high downward force will aid traction. Plus
> you would need disc brakes for the braking force to effect your
> spokes.
A disk brake will transfer braking torque from the hub to the rim
through the spokes and a rim brake will not, but even with a rim brake
the braking force is still transferred from the rim to the fork through
the spokes.
This force would be 1000N for 1g deceleration of 100kg rider+bike-- not
a problem especially when spread over a few spokes.
I'm not really disagreeing, just quibbling.
jobst....@stanfordalumni.org
Rather than expend so many keystrokes to quibble, please review these
effect in "the Bicycle Wheel" this is shown graphically with numbers.
This has been available for more than 25 years.
Jobst Brandt
Ron Ruff
> Rather than expend so many keystrokes to quibble, please review these
> effect in "the Bicycle Wheel" this is shown graphically with numbers.
> This has been available for more than 25 years.
I've already done that in a previous post in this thread. Your FEA
listing doesn't give spoke tensions (in the 2nd edition at least) and
the graph only shows relative radial distortions. Obviously, during a
torque load the tangential displacement is the most important. Your
FEA gives tangential node displacements that are about 6 times lower
(in mm/kg) than the results of a test that you performed. If you can
explain why that should be so, I'd be interested to know.
carl...@comcast.net
Dear Ben & Ron,
I don't think that it will matter how briefly the force is
applied--any reasonable brake allows more (de)acceleration than
stomping on a pedal.
Try to move a rear wheel by stomping on the pedal if a well-adjusted
rear brake is squeezed.
Consider the kind of jolt caused by grabbing a reasonable brake.
Now compare that effect to the pitiful "jolt" of acceleration provided
by even the strongest sprinter heaving down on one pedal.
In any case, calculations that predict that a bicyclist can cause his
spokes to yield just by stomping on one pedal seem to be ignoring
reality.
If this were possible, local bike shops would notice customers
returning again and again with wobbly wheels that never stayed true
for more than one ride because the spokes kept lengthening.
If wheels stay true down a mountain trail, then the spokes are not
yielding under disk or rim braking.
If wheels stay true for a sprint, then the spokes are not yielding
under pedal stomping.
Cheers,
Carl Fogel
jobst....@stanfordalumni.org
>> Rather than expend so many keystrokes to quibble, please review
>> these effect in "the Bicycle Wheel" this is shown graphically with
>> numbers. This has been available for more than 25 years.
This comment was not aimed at your writing but while we are here...
> I've already done that in a previous post in this thread. Your FEA
> listing doesn't give spoke tensions (in the 2nd edition at least)
> and the graph only shows relative radial distortions. Obviously,
> during a torque load the tangential displacement is the most
> important. Your FEA gives tangential node displacements that are
> about 6 times lower (in mm/kg) than the results of a test that you
> performed. If you can explain why that should be so, I'd be
> interested to know.
I don't know what to what tests you are referring. The FEA stands on
its own. I think you should review the whole matter more closely. As
long as spokes do not slacken, their tension does not enter into their
elastic response nor their contribution to supporting loads. If you
used a finite element analysis, you must have noticed that preload
does not affect the outcome. These are not FEA of a collapsing wheel,
but rather one in working condition.
Jobst Brandt
D'ohBoy
D'ohBoy
Ron Ruff
> I don't know what to what tests you are referring.
2nd edition, page 131.
Ron Ruff
> Try to move a rear wheel by stomping on the pedal if a well-adjusted
> rear brake is squeezed.
That was a good idea Carl! But when I tried it I realized that
stomping on the pedal while holding the brake is requiring that I
overcome not only the force of the brake, but also the weight and
friction coefficient of the rear wheel on the ground. So unless you
are doing this in a stand (not possible), then it isn't a true test.
I was amazed at how much my bottom bracket flexed sideways when I
tried it!
> If this were possible, local bike shops would notice customers
> returning again and again with wobbly wheels that never stayed true
> for more than one ride because the spokes kept lengthening.
> If wheels stay true for a sprint, then the spokes are not yielding
> under pedal stomping.
There are several things that have to come together for the spoke load
to get this extreme:
Low spoke count
Small hub flange
Low gear
Extreme (stomping) force on the pedal
I expect that most riders *never* stomp on the pedals this hard in a
low gear... in fact I think it is safe to say that most riders don't
ever stomp on their pedals this hard in *any* gear. But as far as I
can tell it is certainly possible to do so, and if it is possible,
then surely some riders will be doing it... and they may be riding
wheels that aren't up to the task.
Do you think that *no* riders experience spokes breaking and loosening
on the drive side of a low spoke count wheel? I've heard several
reports of this.
> If wheels stay true down a mountain trail, then the spokes are not
> yielding under disk or rim braking.
They must be disc brakes for the comparison to be valid, and disk
brake hubs do not skimp on the flange radius or employ radial spoking
as rear road hubs do. In addition, if traction is sufficient, the
maximum braking force is limited by the initiation of an endo.
carl...@comcast.net
wrote:
>On Feb 15, 12:48 pm, carlfo...@comcast.net wrote:
Dear Ron,
I don't know of any riders who can permanently stretch stainless steel
spokes by stomping on the pedals or by braking.
If "surely some riders will be doing it," then spokes have yielded and
lengthened in use. That can be measured--the spokes will be longer.
It's been pointed out RBT that crashes involving sudden, powerful
forces simply rip the spoke nipples out of the rims. The spokes don't
yield and fail.
I'd love to see a post that reports spokes permanently lengthening
after being placed in a sprinter's wheel, but I doubt that it's going
to happen.
Cheers,
Carl Fogel
Michael Press
<1171529483....@q2g2000cwa.googlegroups.com>,
"Ron Ruff" <rruff...@yahoo.com> wrote:
> That is true, but plenty of wheels are configured with crossed spokes
> on only one side, and the radial spokes then contribute almost nothing
> to the torsional loads, regardless of which side of the hub they are
> on.
Has this calculation been done? As soon as a radial
spoke begins to wind up, it is transmitting torque.
--
Michael Press
Ron Ruff
> Has this calculation been done? As soon as a radial
> spoke begins to wind up, it is transmitting torque.
Of course it is *something*... but the torque transfer of the spokes
is proportional to the effective flange radius. In my model a 500N
force applied tangential to the rim produces ~6mm of deflection there,
with a rim radius of 330mm. The effective flange radius of the radial
spokes is then (6/330)*22 = 0.40mm. Or in other words the radial
spokes add 2% to the wheel stiffness... and that is assuming that the
hub has infinite torsional stiffness.
Ron Ruff
> I'd love to see a post that reports spokes permanently lengthening
> after being placed in a sprinter's wheel, but I doubt that it's going
> to happen.
As I've stated several times already Carl, and showed in a
calculation, a sprinter will not overstress the spokes *unless* he is
in a low gear.
Ben C
jobst....@stanfordalumni.org
>> That is true, but plenty of wheels are configured with crossed
>> spokes on only one side, and the radial spokes then contribute
>> almost nothing to the torsional loads, regardless of which side of
>> the hub they are on.
> Has this calculation been done? As soon as a radial spoke begins to
> wind up, it is transmitting torque.
That's correct for fully radially spoked, but with cross lacing on one
side there won't be significant hub rotation with respect to the rim,
so the effect is close to not there. But this brings up another
point. We have riders who have fully radial wheels and from riding on
them, I could not notice the small rotational wind-up when
accelerating. I haven't seen one of these lately and don't know what
happened to the wheels I rode, but it was interesting to note that
even there the rotation is so small it can't be felt.
I foresaw the main problem with those wheels was flange failure, and
that may have occurred later, but I never saw the wheels a few months
later. In any event, the spokes did not yield on acceleration. as Ron
Ruff would have us believe. Of course they weren't 16-spoke wheels
either.
I'm sure you can build a wheel that won't hold up but I don't thing
that is a reasonable goal.
Jobst Brandt
Ron Ruff
> What about a sprinter sprinting up a steep hill?
Yep... full power stomping sprint in a low gear... that is the
protocol.
What? Nobody has mucked up a low spoke count rear wheel by doing this?
Chalo... anybody?
I'm almost tempted to try it on the new wheels I'm building... but I'm
afraid I'll "succeed"...
Chalo
>
> Yep... full power stomping sprint in a low gear... that is the
> protocol.
>
> What? Nobody has mucked up a low spoke count rear wheel by doing this? Chalo... anybody?
Low spoke count wheel? What, like something with only 36 spokes? No,
not so far anyway.
Chalo
Ron Ruff
> rear brake is squeezed.
Carl, you gave me an idea... and I enlisted my wife to help me out
with a test. I tried just resting my weight on the horizontal pedal
and measuring the change in spoke tension on the drive side for both
leading and trailing spokes on the rear wheel. Unfortunately, all I
have at the moment is a beefy 32 spoke Shimano hub laced 3x drive side
and 2x NDS, and only a Park gauge to check tension, but the results
are in the ballpark at least. At first I couldn't believe it...
what?... the reading only changed by 1.5 marks? But after several
repeat trials and rotating the gauge to remove stiction, that was it.
The original tension was ~21.5 which translates to a tension of 115kg
with WS DB14 spokes. The change in tension per mark is 13 kg, so 1.5
marks is ~19.5kg... with an estimated resolution of ~+- 5kg.
I pulled up the spreadsheet, and put in a force of 150lb, or 667N
(even though I weigh 165lb, I figured something was lost from
steadying myself on the bars). Because the wheel has crossed spokes on
both sides, and the hub barrel is large, I had to guess on the
effective crossed spokes, and selected 24 as a reasonable compromise.
The gears were 34/26. The calculated tension change is 183N or
18.6Kg... so is very close to what would be expected. When I get some
wheels built that are more on the "marginal" side, I should be able
see a bigger effect.
IMO the only important unknown here is how much force I can really put
down on the pedal if I tried. It certainly seems like doubling my
weight would be easy... and I wouldn't be surprised if a 3 or 4x
multiple was possible for an instant.
carl...@comcast.net
wrote:
>On Feb 15, 12:48 pm, carlfo...@comcast.net wrote:
Dear Ron,
Hmmm . . .
I'm not sure that a wife is a good substitute for vises, workbenches,
ropes, weights, bathroom scales, and 2x4's, but all my married friends
were busy yesterday insisting that there are certain advantages, so I
must keep looking into the matter.
I'm just as envious about your tension gauge. My new one has been
languishing for three weeks now at the local university, where I'm
trying to inveigle engineering professors to use their tensile
machines and students to do some spoke testing.
Good luck with your experiments. We'll both be pleased if they show
spokes yielding under pedal forces, and there's little harm done if
your expectations aren't met.
A possible test would be to take a wheel that you don't mind
destroying (a junkyard might provide it), remove a spoke, measure it
carefully, and then try to yield it by tightening.
If the rim yields instead of the spoke, that would suggest that
stomping on a pedal isn't going to cause spokes to yield.
You might only find that the spoke wrench strips the nipple before
anything significant happens, but oil and a heavy weight on the axle
or rim could let you tighten the nipple to the point where something
interesting occurs when the weight is removed.
Obviously, there are some dangers, since the wheel, spoke, nipple, or
hub may fail under considerable tension, so be careful.
Possibly you should have Mrs. Ruff do the tightening, in imitation of
gangsters who take the sensible precaution of letting their spouses
start the car every morning:
"Breslin told us that one of the big responsibilities of a Mafia man's
wife is to start his car in the morning--to be sure it wasn't wired
with a bomb."
http://rogerebert.suntimes.com/apps/pbcs.dll/article?AID=/19711228/REVIEWS/112280301/1023
Happy Valentine's Day to you and your wife!
Carl Fogel
jobst....@stanfordalumni.org
> Good luck with your experiments. We'll both be pleased if they show
> spokes yielding under pedal forces, and there's little harm done if
> your expectations aren't met.
> A possible test would be to take a wheel that you don't mind
> destroying (a junkyard might provide it), remove a spoke, measure it
> carefully, and then try to yield it by tightening.
> If the rim yields instead of the spoke, that would suggest that
> stomping on a pedal isn't going to cause spokes to yield.
> You might only find that the spoke wrench strips the nipple before
> anything significant happens, but oil and a heavy weight on the axle
> or rim could let you tighten the nipple to the point where something
> interesting occurs when the weight is removed.
> Obviously, there are some dangers, since the wheel, spoke, nipple,
> or hub may fail under considerable tension, so be careful.
Let's look at the experience before us. First, you have a text that
investigated these forces and found them to be small in comparison to
radial loads of JRA (just riding along). We also know that a spoke
routinely stressed near yield fails in fatigue soon. MTB riders also
routinely use a 20t CW and a 34t+ sprocket and are often heavy and
strong as they climb over moguls and the like exerting more torque on
the pedals that sprinters.
There are not a raft of spoke failures arising from these actions, so
there must be something else at work that Mr. Ruff is calculating. I
realise that a suitably de-populated spoke complement on a small
flange hub with an old classic noodle for a central shaft could raise
spoke tension to damaging levels, but why are we trying to concoct
such a flimsy wheel?
Jobst Brandt
steve
> On 15 Feb 2007 20:05:52 -0800, "Ron Ruff" <rruffrr...@yahoo.com>
>
> Happy Valentine's Day to you and your wife!
>
>
> - Show quoted text -
Dear Carl,
If you had any particular experiment in mind I do have access to a
tensile tester and multiple tension gages (park, WS, FSA, DT). If you
are unable to fanagle your way into the university I would be more
than happy to help. Good luck!
Steve Sauter
Ron Ruff
> MTB riders also
> routinely use a 20t CW and a 34t+ sprocket and are often heavy and
> strong as they climb over moguls and the like exerting more torque on
> the pedals that sprinters.
MTBs certainly do have low gears, and the ratio of hub torque to crank
torque will be proportional to this. But... in the example I gave we
are already right on the edge of traction with a 34/27 gear, so an
even lower gear with an equivalent stomp will result in wheelspin,
even on pavement. In the dirt the wheel will spin much easier.
Ron Ruff
> I'm not sure that a wife is a good substitute for vises, workbenches,
> ropes, weights, bathroom scales, and 2x4's, but all my married friends
> were busy yesterday insisting that there are certain advantages, so I
> must keep looking into the matter.
I have been with some women that I would have happily traded for a
well-equiped shop... but not this one... I feel that she would be
impossible to replace...
> Good luck with your experiments. We'll both be pleased if they show
> spokes yielding under pedal forces, and there's little harm done if
> your expectations aren't met.
Ideally, I'd like to be able to measure the spoke tension while
stomping, but haven't figured out a decent way to do that, since the
time interval of the stomp is much shorter than the time it takes to
get a good reading on the gauge.
carl...@comcast.net
>Dear Carl,
>If you had any particular experiment in mind I do have access to a
>tensile tester and multiple tension gages (park, WS, FSA, DT). If you
>are unable to fanagle your way into the university I would be more
>than happy to help. Good luck!
>
>Steve Sauter
Dear Steve,
I just emailed you an explanation of what I'm trying to do to some
helpless spokes.
If the question interests you and suits your tensile tester, I can
ship you some spokes to pull apart.
Thanks for your kind offer--but please remember that you're certainly
not obliged to perform bizarre and tedious experiments unless they
appeal to you.
Cheers,
Carl Fogel
joseph.sa...@gmail.com
Or a random fat guy? On seated up-hill stomps, wheelies is my problem,
while for standing stomps if it is steep enough to require the 39x25 I
have on several occasions had problems with traction. I always figured
a pedal spindle to be the weak link in a gorilla sized stomp, if the
traction was there.
Joseph