Claims about BB/Crank Stiffness (XT)
Ian G Batten
Grand claims are made that the new XT Hollowtec cranks and ``Big Pipe''
BB are very stiff. In practical terms, is this true, and does the
stiffness of bottom brackets and cranks make a significant difference to
the rider?
ian
S. Anderson
news:bv5hvt$7he$1...@news-out.ftel.co.uk...
Well, stiffer is better. I don't know how much stiffer they are, or if it's
worth worrying about though. But if you can make it stiffer and you're
upgrading anyway, it's better that way. If you have an old-style and you're
thinking of upgrading solely for this stiffer claim, I wouldn't bother.
Old-style BB's have been around for a long time and they don't, IMHO, seem
to be lacking in stiffness for most people. I'd be more worried about the
pedal eyelet breaking.
Cheers,
Scott..
Qui si parla Campagnolo
BB are very stiff. In practical terms, is this true, and does the
stiffness of bottom brackets and cranks make a significant difference to
The only place stiffness differences can actually be seen on crank arms and BB
spindles is on a test machine and these differences are small.
If anybody rides a bike with this crank or that and comments on how stiff it
feels, or how flexible, they are either feeling the frameset or the wheels or
tires or something but they are not feeling crankarms or steel BB spindles flex
or not.
If they were the next thing they would be feeling is something breaking.
Peter Chisholm
Vecchio's Bicicletteria
1833 Pearl St.
Boulder, CO, 80302
(303)440-3535
http://www.vecchios.com
"Ruote convenzionali costruite eccezionalmente bene"
Werehatrack
Under competition-level hard pedalling, the stiffer cranks and BB are
less susceptible to deflection that can cause unwanted chain behavior.
Yes, for riders capable of generating the kinds of force that can
deflect the chainwheels enough, it's an advantage that's probably
worth the expense. For the pleasure-rider or punter like me, it's
probably not.
--
My email address is antispammed; pull WEEDS if replying via e-mail.
Yes, I have a killfile. If I don't respond to something,
it's also possible that I'm busy.
Words processed in a facility that contains nuts.
Werehatrack
Campagnolo) may have said:
>Ian-<< Grand claims are made that the new XT Hollowtec cranks and ``Big Pipe''
>BB are very stiff. In practical terms, is this true, and does the
>stiffness of bottom brackets and cranks make a significant difference to
>the rider? >><BR><BR>
>
>The only place stiffness differences can actually be seen on crank arms and BB
>spindles is on a test machine and these differences are small.
>
>If anybody rides a bike with this crank or that and comments on how stiff it
>feels, or how flexible, they are either feeling the frameset or the wheels or
>tires or something but they are not feeling crankarms or steel BB spindles flex
>or not.
>
>If they were the next thing they would be feeling is something breaking.
By comparison to high-end conventional stuff, that's probably an
accurate assessment. By comparison to low-end stuff under the strain
of a 6'3" 220-lb rider standing to pedal hard, well, I've seen the
chainwheels deflect enough to cause hard contact with the front der
cage. Swapping from the no-name crank to an old Deore LX crank was
enough to keep it from happening.
There is definitely a point of diminishing returns in such areas, and
I think Shimano passed it a while back. The high-end bike component
market is now starting to resemble the athletic shoe business; nearly
all hype of buzzword features that are of questionable value to the
majority of potential purchasers. (Not that the products are *bad*,
mind you, but just that there's no good reason to spend the extra
money for the merely trendy features most of the time.)
bbense+rec.bicycl...@telemark.slac.stanford.edu
In article <n43d10d5sknv8n7rn...@4ax.com>,
Werehatrack <rau...@earthWEEDSlink.net> wrote:
>On 27 Jan 2004 13:59:22 GMT, vecc...@aol.com (Qui si parla
>Campagnolo) may have said:
>
>>Ian-<< Grand claims are made that the new XT Hollowtec cranks and ``Big Pipe''
>>BB are very stiff. In practical terms, is this true, and does the
>>stiffness of bottom brackets and cranks make a significant difference to
>>the rider? >><BR><BR>
>>
>>The only place stiffness differences can actually be seen on crank arms and BB
>>spindles is on a test machine and these differences are small.
>>
>
[snip]
>
>There is definitely a point of diminishing returns in such areas, and
>I think Shimano passed it a while back. The high-end bike component
>market is now starting to resemble the athletic shoe business; nearly
>all hype of buzzword features that are of questionable value to the
>majority of potential purchasers. (Not that the products are *bad*,
>mind you, but just that there's no good reason to spend the extra
>money for the merely trendy features most of the time.)
>
_ Read Keith Bontrager's rather long rant about "stiffness",
very interesting.
http://www.bontrager.com/keith/index.asp?ck=0&fl=1
_ Booker C. Bense
-----BEGIN PGP SIGNATURE-----
Version: 2.6.2
iQCVAwUBQBav02TWTAjn5N/lAQHbpAP/eNAVDAKti3DPaE/tZafRgQmwY5kHJROI
ReOZtSU1ee10SrHCZzZhUcpucsoQBI8IVO9+RaOCgPJVi5StSZKtpMT/Z2NMyOAs
93IU6DZKXr5F2DDd9a8sofdx1qQCo/2PlS3EtEHT+WDZE30lhIQnfz3SVvrHaUXt
/jaCGgNGuao=
=3ioS
-----END PGP SIGNATURE-----
A Muzi
I do not know.
Tonight at a Shimano tech show, the speaker said that
because frames have gotten so light and flexible, the
stiffer new Shimano crank returns stiffness to previous levels.
With a straight face.
Me? I can't flex my TA Cyclotouriste single ring enough to
notice.
--
Andrew Muzi
www.yellowjersey.org
Open every day since 1 April, 1971
Mark Hickey
>Tonight at a Shimano tech show, the speaker said that
>because frames have gotten so light and flexible, the
>stiffer new Shimano crank returns stiffness to previous levels.
>
>With a straight face.
Heh. My favorite Shimano pitch was at Interbike back in the mid-80's.
The pitch man was explaining how Biopace was the end-all for cyclists
and would make them healthy, wealthy and wise... but for your more
discriminating boy racer customers, Shimano had once again led the
innovative push toward (get this...) a ROUNDER chainring for the Dura
Ace gruppo.
Eventually they invented the totally round chainring. Again.
>Me? I can't flex my TA Cyclotouriste single ring enough to
>notice.
I dream of the day when I can flex a crank enough to notice.
Mark Hickey
Habanero Cycles
http://www.habcycles.com
Home of the $695 ti frame
Qui si parla Campagnolo
stiffer new Shimano crank returns stiffness to previous levels.
With a straight face.
I don't get invited to these things when they come to Boulder...I think they
know what I would be saying..Tough questions that the lizards have no answers
for.
jim beam
guys are biting...
larger diameter pipe spindles /are/ stiffer, both in torsion & bending.
that's fact. [i discussed some of the torsion math with chalo a while
back.]
there is also a debate on whether it's noticeable; in my personal and
highly subjective judgment, and as a clydesdale, i definitely think so.
when i first put an xt hollowtech crank on my commuter, i was new to
this new fangled spindle stuff and not expecting any difference, but my
route home from work takes me up a certain san francisco cable car grade
and it was immediately apparent that there was less flex.
does it make a difference? probably not to efficiency because the flex
is effectively 100% elastic. does it help a maladjusted front
derailleur not to grind? yes. does it "feel" better? yes, in my opinion.
but all this is irrelevant to what's really important: the fact that
hollow spindles [and cranks] allow a lighter component to be made with
similar or slightly better mechanical properties than solid ones. just
like larger diameter thinner wall frame tube on big frames is a good
thing - hollow components allow the drive train to be made with the same
or slightly better stiffness but less weight.
less weight is good therefore hollow is good.
jb
Ian G Batten
jim beam <u...@ftc.gov> wrote:
> in my opinion, the o.p. has a troll coefficient of 0.8, but since you
> guys are biting...
It wasn't intended as such. A bike I am consider buying (Marin Mount
Vision, for reasons unrelated to its bottom bracket) has Shimano new
XT. The salesman, in passing, made claims of stiffness. I was curious
as to the impact of this.
ian
Qui si parla Campagnolo
Nope, items of less weight are...lighter.
Chalo
> Tonight at a Shimano tech show, the speaker said that
> because frames have gotten so light and flexible, the
> stiffer new Shimano crank returns stiffness to previous levels.
>
> With a straight face.
>
> Me? I can't flex my TA Cyclotouriste single ring enough to
> notice.
Well, if Shimano is restoring a bit of stiffness to their bottom
brackets, it's stiffness they themselves took away with the advent of
cartridge BBs.
I was a shop mechanic when cartridge BBs hit the market. When I went
out to test ride the first bike I had built that used one, I thought
I'd broken it with the first few pedal strokes. The flex was so much
more than what I was attuned to, that it felt as if the pedals were
swinging under the bike past the centerline. The Specialized
Rockhopper CrMo frame was of utterly normal status-quo construction,
so what I was feeling was the bottom bracket itself.
I can tell the difference in stiffness between different BBs, clear as
day. I have a single-speed whose sprockets catch on the chain's
sideplates if I pedal too hard. That's a problem IMO, and it's
partially related to BB spindle and bearing flex.
Chalo Colina
Mike Jacoubowsky
> notice.
That surprises me. Well, no it doesn't, because it's a single ring, and
where I notice the difference between a stiffer and less-stiff crank is in
chain rub on the front derailleur. You have no front derailleur, thus no
chain rub to serve as an indication of chain rub.
Larger, stronger riders most definitely notice the symptom of flex in the
system. Whether it's from the bottom bracket (frame construction), axle or
chainrings... the problem of flexing enough to cause grief with chain rub on
the derailleur is real, not imaginary. Anything done to stiffen up those
areas of the bicycle is a good thing in my book.
--Mike-- Chain Reaction Bicycles
http://www.ChainReactionBicycles.com
Qui si parla Campagnolo
the derailleur is real, not imaginary. Anything done to stiffen up those
True but the 'flex' isn't in the aluminum crankarm or the 'twist' of the steel
spindle. If that 'flex' was that noticable, the next thing noticed would be the
spindle or the aluminum crankarm breaking.
As I said before, these two items can be tested and compared with other
crankarms and BB spindles on a flex machine, that exerts a bunch more force
than any rider...and then these values or differences found on the machine are
small, small.
For a rider to say he feels arms bend or spindles twist is just bugleoil.
jim beam
see from this thread, opinions differ on its relevance.
xt is great kit. won't let you down. have fun!
Rick Onanian
>the derailleur is real, not imaginary. Anything done to stiffen up those
>areas of the bicycle is a good thing in my book. >><BR><BR>
Campagnolo) wrote:
>True but the 'flex' isn't in the aluminum crankarm or the 'twist' of the steel
>spindle. If that 'flex' was that noticable, the next thing noticed would be the
>spindle or the aluminum crankarm breaking.
>
I have no opinion on the issue of spindle or arm flex, but I _have_
observed a rub between the front derailer and chain as Mike
described when putting a lot of force unevenly into the pedals, as I
sometimes do in tall gears. I weigh 210 pounds and often mash very
tall gears.
That said, what else could cause that rub? Frame flex? If it only
happened on my road bike, a plausible explanation could be that my
hands are flexing the hoods so much as to change the tension in the
derailer cable; but it has happened on my mountain bike too.
>Peter Chisholm
--
Rick Onanian
jlscott3
> where I notice the difference between a stiffer and less-stiff crank
> is in chain rub on the front derailleur. [...] the problem of flexing
> enough to cause grief with chain rub on the derailleur is real, not
> imaginary. Anything done to stiffen up those areas of the bicycle is a
> good thing in my book.
FD chain rub is something that I've always attributed exclusively to
frame flex in the BB shell area. It seems unlikely to me that a short
steel (or Ti) cylinder supported near both ends is going to flex more
than the frame. But, I'm just supposing here. The crank transmits torque
to the BB spindle but it will also apply some bending force...has
anybody (outside of Shimano's R&D lab) modeled or measured that?
Anybody else with anecdotes about how replacing a conventional BB with a
large-diameter hollow spindle has reduced chain rub?
JLS
--
Chalo
> For a rider to say he feels arms bend or spindles twist is just bugleoil.
The hell! When I mash down on a platform pedal attached to a regular
aluminum crank and a square-taper BB, I can not only feel the sucker
sag down under my foot, I can _see_ it sag down. Some of that sag is
in the spindle, and some in the frame, but a lot of it is due to flex
in the crank arm and spindle.
I have broken square-taper spindles before. Do you suggest that they
broke without ever having flexed noticeably?
I am not claiming that Octalink or Isis are significantly better in
that regard, but there is a big difference in stiffness between a
traditional crank and BB and one that is built for rigidity (Bullseye,
Primo, Profile, Redline Flite, etc.)
Chalo Colina
John Dacey
On 29 Jan 2004 13:50:56 GMT, vecc...@aol.com (Qui si parla
Campagnolo) wrote:
>Mike J-<< the problem of flexing enough to cause grief with chain rub on
>the derailleur is real, not imaginary. Anything done to stiffen up those
>areas of the bicycle is a good thing in my book. >><BR><BR>
>
>True but the 'flex' isn't in the aluminum crankarm or the 'twist' of the steel
>spindle.
Where do you suppose it is, then?
>If that 'flex' was that noticable, the next thing noticed would be the
>spindle or the aluminum crankarm breaking.
Not necessarily. People have commented for years about flex issues
with frames, brake calipers, handlebar stems and other parts. Simply
identifying that flex was happening didn't always mean that failure of
the subject part was imminent.
>As I said before, these two items can be tested and compared with other
>crankarms and BB spindles on a flex machine, that exerts a bunch more force
>than any rider...and then these values or differences found on the machine are
>small, small.
Can you point to a test that shows this? The last such comparison I
recall was done by a magazine and it occurred not just a few years ago
(latter '80s?). It well preceded any current model road or mountain
crank from Shimano or Campagnolo.
>For a rider to say he feels arms bend or spindles twist is just bugleoil.
Well (to keep with your brassy metaphor), I think you can blow that
one out your spitvalve. A couple of months ago, I fitted my own bike
with a set of the new Dura-Ace 7800 cranks. Only the crank was changed
(and the bearings to match); I even used my old chainrings. The
outgoing crank (Dura-Ace spined) was somewhat cosmetically challenged;
but it, and its bottom bracket (Ultegra splined) were mechanically
uncompromised.
With the new crank, when moving away from a stop, the first eight or
ten standing pedal strokes have a distinctly different feel. I don't
harbour any delusions about how much free speed I've found, but that
there has been a reduction in the amount of sag with the new cranks is
unambiguous. It's apparent even with relatively modest efforts, which
leads me to think that much of what I feel is due to the relocation of
the 7800's crank bearings outboard of the bottom bracket shell.
-------------------------------
John Dacey
Business Cycles, Miami, Florida
Now in our twenty-first year.
Our catalogue of track equipment: eighth year online.
http://www.businesscycles.com
Jose Rizal
> vecc...@aol.com (Qui si parla Campagnolo) wrote:
>
> > For a rider to say he feels arms bend or spindles twist is just bugleoil.
>
> The hell! When I mash down on a platform pedal attached to a regular
> aluminum crank and a square-taper BB, I can not only feel the sucker
> sag down under my foot, I can _see_ it sag down. Some of that sag is
> in the spindle, and some in the frame, but a lot of it is due to flex
> in the crank arm and spindle.
> I have broken square-taper spindles before. Do you suggest that they
> broke without ever having flexed noticeably?
I think the opposite is what Pete has said many times before, that
noticeable flex will result in broken parts soon enough.
An interesting question for you is, how long do your cranks last with
noticeable flexing before you break them?
Jose Rizal
> >If that 'flex' was that noticable, the next thing noticed would be the
> >spindle or the aluminum crankarm breaking.
>
> Not necessarily. People have commented for years about flex issues
> with frames, brake calipers, handlebar stems and other parts. Simply
> identifying that flex was happening didn't always mean that failure of
> the subject part was imminent.
It is with aluminum components. Flex is not a good thing with Al.
> >For a rider to say he feels arms bend or spindles twist is just bugleoil.
>
> Well (to keep with your brassy metaphor), I think you can blow that
> one out your spitvalve. A couple of months ago, I fitted my own bike
> with a set of the new Dura-Ace 7800 cranks. Only the crank was changed
> (and the bearings to match); I even used my old chainrings. The
> outgoing crank (Dura-Ace spined) was somewhat cosmetically challenged;
> but it, and its bottom bracket (Ultegra splined) were mechanically
> uncompromised.
>
> With the new crank, when moving away from a stop, the first eight or
> ten standing pedal strokes have a distinctly different feel. I don't
> harbour any delusions about how much free speed I've found, but that
> there has been a reduction in the amount of sag with the new cranks is
> unambiguous. It's apparent even with relatively modest efforts, which
> leads me to think that much of what I feel is due to the relocation of
> the 7800's crank bearings outboard of the bottom bracket shell.
If true, it then means that what you felt was axial bending of the
spindle, not cranks bending or spindles twisting.
What's contentious is the notion that aluminum components in cranks and
BB are noticeably felt to flex without the components breaking soon
enough. This just isn't possible, hence the attribution of most flex to
something else.
Paul Kopit
wrote:
>That said, what else could cause that rub? Frame flex? If it only
>happened on my road bike, a plausible explanation could be that my
>hands are flexing the hoods so much as to change the tension in the
>derailer cable; but it has happened on my mountain bike too.
Stand perpendicular to your bicycle holding the seat and bars. Put
your foot on the pedal and push. You will see how little force it
takes to make the bb move away from you. That would make the front
derailleur rub. As far as crankarm stifness, I would guess that a
piece of wood that replaced the crankarm would still let you flex the
bicycle at the bb.
Mark Hickey
>Stand perpendicular to your bicycle holding the seat and bars. Put
>your foot on the pedal and push. You will see how little force it
>takes to make the bb move away from you. That would make the front
>derailleur rub. As far as crankarm stifness, I would guess that a
>piece of wood that replaced the crankarm would still let you flex the
>bicycle at the bb.
The trouble with this test is that the tires and wheels flex MUCH more
than the rest of the bike, giving a very false impression of what is
flexing. Slap bare hubs in the frame and fork, support them on
something solid and try the test again - I think you'll be amazed "how
much stiffer the frame got"... ;-) Heck, you should probably support
the bike by the rear dropouts and the head tube if you want to
eliminate the fork, which WILL flex laterally.
I once showed a LBS guy one of my (fully built) MTBs. He did the test
you mention, and commented about how flexy the frame was. I guess
fact he was pushing against the 2.1" low pressure tires didn't occur
to him... ;-)
Mark Hickey
>FD chain rub is something that I've always attributed exclusively to
>frame flex in the BB shell area. It seems unlikely to me that a short
>steel (or Ti) cylinder supported near both ends is going to flex more
>than the frame. But, I'm just supposing here. The crank transmits torque
>to the BB spindle but it will also apply some bending force...has
>anybody (outside of Shimano's R&D lab) modeled or measured that?
The BB shell is plenty stiff, and I think it's safe to say that they
don't flex enough even for Chalo to notice. The flex that occurs is
when the BB shell deflects due to the seat tube (primarily) and down
tubes not being able to keep it from doing so. The amount of flex in
the seat tube between the BB shell and the front derailleur is what
determines if the FD cage will rub or not.
>Anybody else with anecdotes about how replacing a conventional BB with a
>large-diameter hollow spindle has reduced chain rub?
Does the BB spindle and crank flex somewhat? I'm sure they do, but
you can get a relative idea of the significance of BB/crank vs. frame
flex by comparing the two following theories.
#1) If frame flex is the predominant problem, you'll have anecdotal
evidence of a large number of a particular type of frame exhibiting FD
rubbing regardless of what type of crank and BB is installed.
#2) If BB/crank flex is the predominant problem, you'll have anecdotal
evidence of a large number of particular BB/cranks causing FD rubbing
regardless of what type of frame they're installed on.
#1 is clearly true, while I don't think I've heard anyone make the
claim in theory #2.
jlscott3
>> jlscott3 <usenet...@cyclingforums.com> wrote:
>>FD chain rub is something that I've always attributed exclusively to
>>frame flex in the BB shell area. It seems unlikely to me that a short
>>steel (or Ti) cylinder supported near both ends is going to flex more
>>than the frame. But, I'm just supposing here. The crank transmits
>>torque to the BB spindle but it will also apply some bending
>>force...has anybody (outside of Shimano's R&D lab) modeled or
>>measured that?
> The BB shell is plenty stiff, and I think it's safe to say that they
> don't flex enough even for Chalo to notice. The flex that occurs is
> when the BB shell deflects due to the seat tube (primarily) and down
> tubes not being able to keep it from doing so. The amount of flex in
> the seat tube between the BB shell and the front derailleur is what
> determines if the FD cage will rub or not.
Yup, we're in violent agreement on that one, though you phrased it much
better than I did. The double-diamond frame isn't very well
triangulated to resist flex in that direction, and I think FD-rub-
causing flex is mostly in the seat tube. It seems to me that if the BB
spindle flexed enough to cause rub, we'd see bearings bind or some
other really nasty symptom.
I'm working on something that will model the forces applied to the BB
spindle, I hope I'll have something to report in a few days. If we're
going to argue about it we may as well have some data...
JLS
--
Carl Fogel
Dear Jose, John, and Chalo,
What's going on with the two sets of bearings
and races in the pedal under load? And the two
sets of bearings and races supporting the spindle
under the same load?
If the spindle and arm are bending significantly,
as some people say, are the bearings suffering,
too? The bending force has to be transmitted from
the foot through them, too, doesn't it?
Do the bearings fail at the same time, or does
something protect them that is going to fascinate
me?
Or are the failures of cranks and spindles more
a matter of fatigue, corrosion, or manufacturing
cracks unrelated to the kind of bending fatigue
that's been mentioned?
Come to think of it, do the pedal shafts bend
significantly, too?
I'm curious because I recently read Bontrager's
comments about stiffness, where he says that in
a system of springs of different stiffness, the
whole system functions only a little more stiffly
than its softest spring. (Not a theory embraced
by people who believe that different frame materials
affect the harshness of the ride of rubber-tired
vehicles.)
What's the softest spring, so to speak, in this
system, the one that supposedly should bend quite
a bit before any other component begins to bend?
I keep coming back to the frame, but I'm interested
in what you're all saying.
Has anyone shoved a crankset in a vise and tested
it to see what heaving on the pedal actually bends?
I have one that I may be willing to sacrifice and
a fair-sized vise, but it seems better to wait for
suggestions about how to abuse it.
Carl Fogel
Qui si parla Campagnolo
Sure, the BB shell is 'suspended' below the front der and can move left and
right, affecting the distance between the ring and the front der.
Qui si parla Campagnolo
aluminum crank and a square-taper BB, I can not only feel the sucker
sag down under my foot, I can _see_ it sag down. Some of that sag is
in the spindle, and some in the frame, but a lot of it is due to flex
garbage....you are not 'bending', 'twisting' or 'flexing' the steel spindle or
aluminum crankarm...
Chump-<< I have broken square-taper spindles before. Do you suggest that they
broke without ever having flexed noticeably? >><BR><BR>
you bet, probably every one.
jim beam
<snip>
> you are not 'bending', 'twisting' or 'flexing' the steel spindle or
> aluminum crankarm...
with respect, that's not a technically accurate statement. just because
it may not be visible doesn't mean it's not measurable or relevant.
they bend elastically even under the tiniest loads. and that bending is
part of their fatigue scenario.
jb
Chalo
> When I mash down on a platform pedal attached to a regular
> aluminum crank and a square-taper BB, I can not only feel the sucker
> sag down under my foot, I can _see_ it sag down.
and when I said:
> Some of that sag is
> in the spindle,
I meant the _pedal_ spindle
> and some in the frame, but a lot of it is due to flex
> in the crank arm and spindle.
as opposed to the crank spindle referenced here.
Chalo Colina
Qui si parla Campagnolo
it may not be visible doesn't mean it's not measurable or relevant.
they bend elastically even under the tiniest loads. and that bending is
Yes and glass is a liquid and is 'flowing'. Crank arm flex and BB spindle twist
is barely measurable, even with test equipment, and IMO, is not relevant only
to the marketing claim that one crank/BB combo is 'stiffer' than another,
meaning it is the one to get.
Ya know, a plastic spoiler on a compact car exerts downforce when ya drive it
at 55-75 MPH, but is it significant or relevant?...no....
Chalo
> Chump-<< I have broken square-taper spindles before. Do you suggest that they
> broke without ever having flexed noticeably? >><BR><BR>
>
> you bet, probably every one.
Crank spindles are not magic; they are made of metal, like any other
machine part. You can't stress them to failure without having
stressed them to a high level of strain (manifested as flex).
You seem to have an inflated sense of the structural integrity of
these parts, which are small, after all. The tip of a square taper
spindle is smaller than the tip of my pinky finger, and it's drilled
out to boot. And I have seen no recent road or MTB crankarm that was
thicker than about a half inch. Those suckers wind up longitudinally
under pedal loading; if you believe they don't, you're fooling
yourself.
You could just observe what I'm talking about by having a large rider
hold the brakes on a floor model while pushing down on a pedal in the
forward position. If you sight down the length of the arm, you'll see
both the windup and the spindle deflection.
Chalo Colina
Appkiller
during high load situations will result in the following observations:
1. The bb shell moves (if you get any motion at all).
2. The crankset displacement equals the bb shell displacement.
3. The fder does not move, nor does any frame component north of
there.
Summed up: the frame is flexing in the area of the bb. The crank is
not.
And if you pay too close attention, you will get to observe the
asphalt close-up too.
App
Chalo
> What's going on with the two sets of bearings
> and races in the pedal under load? And the two
> sets of bearings and races supporting the spindle
> under the same load?
Yes, and they are elastic too. Under load, bearing balls "squish" and
rebound, and raceways indent underneath them. Just like your tires on
the pavement, bearing balls must establish an elliptical "contact
patch" to support a load.
> If the spindle and arm are bending significantly,
> as some people say, are the bearings suffering,
> too? The bending force has to be transmitted from
> the foot through them, too, doesn't it?
There is a measure of ball-and-socket type motion in a cup & cone
bearing or angular contact cartridge bearing. Jobst Brandt has
observed that ideally, the cups and cones of a traditional BB would be
reversed, with the cups inboard to facilitate such motion. That would
impose difficulties in assembly, of course. :-)
> Do the bearings fail at the same time, or does
> something protect them that is going to fascinate
> me?
There is enough capacity in most BB bearings to withstand even the
forces that cause cranks and spindles to fail, provided the
lubrication holds up.
> Come to think of it, do the pedal shafts bend
> significantly, too?
Yes, and the differences between them can be felt too.
> I'm curious because I recently read Bontrager's
> comments about stiffness, where he says that in
> a system of springs of different stiffness, the
> whole system functions only a little more stiffly
> than its softest spring....
>
> What's the softest spring, so to speak, in this
> system, the one that supposedly should bend quite
> a bit before any other component begins to bend?
> I keep coming back to the frame, but I'm interested
> in what you're all saying.
Frames vary in their stiffness at the BB much more than BB spindles
do. If you ride some featherweight CFRP marvel, then surely the BB
sway in the frame would overshadow that in the crank assembly. I ride
mostly big-tube frames in both steel and aluminum, and the differences
between cranks are stark.
Even with my Surly 1x1, a conventional MTB frame by any reasonable
standard, I observed big feel differences by swapping cranks and BB.
I started with a Truvativ ISIS DH bottom bracket, heavy walled and at
least as stiff as any English threaded unit BB around. My first crank
was a Truvativ Stylo SS, but it wagged around so freely I was scared
by it. So I swapped to a Truvativ Hussefelt DH crank on the same BB,
and there was a noticeable improvement.
I would have left it at that, but after having crank bolt loosening
problems, I swtched to a Redline Dual Slalom tubular cromoly crank
with splined spindle and a dedicated bottom bracket. Even compared to
the beefiest ISIS crank on the market at that time, the improvement in
stiffness was easily discerned.
I think that Peter should try a Bullseye crank on his most familiar
ride before poo-poohing the notion that there are noticeable crank
stiffness differences. Heck, I think everybody should try a Bullseye
crank. It ain't for no reason that Shimano flagrantly ripped off the
basic design of the Bullseye once the relevant patents had expired.
> Has anyone shoved a crankset in a vise and tested
> it to see what heaving on the pedal actually bends?
> I have one that I may be willing to sacrifice and
> a fair-sized vise, but it seems better to wait for
> suggestions about how to abuse it.
That would be a valuable way to do some comparison among different
systems. Sounds like a job for Kraig Willett. I'd be willing to lend
a variety of Very Mighty cranks to whoever undertook such a study.
Chalo Colina
jobst....@stanfordalumni.org
> What's going on with the two sets of bearings and races in the pedal
> under load? And the two sets of bearings and races supporting the
> spindle under the same load?
> If the spindle and arm are bending significantly, as some people
> say, are the bearings suffering, too? The bending force has to be
> transmitted from the foot through them, too, doesn't it?
What you mean "suffering"? This is an elastic system and regardless
of the weakest elements, under steady state (which is essentially it
with pneumatic tires between road shock and metal parts) with no
significant spike loads. Just imagine a series of tension springs
hooked up in series hanging from a hook. Each one stretches according
to its elasticity more or less but the same as it would alone under
the same load.
> Do the bearings fail at the same time, or does something protect
> them that is going to fascinate me?
Yes. The bearings, or any element in the combination wears the same
as it would entirely alone. As I said, once there are pneumatic tires
between road and bicycle it is essentially a static load since no
significant spikes can get past the tires and the human engine is too
soft to cause any without using a hammer or the like. That's why we
use tools.
> Or are the failures of cranks and spindles more a matter of fatigue,
> corrosion, or manufacturing cracks unrelated to the kind of bending
> fatigue that's been mentioned?
How about refining "the kind of bending fatigue that's been
mentioned". This is all fatigue except the forced ruptures that occur
from overload of a structural element. What is not mentioned here is
that composites have internal fatigue that looks different from metal
failures in that the matrix dissolves and leaves the element with a
bunch of free fibers that we commonly know as cloth in clothing. This
is before anything breaks but the structure has been compromised.
That is the basic problem with composites. The don't show crack
propagation.
> Come to think of it, do the pedal shafts bend significantly, too?
Yes???
> I'm curious because I recently read Bontrager's comments about
> stiffness, where he says that in a system of springs of different
> stiffness, the whole system functions only a little more stiffly
> than its softest spring. (Not a theory embraced by people who
> believe that different frame materials affect the harshness of the
> ride of rubber-tired vehicles.)
The whole system functions MORE ELASTICALLY than the softest
element... with the definition meaning total displacement of that
element. That is, the size of the element also plays into the total
deflection of, for instance, the pedal when standing on it. In that
case the frame of most bicycles is the greatest deflection as you can
easily see statically by pushing on one pedal when standing next to
the bicycle.
> What's the softest spring, so to speak, in this system, the one that
> supposedly should bend quite a bit before any other component begins
> to bend? I keep coming back to the frame, but I'm interested in
> what you're all saying.
Not so. They all flex simultaneously. As I said, consider it a
series of coil springs hooked in series as you load the end one.
> Has anyone shoved a crankset in a vise and tested it to see what
> heaving on the pedal actually bends? I have one that I may be
> willing to sacrifice and a fair-sized vise, but it seems better to
> wait for suggestions about how to abuse it.
That depends on whose crank (aka crank-set-arm) you choose. The old
aluminum cranks with a vanity groove down the middle were
substantially softer in torsion from standing on the pedal than in
bending. Torsional stiffness being approximately that of a crank of
the largest inscribed circle within the actual cross section. Even
without this feature, most are probably still more torsionally soft
from outboard pedal loads than in crank bending.
Bearing failures generally take the form of spalling, the cracking off
of outer hardness layers of both balls and races. These are fatigue
failures that can be seen on various web sites if you don't have a
bunch of them in your collection of failed parts, that you should have
if you ride much.
Jobst Brandt
jobst....@stanfordalumni.org
Matt O'Toole
> Carl Fogel writes:
>> Has anyone shoved a crankset in a vise and tested it to see what
>> heaving on the pedal actually bends? I have one that I may be
>> willing to sacrifice and a fair-sized vise, but it seems better to
>> wait for suggestions about how to abuse it.
>
> That depends on whose crank (aka crank-set-arm) you choose. The old
> aluminum cranks with a vanity groove down the middle were
> substantially softer in torsion from standing on the pedal than in
> bending. Torsional stiffness being approximately that of a crank of
> the largest inscribed circle within the actual cross section. Even
> without this feature, most are probably still more torsionally soft
> from outboard pedal loads than in crank bending.
So why are more cranks not round in cross section, especially if weight is a
consideration?
Matt O.
Rick Onanian
Campagnolo) wrote:
>Yes and glass is a liquid and is 'flowing'.
This is actually a myth. The common evidence is old window panes
that are thicker at one end -- but they were manufactured that way.
Sometimes they are found with the thick side up.
The manufacturing process that produces the uneven panes is
described here:
http://math.ucr.edu/home/baez/physics/General/Glass/glass.html
There is no conclusive evidence, AFAIK, that glass is actually a
liquid.
>Ya know, a plastic spoiler on a compact car exerts downforce when ya drive it
>at 55-75 MPH, but is it significant or relevant?...no....
It is significant and relevant, just not good -- if they were less
dumb, they'd put it on the front of the car where it has less
traction; and if they were smart, they'd avoid the silly thing
entirely. If said wings have any effect on traction at all, it is to
worsen the understeer condition found on such cars.
Those are good for locking your bike rack to, until a potential
thief realizes that he can just break the spoiler off...
--
Rick Onanian
jobst....@stanfordalumni.org
Because they need to be oval in cross section for strength purposes.
As you see, Campagnolo and Shimano stopped using using vanity grooves
on their solid cranks long ago and now have cross sections that last
longer than those shown. The failure mode is not in the softer
direction (torsion) but in bending. See failures:
http://technology.open.ac.uk/materials/mem/mem-ccf4.html
http://pardo.net/pardo/bike/pic/fail/000.html
http://www.yellowjersey.org/photosfromthepast/DEADCRNX.JPG
Jobst Brandt
jobst....@stanfordalumni.org
Jon
All of us mechanical engineers know that any system of springs connected
in series (like say your foot, shoe, pedal, crank, BB, frame) is
derailled entirely by the weakest (most flexible) part as stiffness is
inversely additive. Do the math and it becomes very obvious that you're
being suckered by marketing hype yet again by Shimano.
It would be a nice change if Shimano engineers actually designed a
product that solved a problem instead of just forcing another exclusive
'standard' on the poor suckers with too much money and not enough brains.
Jon
A Muzi wrote:
> Tonight at a Shimano tech show, the speaker said that because frames
> have gotten so light and flexible, the stiffer new Shimano crank returns
> stiffness to previous levels.
>
> With a straight face.
>
> Me? I can't flex my TA Cyclotouriste single ring enough to notice.
A Muzi
>>>Has anyone shoved a crankset in a vise and tested it to see what
>>>heaving on the pedal actually bends? I have one that I may be
>>>willing to sacrifice and a fair-sized vise, but it seems better to
>>>wait for suggestions about how to abuse it.
> jobst....@stanfordalumni.org wrote:
>>That depends on whose crank (aka crank-set-arm) you choose. The old
>>aluminum cranks with a vanity groove down the middle were
>>substantially softer in torsion from standing on the pedal than in
>>bending. Torsional stiffness being approximately that of a crank of
>>the largest inscribed circle within the actual cross section. Even
>>without this feature, most are probably still more torsionally soft
>>from outboard pedal loads than in crank bending.
Matt O'Toole wrote:
> So why are more cranks not round in cross section, especially if weight is a
> consideration?
You might ask the same of frame tubes. Or rim sections. Or
handlebar sections.
Marketing.
Did you see the BUYcycling review of Waterford's R33? On
one page they praised components which they found 'rigid yet
flexible' several times and then gushed over the stiffness,
yet compliant comfort, of the wheels with tied spokes.(!)
These sort of writers determine a product's success now.
Merely being efficiently designed would not impress these
guys, I think.
You want components designed for maximum efficiency?
(material weight to torsional strength) Look to the round
section cranks in the twenty inch world.
--
Andrew Muzi
www.yellowjersey.org
Open every day since 1 April, 1971
Qui si parla Campagnolo
rebound, and raceways indent underneath them. Just like your tires on
the pavement, bearing balls must establish an elliptical "contact
Yikes, Chalo 'Godzilla' Colina
Jonesy
>
> There is no conclusive evidence, AFAIK, that glass is actually a
> liquid.
Like one of the forms of elemental sulfur, glass is what's known as an
"amorphous solid."
As a crank or BB material, I would guess that glass (or sulfur, for
that matter) would not be ideal. Just a guess.
--
Jonesy
Carl Fogel
Dear Andrew,
I vaguely thought that oval frame tubes were a
compromise between the all-directional strength
that they'd have if rounded and the improved
aerodynamics that they offered.
Modern racing pavement motorcycles also use
rectangular-section frame members (under far
greater stress). There, my impression is that
the direct aerodynamics don't matter much, since
the frame is under a fairing--it's a matter of
avoiding a bulging frame by squashing what would
be a four inch round tube down into a flattened
rectangle.
In any case, most of these parts face far greater
loads in one direction--they need not be equally
strong in every direction. The cranks, for example,
typically face most of their stress fore and aft,
not side to side laterally and not in barber-pole
torsion , so it makes sense to widen them in the
traditional manner.
A reasonably large circular-section crank would
break sooner than an oval-section crank with its
long dimension running fore and aft.
Think of what would happen if some crazed designer
rotated the traditional oval cross-section crank
ninety degrees, putting its long axis parallel to
the spindle.
Chalo probably couldn't bend such a crank inward,
but he might snap such cranks just by jumping up
and down on both pedals once.
Carl Fogel
onefred
news:bv5hvt$7he$1...@news-out.ftel.co.uk...
>
> Grand claims are made that the new XT Hollowtec cranks and ``Big Pipe''
> BB are very stiff. In practical terms, is this true, and does the
> stiffness of bottom brackets and cranks make a significant difference to
> the rider?
>
> ian
If you are to believe this article,
http://tinyurl.com/2qxlj
Then Shimano's Hollowtech II cranks are at least 42% stiffer than other
designs. Nonetheless, Octalink and ISIS are still very stiff interfaces.
Dave
jobst....@stanfordalumni.org
> I vaguely thought that oval frame tubes were a compromise between
> the all-directional strength that they'd have if rounded and the
> improved aerodynamics that they offered.
The whole non-round frame tube jive has been here before and plays on
the fear of wrinkling a downtube in the event of a collision with the
rear bumper of a car (the usual cause) or running into a cross drain
somewhere that just fits the front tire.
There is a far greater fear for those who have been there historically
and that is the failure of a fork crown or steer tube because the
frame is too rigid to take up any in-plane shock. In the days of
yore, fork failures occurred without warning. Such a failure usually
causes a head-first into the road, something for which a helmet does
little since the face is the first line of collision. I am
disappointed to see that no lesson was learned from the days of the
fork being the weakest link. Today it is once more in that spot with
some of the extremely rigid (high section) down tubes. This of course
is a road bicycle problem, the suspension fork ameliorating any
rigidity of the frame.
> Modern racing pavement motorcycles also use rectangular-section
> frame members (under far greater stress). There, my impression is
> that the direct aerodynamics don't matter much, since the frame is
> under a fairing--it's a matter of avoiding a bulging frame by
> squashing what would be a four inch round tube down into a flattened
> rectangle.
First, motorcycle frames are not "under far greater stress" or they
would fail regularly, considering the hours of use and miles they
travel. The bicycle is traditionally designed close to the limit of
failure in an effort to save weight where fatigue failures are common
on most components from ball bearings to frames. Rectangular tubes
have benefits for mechanical attachments and narrow, tall cross
sections that are typical of single track vehicles because they have
only light side loads. In contrast the bicycle has torsional loads
that round tubes take up best for single plane frames, unlike those of
motorcycles that have width.
> In any case, most of these parts face far greater loads in one
> direction--they need not be equally strong in every direction. The
> cranks, for example, typically face most of their stress fore and
> aft, not side to side laterally and not in barber-pole torsion , so
> it makes sense to widen them in the traditional manner.
I think you are guessing about this. If you look at crank failures
you'll see that your assessment of stresses are off the mark. Cranks
have arrived where they are by two main conditions, those of failure
resistance and flexibility. The second one has less significance
today with aluminum cranks but it was a problem with slender steel
ones.
> A reasonably large circular-section crank would break sooner than an
> oval-section crank with its long dimension running fore and aft.
That depends on what diameter you chose, not the shape in itself.
Besides, you would have to blend that cross section into a chainwheel
spider at some point and there is where you would start considering a
different cross section. Armchair design often arrives at bizarre
creations that fortunately don't make it to market in most cases.
Either that or they are short lived as the customer does the road
tests.
> Think of what would happen if some crazed designer rotated the
> traditional oval cross-section crank ninety degrees, putting its
> long axis parallel to the spindle.
And WHO is going to do this? There is a difference between a KBD
and useful hardware. I think you have got them too mixed up.
> Chalo probably couldn't bend such a crank inward, but he might snap
> such cranks just by jumping up and down on both pedals once.
You don't seem to understand fatigue failures. You can make a pretty
scrawny crank that will not break under high load but won't last long.
Consider that riding up the 30%+ grade of Filbert St in SF in a 47-20t
gear required more than 300lbf on the pedal just to stand still.
These cranks survived that routine for a couple of runs but cracked a
few hundred miles later from 8-10% grades on ordinary highways.
Jobst Brandt
jobst....@stanfordalumni.org
David Reuteler
: Consider that riding up the 30%+ grade of Filbert St in SF in a 47-20t
: gear required more than 300lbf on the pedal just to stand still.
good christ. that's a 63.5" gear. ouch, ouch, ouch, ouch, ouch.
: These cranks survived that routine for a couple of runs but cracked a
: few hundred miles later from 8-10% grades on ordinary highways.
did the cyclist make out any better?
--
david reuteler
reut...@visi.com
Richard Chan
>
> It would be a nice change if Shimano engineers actually designed a
> product that solved a problem instead of just forcing another exclusive
> 'standard' on the poor suckers with too much money and not enough brains.
Help me out here, I have a bunch of tapered cranks here that require
102mm axles. Which one of the Nouvo Record, Super Record, C-Record BBs
can I use?
How am I not forced to use a "new standard"?
Carl Fogel
[snip]
> First, motorcycle frames are not "under far greater stress" or they
> would fail regularly, considering the hours of use and miles they
> travel.
[snip]
Dear Jobst,
Possibly I failed to make myself clear?
I meant that motorcycle frames are under far greater
stress than bicycle frames.
Was this what you disagreed with?
Carl Fogel
Ryan Cousineau
carl...@comcast.net (Carl Fogel) wrote:
> A Muzi <a...@yellowjersey.org> wrote in message
> news:<101mf6k...@corp.supernews.com>...
> > >>Carl Fogel writes:
> > >>>Has anyone shoved a crankset in a vise and tested it to see what
> > >>>heaving on the pedal actually bends? I have one that I may be
> > >>>willing to sacrifice and a fair-sized vise, but it seems better to
> > >>>wait for suggestions about how to abuse it.
> >
> > jobst....@stanfordalumni.org wrote:
> > >>That depends on whose crank (aka crank-set-arm) you choose. The old
> > >>aluminum cranks with a vanity groove down the middle were
> > >>substantially softer in torsion from standing on the pedal than in
> > >>bending. Torsional stiffness being approximately that of a crank of
> > >>the largest inscribed circle within the actual cross section. Even
> > >>without this feature, most are probably still more torsionally soft
> > >>from outboard pedal loads than in crank bending.
> >
> > Matt O'Toole wrote:
> > > So why are more cranks not round in cross section, especially if weight
> > > is a
> > > consideration?
> > You want components designed for maximum efficiency?
> > (material weight to torsional strength) Look to the round
> > section cranks in the twenty inch world.
>
> Dear Andrew,
>
> I vaguely thought that oval frame tubes were a
> compromise between the all-directional strength
> that they'd have if rounded and the improved
> aerodynamics that they offered.
>
> Modern racing pavement motorcycles also use
> rectangular-section frame members (under far
> greater stress). There, my impression is that
> the direct aerodynamics don't matter much, since
> the frame is under a fairing--it's a matter of
> avoiding a bulging frame by squashing what would
> be a four inch round tube down into a flattened
> rectangle.
Modern racing pavement motorcycles usually use _two_ rectangular-section
frame members, down either side of the engine, and the engine itself is
used as a stressed member. In other words, motorcycle frames have plenty
of other torsional stiffness advantages.
Frame width is important, because it has major effects on the comfort
and fit of the motorcycle, the turning radius, and a considerable effect
on the aerodynamics (frontal area considerations, mostly).
Be very wary of trying to translate racing motorcycle design (where the
heaviest load is the engine and transmission) to bicycles.
--
Ryan Cousineau, rcou...@sfu.ca http://www.sfu.ca/~rcousine
President, Fabrizio Mazzoleni Fan Club
dianne_1234
wrote:
I think what you might mean is they are under far greater *forces*.
Stress has an engineering definition that is very specific:
http://www.vishay.com/brands/measurements_group/guide/glossary/stress.htm
"stress - The force acting across a unit area in a solid material in
resisting the separation, compacting, or sliding that tends to be
induced by application of forces."
http://www.m-w.com/cgi-bin/dictionary
"force - 4 a : an agency or influence that if applied to a free body
results chiefly in an acceleration of the body and sometimes in
elastic deformation and other effects "
Qui si parla Campagnolo
102mm axles. Which one of the Nouvo Record, Super Record, C-Record BBs
can I use?
How am I not forced to use a "new standard"? >><BR><BR>
102mm BBs are Campagnolo Record and Chorus produced starting in 1995 until
today. You cannot use the NR for anything but NR, Super Record or Gran Sport.
You can use the C-Record for any Campagnolo crank below the Chorus level
produced since 1995, any crank from 1984 until 1995, then all below Chorus...
Tom Sherman
...
> Yes and glass is a liquid and is 'flowing'....
Glass is a non-crystalline material. There are four commonly used
transition points:
1. "Working point" - above this temperature the glass may be drawn or
pressed.
2. "Softening point" - above this temperature the glass will deform
under its own weight.
3. "Anneal point" - above this temperature the glass will creep and
stresses can be relieved through the conversion of elastic deformation
to plastic deformation.
4. "Strain point" - below this temperature the glass behaves
elastically, and permanent plastic deformation will result in fracture.
Above the anneal point, glass behaves as a visco-elastic solid, and
below the strain point, it behaves as an elastic solid.
Tom Sherman - Quad Cities
Tom Sherman
> ...
> Consider that riding up the 30%+ grade of Filbert St in SF in a 47-20t
> gear required more than 300lbf on the pedal just to stand still....
Using such gearing for a very steep grade such as Filbert Street is just
nuts (pun intended). ;)
Carl Fogel
Dear Jobst,
I'm still puzzled.
Can you explain what differences you do or do not
see in the stresses or forces on bicycle and motorcycle
frames?
Carl Fogel
Carl Fogel
Dear Ryan,
I'm not sure that we disagree.
But it may be worth pointing out that the
disappearance of round frame tubing is not
confined to pavement motorcycles.
Almost all off-road motorcycles now use
either oval or rectangular frame members,
from motocross to trials bikes. Only the
telescopic front fork-tubes seem to have
remained round.
While using the engine as a frame member
adds enormous strength, it does not explain
the almost universal use of rectangular
swing-arm sections.
Strength and weight are just as important
to motorcycles as bicycles. That is, if
round frame tubes offered the same strength
(in the directions needed) for the same weight,
modern trials bikes would be using round frame
tubes.
Whether the roughly 400-pound load consists
of an ordinary rider and an engine, or of Chalo
and a bicycle, the motorcycle frame will face
greater forces because even a gentle engine
puts out more than twenty times the horsepower,
which means that the rider and two-wheeled vehicle
accelerate much faster, reach greater speeds,
must brake from these greater speeds, hit the
bumps harder, and jump and land far more often
from far greater heights.
Since claims about jumping may rouse the ire of
huckers, I should explain that most bicycle jumping
involves a steep downhill approach to a ramp--a
fairly infrequent situation. Bicycles climbing
steadily up a mountain simply lack the power to
go fast enough to fling themselves into the air
repeatedly like ordinary off-road motorcycles.
I expect that the matter of shape and strength
is more complicated suggested by either idealized
omni-directional theory or appeals to practical
application--which is what makes it interesting.
Carl Fogel
Peter Serjeant
> For a rider to say he feels arms bend or spindles twist is just bugleoil.
>
> Peter Chisholm
Sorry,but speaking as another Clydsdale, I have recently upgraded my
bike from the original(87)Shimano 105 group to Ultegra with hollow
cranks/spindle and was pleasantly surprised to find the difference was
very evident. Same pedals BTW.
Yes I DO use BugleOil, but only to lube crank tapers.
Ryan Cousineau
carl...@comcast.net (Carl Fogel) wrote:
The swing-arm needs to be very strong vertically (so that a bump at the
rear wheel doesn't break it in half), but torsional strength is provided
by the rear axle, and often with a brace connecting the two swingarm
sides. Combined with the need for the tire to fit between the swingarms,
this encourages a design which is skinny on the sides, because it gets
torsional strength from elsewhere.
A bicycle's rear triangle, by contrast, consists of the thinnest tubing
on the bike. Jobst has explained in the FAQ that this is because the
rear triangle (double-triangle, really) and the rear axle form a
tetrahedron, which is a really good shape.
Rear triangle tubes are tiny and round because they are plenty strong
when arranged the way they are. You couldn't make them much smaller or
they'd start trying to flex laterally, or beercan, or otherwise deform.
As for the use of oddly-shaped frame members, note once again that the
frame design of a motorcycle has very different needs than that of a
bicycle. Motorcycles carry their engine in the middle of the frame, and
most performance bikes use that engine as a stressed member. BMW takes
this to the extreme with their boxer motorcycles, in which there are
just a few subframes that bolt onto the engine, which is the only
significant "frame member" between the front suspension and the rear
suspension.
There are various reasons why most motorcycles have double top tubes and
downtubes, but it amounts to the fact that motorcycles have to be about
30-90 cm wide or so to accomodate the engine, so it's better to use that
width to run two tubes, at a notable increase in torsional and lateral
stiffness.
I know of one recent bike that gets by with a single top tube, and it
uses a steel tube about 3" in diameter. Which is one way of doing it.
> Whether the roughly 400-pound load consists
> of an ordinary rider and an engine, or of Chalo
> and a bicycle, the motorcycle frame will face
> greater forces because even a gentle engine
> puts out more than twenty times the horsepower,
> which means that the rider and two-wheeled vehicle
> accelerate much faster, reach greater speeds,
> must brake from these greater speeds, hit the
> bumps harder, and jump and land far more often
> from far greater heights.
Yes.
> I expect that the matter of shape and strength
> is more complicated suggested by either idealized
> omni-directional theory or appeals to practical
> application--which is what makes it interesting.
--
dianne_1234
wrote:
Dear Carl,
Your curiosity has brought you to a point where further understanding
requires mutually agreed upon and carefully adhered to definitions so
that everyone understands one another. Stress has been very narrowly
defined for just this purpose.
Read below, and you should see that you have (inadvertently) claimed
motorcycles frames are under higher stress, when I'm fairly certain
all you meant was that they are under higher loads. Just guessing
here, but I think Jobst is probably assuming that the MC frame tubes
probably have, e.g., greater cross sectional area -- to reduce the
stress.
No doubt the motorcycle designer did this on purpose, precisely in
order to reduce the stress. That was his intent: he doesn't want the
MC to break as frequently as, say, some light weight bike frames,
whose tube walls are so thin that the stress (even from the smaller
loads a bicycle sees) is still higher than a MC designer wants in his
own design.
From
http://www.sciencedaily.com/encyclopedia/Stress_(physics)
=== Begin quoted text =================
Stress (physics)
From Wikipedia, the free encyclopedia.
In physics, the stress at a point in a material is the applied force
per unit area. The stress unit is the Pascal (symbol Pa). To be exact,
the stress at a point may be determined by taking the limit of the
load being carried by a particular cross section, divided by that
cross section, as the area of the cross section approaches zero. In
general the stress may vary from point to point, but for simple cases,
such as circular cylinders with pure axial loading, the stress is
constant and equal to the cross-sectional area divided by the applied
load.
Stress is described by a symmetric tensor.
For instance, if we have a steel bolt with a diameter of 5 mm, it has
a cross-sectional area of 2*10-5m2. Suppose that the load is 50 kN,
the stress (force distributed across the cross-section) is about 2.5
MPa.
That means each m2 of bolt would support 2.5 MN of the total load.
In another bolt with half the diameter, and hence a quarter the
cross-sectional area, carrying the same 50 kN load, the stress will be
quadrupled (10 MPa).
The ultimate tensile strength of a material is the value of the
tension stress causing the material's fracture. The compression
strength is analogous for compression strain. The yield strength is
the value of stress causing plastic deformation. These values are
determined experimentally using the measurement procedure known as the
tensile test.
============= End of Quoted Text ==============
jobst....@stanfordalumni.org
>> Consider that riding up the 30%+ grade of Filbert St in SF in a
>> 47-20t gear required more than 300lbf on the pedal just to stand
>> still.
> good christ. that's a 63.5" gear. ouch, ouch, ouch, ouch, ouch.
>> These cranks survived that routine for a couple of runs but cracked
>> a few hundred miles later from 8-10% grades on ordinary highways.
> did the cyclist make out any better?
I had so many crank failures that I inspected for cracks before every
weekend ride and in that mode caught most of them before separation.
Since then I modified my cranks to prevent the common pedal eye
failure that occurs at the bottom of the stroke (where nearly all my
dozens of cranks failed).
The modification has made cranks last tenfold or more. I countersink
the crank thread and essentially make the pedal shaft have a conical
face so that it rides like a lug nut of a car in a non-squirming fit.
That this is a moving fit is borne out by the need for left hand
threads on left pedals and the erosion of crank faces by pedals, the
place where cracks are initiated.
Jobst Brandt
jobst....@stanfordalumni.org
jobst....@stanfordalumni.org
>> First, motorcycle frames are not "under far greater stress" or they
>> would fail regularly, considering the hours of use and miles they
>> travel.
> Possibly I failed to make myself clear?
> I meant that motorcycle frames are under far greater stress than
> bicycle frames.
Possibly I failed to make myself clear. I meant what I said. Bicycle
frames regularly fail from fatigue while a motorcycle failure is
practically unheard of. Therefore, motorcycle frame are not stressed
as highly as bicycle frames.
> Was this what you disagreed with?
Among other points.
Jobst Brandt
jobst....@stanfordalumni.org
jobst....@stanfordalumni.org
>> Consider that riding up the 30%+ grade of Filbert St in SF in a
>> 47-20t gear required more than 300lbf on the pedal just to stand
>> still....
> Using such gearing for a very steep grade such as Filbert Street is
> just nuts (pun intended). ;)
It was an interesting outing with a bunch of bikies who were young and
fast racers. Few were able to ride up that street (1/2 block) between
Leavenworth and Hyde. We did that a couple of times for a Sunday ride
up El Camino Real from Palo Alto and back HWY 1 on the coast. No, our
tubulars do not slip. We also rode up the bricks of the (one-way down)
Lombard curlicue.
Jobst Brandt
jobst....@stanfordalumni.org
jobst....@stanfordalumni.org
>>> I vaguely thought that oval frame tubes were a compromise between
>>> the all-directional strength that they'd have if rounded and the
>>> improved aerodynamics that they offered.
>>> Modern racing pavement motorcycles also use rectangular-section
>>> frame members (under far greater stress). There, my impression is
>>> that the direct aerodynamics don't matter much, since the frame is
>>> under a fairing--it's a matter of avoiding a bulging frame by
>>> squashing what would be a four inch round tube down into a
>>> flattened rectangle.
>> Modern racing pavement motorcycles usually use _two_
>> rectangular-section frame members, down either side of the engine,
>> and the engine itself is used as a stressed member. In other words,
>> motorcycle frames have plenty of other torsional stiffness
>> advantages.
>> Frame width is important, because it has major effects on the
>> comfort and fit of the motorcycle, the turning radius, and a
>> considerable effect on the aerodynamics (frontal area
>> considerations, mostly).
>> Be very wary of trying to translate racing motorcycle design (where
>> the heaviest load is the engine and transmission) to bicycles.
> I'm not sure that we disagree.
I am.
> But it may be worth pointing out that the disappearance of round
> frame tubing is not confined to pavement motorcycles.
> Almost all off-road motorcycles now use either oval or rectangular
> frame members, from moto-cross to trials bikes. Only the telescopic
> front fork-tubes seem to have remained round.
I think there is a misunderstanding of stress, load, and torsion and
how these arise in bicycle and motorcycle frames. The motorcycle
frame has practically no torsional loads, ones that occur greatly in
the main tubes of bicycle frames that are for that reason round.
Stress, as has been mentioned is in psi not lbs and even then the
bicycle, under thrusting loads of a ~200lb rider without suspension,
causes grater forces than a motorcycle, considering the less than
0.040" wall thicknesses on bicycle tubes and their size.
> While using the engine as a frame member adds enormous strength, it
> does not explain the almost universal use of rectangular swing-arm
> sections.
These elements are stressed mainly in tension and compression, like
those of the bicycle's "rear triangle". To make up for that, the M/C
frame has larger tubes with greater wall thickness.
> Strength and weight are just as important to motorcycles as
> bicycles. That is, if round frame tubes offered the same strength
> (in the directions needed) for the same weight, modern trials bikes
> would be using round frame tubes.
Not so. On a motorcycle reliability is far more important considering
that at 100mph a greater disaster with far greater public scrutiny can
occur. We don't hear of such events often on bicycles because they
are not as exposed to highway traffic at such speeds. However, if you
interview an experienced bicycle mechanic, you will discover many
parts and frame failures.
> Whether the roughly 400-pound load consists of an ordinary rider and
> an engine, or of Chalo and a bicycle, the motorcycle frame will face
> greater forces because even a gentle engine puts out more than
> twenty times the horsepower, which means that the rider and
> two-wheeled vehicle accelerate much faster, reach greater speeds,
> must brake from these greater speeds, hit the bumps harder, and jump
> and land far more often from far greater heights.
Quit picking on the physique of some riders. If you pick up a bicycle
frame and an M/C frame you can see that there is a great difference in
weight.
> Since claims about jumping may rouse the ire of huckers, I should
> explain that most bicycle jumping involves a steep downhill approach
> to a ramp--a fairly infrequent situation. Bicycles climbing steadily
> up a mountain simply lack the power to go fast enough to fling
> themselves into the air repeatedly like ordinary off-road
> motorcycles.
That has little to do with frame design and failures, since these are
forced ruptures if any, not classic failures of metal fatigue, the
ones for which most designs are critical.
> I expect that the matter of shape and strength is more complicated
> suggested by either idealized omni-directional theory or appeals to
> practical application--which is what makes it interesting.
Don't try to make this more complicated than it is. That's called
obfuscation.
Jobst Brandt
jobst....@stanfordalumni.org
A Muzi
>>>>>Has anyone shoved a crankset in a vise and tested it to see what
>>>>>heaving on the pedal actually bends?
>> jobst....@stanfordalumni.org wrote:
>>>>That depends on whose crank (aka crank-set-arm) you choose. The old
>>>>aluminum cranks with a vanity groove down the middle were
>>>>substantially softer in torsion from standing on the pedal than in
>>>>bending. Torsional stiffness being approximately that of a crank of
>>>>the largest inscribed circle within the actual cross section. Even
>>>>without this feature, most are probably still more torsionally soft
>>>>from outboard pedal loads than in crank bending.
>>Matt O'Toole wrote:
>>>So why are more cranks not round in cross section, especially if weight is a
>>>consideration?
> A Muzi <a...@yellowjersey.org> wrote in message news:<101mf6k...@corp.supernews.com>...
>>You might ask the same of frame tubes. Or rim sections. Or
>>handlebar sections.
>>Marketing.
-snip-
Carl Fogel wrote:
> I vaguely thought that oval frame tubes were a
> compromise between the all-directional strength
> that they'd have if rounded and the improved
> aerodynamics that they offered.
snip-
> A reasonably large circular-section crank would
> break sooner than an oval-section crank with its
> long dimension running fore and aft.
> -snip-
Jobst succinctly pointed out it's the largest circle which
will fit in the section determines the torsional resistance
Changing a round tube for a larger oval is one thing.
Ovalling a tube of the same circumference is quite another.
And for efficiency ( least amount of material for a certain
level of torsional resistance) round tubes are clearly
better.
And cranks torque in use. There's not just load in one
plane. Really. An oval crank is a less efficient use of
material.
Carl Fogel
Dear Ryan,
When I search the photo database at that nice site
that you mentioned a few months ago:
http://www.pinkbike.com/modules/photo/
for "swingarm," it seems as if the great majority
of bicycle suspension swingarms for hucking resemble
modern motorcycle swingarms. That is, rectangular,
box-section swing arms with the kind of bracing that
you mention just foward of the tire.
I'd ask all sorts of questions, but I don't want to
give the impression that I'm arguing or playing gotcha.
I'm just interested in whether my impression that the
rectangular swingarm is preferred for this kind of
riding is correct, and if so, what your thoughts are
on why it's preferred.
One question that I need to ask, again without putting
it as an argument, is how the bracing just forward
of the tire between the monoshock swing arms differs
from ordinary rigid chainstays meeting at the bottom
bracket--the welding seems to occur at about the same
place.
Thanks,
Carl Fogel
A Muzi
>>It would be a nice change if Shimano engineers actually designed a
>>product that solved a problem instead of just forcing another exclusive
>>'standard' on the poor suckers with too much money and not enough brains.
Richard Chan wrote:
> Help me out here, I have a bunch of tapered cranks here that require
> 102mm axles. Which one of the Nouvo Record, Super Record, C-Record BBs
> can I use?
> How am I not forced to use a "new standard"?
Somewhere in snipping I lost your point. (?)
Campagnolo over the years has shortened spindles just as
everyone else has. The appropriate year of crank matches
that year of spindle.
The asymmetric 1046a Nuovo Record BB assembly has an
identical spindle with the Shimano BB7400 for the first New
Dura Ace. That pattern is correct to another dozen cranks
of top quality in that era. Or an original Phil Wood #2 BB.
That was a de facto standard among professional quality
bikes for over twenty years, across nearly every major
brand. (OK, not the two big French houses but they were
always intractable on purpose)
Sadly, nowadays everyone wants to reinvent the wheel it
seems. We'll never see such broad interchangeability again.
(for your cranks which need 102, use the current Record or
Chorus)
David Reuteler
:> did the cyclist make out any better?
:
: I had so many crank failures that I inspected for cracks before every
: weekend ride and in that mode caught most of them before separation.
: Since then I modified my cranks to prevent the common pedal eye
: failure that occurs at the bottom of the stroke (where nearly all my
: dozens of cranks failed).
had to have hurt. a lot. i climbed that hill myself in what i'd guess was
a gear in the low 30s. being a native midwesterner & given the chance i'm
kinda into that particuliar brand of self abuse. great fun.
: The modification has made cranks last tenfold or more. I countersink
: the crank thread and essentially make the pedal shaft have a conical
: face so that it rides like a lug nut of a car in a non-squirming fit.
: That this is a moving fit is borne out by the need for left hand
: threads on left pedals and the erosion of crank faces by pedals, the
: place where cracks are initiated.
i've seen pictures? or heard you talk on this before. you should, uh, get
a patent on this and donate it to shimano so the rest of us can benefit.
--
david reuteler
reut...@visi.com
jobst....@stanfordalumni.org
>>> did the cyclist make out any better?
>> I had so many crank failures that I inspected for cracks before
>> every weekend ride and in that mode caught most of them before
>> separation. Since then I modified my cranks to prevent the common
>> pedal eye failure that occurs at the bottom of the stroke (where
>> nearly all my dozens of cranks failed).
> Actually I was referring to your body. Climbing that grade in a 63"
> gear had to have hurt. A lot. I climbed that hill myself in what
> I'd guess was a gear in the low 30s. Being a native midwesterner &
> given the chance I'm kinda into that particuliar brand of self
> abuse. Great fun.
Well when you are young and strong, it isn't hard or abusive. I
didn't do that to suffer but rather to show the other guys that their
excuses of insufficient gears wouldn't hold water. They were after
all bicycle racers who could beat me most any time. I was an oldster
(retired racer) at that time already.
>> The modification has made cranks last tenfold or more. I
>> countersink the crank thread and essentially make the pedal shaft
>> have a conical face so that it rides like a lug nut of a car in a
>> non-squirming fit. That this is a moving fit is borne out by the
>> need for left hand threads on left pedals and the erosion of crank
>> faces by pedals, the place where cracks are initiated.
> I've seen pictures? or heard you talk on this before. You should,
> uh, get a patent on this and donate it to Shimano so the rest of us
> can benefit.
It isn't patentable and as I have mentioned here before, neither
Shimano or Campagnolo are interested. Truvative gave me the usual
routing of "who are you" rather than looking at the matter on its
merit rather than who said it. Only the Japanese higher ups that
usually attend InterBike were approachable (Shimano employees are
forbidden to talk technically with non Shimano people) and I get the
impression these people understood the problem and how this solves it.
I don't know whether they will ever address it though.
Jobst Brandt
jobst....@stanfordalumni.org
Ryan Cousineau
> Ryan Cousineau <rcou...@sfu.ca> wrote in message
> news:<rcousine-18D893...@morgoth.sfu.ca>...
> > In article <8bbde8fc.04020...@posting.google.com>,
> > carl...@comcast.net (Carl Fogel) wrote:
> > > Dear Ryan,
> > The swing-arm needs to be very strong vertically (so that a bump at the
> > rear wheel doesn't break it in half), but torsional strength is provided
> > by the rear axle, and often with a brace connecting the two swingarm
> > sides. Combined with the need for the tire to fit between the swingarms,
> > this encourages a design which is skinny on the sides, because it gets
> > torsional strength from elsewhere.
> > > I expect that the matter of shape and strength
> > > is more complicated suggested by either idealized
> > > omni-directional theory or appeals to practical
> > > application--which is what makes it interesting.
>
> Dear Ryan,
>
> When I search the photo database at that nice site
> that you mentioned a few months ago:
>
> http://www.pinkbike.com/modules/photo/
>
> for "swingarm," it seems as if the great majority
> of bicycle suspension swingarms for hucking resemble
> modern motorcycle swingarms. That is, rectangular,
> box-section swing arms with the kind of bracing that
> you mention just foward of the tire.
>
> I'd ask all sorts of questions, but I don't want to
> give the impression that I'm arguing or playing gotcha.
> I'm just interested in whether my impression that the
> rectangular swingarm is preferred for this kind of
> riding is correct, and if so, what your thoughts are
> on why it's preferred.
At some point, what you'll "gotcha" is that I'm not an engineer. My
understanding of the physics is a bit colloquial, and I often
"reverse-engineer" these things, figuring out what the smarter engineers
did and what forces they must have found in testing to design a
structure that way. It's not always reliable.
Here's a nice pic of a fairly typical freeride bike frame:
http://www.mountaincycle.com/images/large_image_popups/sin/sin_large_popu
p.jpg
What you may notice is that this frame looks a lot like an off-road
motorcycle frame. The swingarm is big and beefy, and it pretty much is
at the limit of how wide it can be. Any wider, and it would start
hitting stuff (chains, cranks, your foot...). The only way to make it
stiffer is to go taller, go around (via a brace), or as some motorcycle
frame members do, use internal bracing.
The basic issue is really heavy landing loads (somewhat mitigated by
suspension), and you may also notice the lack of any seat stays. This
means the frame has to be very sturdy.
> One question that I need to ask, again without putting
> it as an argument, is how the bracing just forward
> of the tire between the monoshock swing arms differs
> from ordinary rigid chainstays meeting at the bottom
> bracket--the welding seems to occur at about the same
> place.
>
> Thanks,
Firstly, really stiff motorcycle setups actually use a brace that is
somewhat removed from the swingarm, like this:
http://www.yamaha-motor.com/products/UnitImage.asp?lid=2&lc=mcy&cid=5&mid
=6&iid=4954
Note the underslung swingarm brace. Sometimes they go over the top
instead, or even have both.
Carl Fogel
Dear Jobst,
I think that now I understand what you're saying,
but correct me if I'm wrong: given their relative
designs, bicycle frames are closer to breaking under
their normal use than motorcycle frames, the latter
being more robust.
But now I'm curious about the idea that bicycles
frames fail regularly while a motorcycle failure
is practically unheard of.
In this survey, the first that I found through
Google, 6% of Ducati street motorcycle owners
reported that their frames had broken:
http://www.duc.org/900ssduc_owner.htm
It looks like a 1998 survey with 84 responses for
an 8-year-old model, with most of the breaks afflicting
owners who said that they rode on tracks.
In my experience, motorcycle frames break without crashing.
My faithful Honda trials machine, for example, has a neat
weld where a friend noticed while loading the bikes on the
trailer that a round tube (the left fork of the backbone
descending to the footpeg) had cracked completely through.
What kind of non-crash frame-break rate would you say
that bicycles suffer? A quick search for "frame broke"
in rec.bicycles.tech turns up about 78 posts since 1992,
with apparently duplicated posts omitted.
I should add that I'm not arguing relative or absolute
rates. I'm just curious how often you (and others) with
more bicycle experience expect frames to break in normal
use.
Thanks,
Carl Fogel
Qui si parla Campagnolo
bike from the original(87)Shimano 105 group to Ultegra with hollow
cranks/spindle and was pleasantly surprised to find the difference was
No doubt it was but you may want to look at bearing design and placement as
well as moment arm of the crank spindle/crank and bearings...as for actually
flexing and bending and twisting the 1987 105 crankarms and steel
spindle...sorry...
jim beam
bike passion, but you can't say this stuff because it's not correct.
if you want colloquial evidence, check out the broken solid spindle pics
on the net. by /definition/, for those things to fatigue, they /have/
to flex.
if you want to go down the engineering route, look up the math for tube
torsion on the net & compare hollow spindles to solid. for the solids,
set the internal radius to zero. you could do the same for crank arms,
but they're not a round section so it's more complicated.
again, i respect your advice and encyclopedic campy experience, so i'm
only seeking technical accuracy here.
jb
Qui si parla Campagnolo wrote:
Richard Chan
> > Jon <jek...@NOSPAMucalgary.ca> wrote in message news:<bveljl$dvu$1...@news.ucalgary.ca>...
> >>It would be a nice change if Shimano engineers actually designed a
> >>product that solved a problem instead of just forcing another exclusive
> >>'standard' on the poor suckers with too much money and not enough brains.
>
> Richard Chan wrote:
> > Help me out here, I have a bunch of tapered cranks here that require
> > 102mm axles. Which one of the Nouvo Record, Super Record, C-Record BBs
> > can I use?
> > How am I not forced to use a "new standard"?
>
> Somewhere in snipping I lost your point. (?)
My point is Jon is accusing Shimano of designing/manufacturing new
standards and forcing consumers to adopt them. In this particular
example, i.e., cranks, that just because it is square tapered, you
still have to buy new matching BB for modern Campy cranks (meaning you
can't use older units). 102 mm is not 11x mm. It is a new standard. It
is a non issue for me if he states that ALL manufacturers do this.
jim beam
vehicles fail regularly, unibody ones less so. cranks, con rods,
suspension arms, lug nuts - they all fail.
i think what you're getting at is what jobst is saying: the closer you
operate a component to yield, the sooner it's going to fail. it doesn't
really matter if it's a bike, motorbike, landing gear on an f16 or a
turbine shaft on an aircraft carrier.
fatigue management all depends on expectations. formula one cars for
example are run at only 80% reliability - anything more in not
considered to be pushing the competitive envelope enough. other
applications like nuclear, military or aerospace aim for near 100%.
testing is rigorous and expensive and users are prepared to trade
performance for reliability. users of bikes, whether powered or not,
are faced with the dilemma of wanting the cutting edge on perfomance
without the testing costs. not a great combo, but you pays your money &
takes your choice.
jb
dvt
>>> ...a plastic spoiler on a compact car...
> Those are good for locking your bike rack to, until a potential
> thief realizes that he can just break the spoiler off...
... at which point he/she is no longer a *potential* thief...
Dave
dvt at psu dot edu
bfd
Interestingly, I'm using an *old* (like from mid80s) 115mm Edco
Competition sealed BB with my Campy Record 9 crank. It works well on
my Carbonframes (now Calfee) frame which required a longer spindle due
to the wide chainstay.
Personally, I like Shimano stuff. I think the "criticism" here is that
Shimano came out with this new BB design, Octalink, in around 1996 or
1997. Now in 2004, they seem to be heading away from that design with
the new DA 10 crank/BB combo and the rumoured trickle down effect
expected over the next several years.
Will octalink die? I doubt it. Shimano still offers square taper bb -
UN5x/7x. You may not get the assortment of octalink bb that was once
offered, but Shimano probably will offer some sort of 105/LX version
of Octalink for a long time....
Peter Serjeant
> pser-<< Sorry,but speaking as another Clydsdale, I have recently upgraded my
> bike from the original(87)Shimano 105 group to Ultegra with hollow
> cranks/spindle and was pleasantly surprised to find the difference was
> very evident. >><BR><BR>
>
> No doubt it was but you may want to look at bearing design and placement as
> well as moment arm of the crank spindle/crank and bearings...as for actually
> flexing and bending and twisting the 1987 105 crankarms and steel
> spindle...sorry...
Of course over those umpteen years I have been through a few BB's. I
would assume that the cartridge version I was using of late had very
similar bearing placement and bending moments to the new Ultegra one.
Pete
Qui si parla Campagnolo
still have to buy new matching BB for modern Campy cranks (meaning you
Modern, I guess that means for tha lst 9 years. And it is for Record and
Chorus, not all other Campagnolo cranks since 1994...
Qui si parla Campagnolo
on the net. by /definition/, for those things to fatigue, they /have/
to flex. >><BR><BR>
I never said they don't but I don't believe a cyclist can 'feel' the teeny
differences from one crankarm or BB spindle to another. For crank makers along
with wheel makers to advertise stiffness claims or weight location claims like
it makes a whole bunch of difference is gar-bage'.
The stiffness differences for these things are genuine and measureable, just
like the aero differneces of a wheel or frameset but does it matter to a
cyclist and their performance? Probably not.
I guess my beef is with the stuff coming out of advertising departments rather
than R/D, racing departments.
Todd Kuzma
> I guess my beef is with the stuff coming out of advertising departments rather
> than R/D, racing departments.
Well, Peter, you can believe that our new Heron
Super-Stiff(tm) Housing Stops* are tested in the lab to be
20% stiffer than standard housing stops. Why not braze on a
set today?
*Patent pending. Void where prohibited. UCI approval pending.
Todd Kuzma
Heron Bicycles
LaSalle, Il 815-223-1776
http://www.heronbicycles.com
Peter Storey
How do these enhance power transfer? I think you want to focus on
stiffening the cable -- all the ones I've seen are really flexible.
Personally, I'd like to see someone develop a handlebar tape that is
stiff when you pull up on the bars, but flexible when you push down.
Peter Storey
Todd Kuzma
> Todd Kuzma <tul...@TheRamp.net> wrote in message news:<401FD255...@TheRamp.net>...
>
>>Qui si parla Campagnolo wrote:
>>
>>
>>>I guess my beef is with the stuff coming out of advertising departments rather
>>>than R/D, racing departments.
>>
>>Well, Peter, you can believe that our new Heron
>>Super-Stiff(tm) Housing Stops* are tested in the lab to be
>>20% stiffer than standard housing stops. Why not braze on a
>>set today?
>>
>>*Patent pending. Void where prohibited. UCI approval pending.
>>
>
> stiffening the cable -- all the ones I've seen are really flexible.
We are hard at work on the cable flex issue. Here is a
picture of a prototype of our new "rod-operated" braking
system:
<http://www.retroraleighs.com/catalogs/1951-england/pages/20-dawn.html>
Mark Hickey
>We are hard at work on the cable flex issue. Here is a
>picture of a prototype of our new "rod-operated" braking
>system:
>
><http://www.retroraleighs.com/catalogs/1951-england/pages/20-dawn.html>
Great stuff... I had a bike in China with those "brakes". They
actually worked surprisingly well, but I couldn't help wonder about
what would happen if certain of those linkages were to break suddenly.
Not to mention it's kinda tough to get brake pads that pull straight
up against the rim to work very well with deep dish rims! That kept
me from releasing my "aero pigeon" model.
Mark Hickey
Habanero Cycles
http://www.habcycles.com
Home of the $695 ti frame
jlscott3
modeling package from www.kisssoft.ch that turned out to be more and
less than what I expected. It allows one to model shafts with variable
diameters and bearing points, but the demo version doesn't allow choice
of materials or saving files. It also managed to blow up on me often
enough that I decided to give it up and dig up the fundamental equation.
The equation for torsional stiffness of a cylindrical shaft is
given as: J*G/L
Where J = area moment of inertia for the shaft = pi * (r1^4 - r2^4)/2,
where r1 = outer radius r2 = inner radius G = shear modulus of the
material (80KN/mm^2 for steel) L = length of the shaft
(Note, this equation appears in Jobst's book, p.137 in my 7th printing)
I just so happen to have a Suntour Superbe Pro BB spindle sitting on my
desk, with an outer radius of ~8mm and inner radius of ~3.5mm, and
length of ~111mm. Plugging those numbers into the equation yields a
stiffness of 77.97 Nm/deg, or 7.956Kg-m/deg here on earth. That's
equivalent to 56.5 ft-lb/deg.
So you'd have to apply 46.8 Kg of force (458.6 N) to a 170mm crank to
twist this BB spindle 1 degree - actually a bit more since the effective
length of the BB would be shorter than the overall length. Seems to me
that a person of 75Kg or more might be able to perceive the twist
induced by standing on the pedals.
I don't have an Octalink or Hollowtech BB around to measure, so I can't
compare stiffness. If somebody's got relevant measurements you can post
them here or email them to jls.u...@doppke.com.
IANAME so those of you who actually know what you're talking about
should feel free to correct my work.
JLS
--
dvt
> The equation for torsional stiffness of a cylindrical shaft is
> given as: J*G/L
> I just so happen to have a Suntour Superbe Pro BB spindle sitting on my
> desk, with an outer radius of ~8mm and inner radius of ~3.5mm, and
> length of ~111mm. Plugging those numbers into the equation yields a
> stiffness of 77.97 Nm/deg, or 7.956Kg-m/deg here on earth. That's
> equivalent to 56.5 ft-lb/deg.
I didn't check your numbers, so I'll assume your math is correct. The ID
is only important if it is hollow all the way through. Is it?
> So you'd have to apply 46.8 Kg of force (458.6 N) to a 170mm crank to
> twist this BB spindle 1 degree - actually a bit more since the effective
> length of the BB would be shorter than the overall length. Seems to me
> that a person of 75Kg or more might be able to perceive the twist
> induced by standing on the pedals.
If you shorten the length a bit use a solid shaft, you might get
slightly higher numbers. But the basic conclusion would be the same: the
average rider should be able to get a degree or two of torsion out of
the BB. One degree of torsion gives 1/360*2*pi*170mm = 3 mm of arc
length. Note that the left side will demonstrate this flex, but loads on
the right side don't apply any torque to the BB spindle.
Can the rider feel a "squish" of 3 mm on one side? I would think it
wouldn't be terribly obvious, but it might be noticeable if everything
else (cranks, pedals) were very stiff.
jlscott3
> I didn't check your numbers, so I'll assume your math is correct.
Thanks :-) I did double-check my calc against the example in /The
Bicycle Wheel/, which I thought was wrong until I realized that Jobst
had factored the gravitational constant out of the shear modulus.
> The ID is only important if it is hollow all the way through. Is it?
Yes, I can see daylight through the bolt hole.
> One degree of torsion gives 1/360*2*pi*170mm = 3 mm of arc length.
> Note that the left side will demonstrate this flex, but loads on the
> right side don't apply any torque to the BB spindle.
Hmmm...ya lost me there. Are you saying that the chain prevents right-
side torque transmission to the BB spindle?
> Can the rider feel a "squish" of 3 mm on one side? I would think it
> wouldn't be terribly obvious, but it might be noticeable if everything
> else (cranks, pedals) were very stiff.
I think that bending of the pedal spindle is probably the biggest
culprit under a static load, but that's a calculation for another post.
JLS
--
Carl Fogel
Dear Dave,
Applying my long-suffering dial calipers, I
determined that a 3mm squish would be roughly
equal to wearing two ordinary pairs of socks
instead of one.
The squish and rebound effect would last for
under a second at even a lowly 60 rpm.
I'm dubious that anyone can feel a 170 mm crank
flexing 3 mm at normal cadences under the sole
of the foot, a notoriously insensitive member.
I searched, but failed to find what the Just
Noticeable Difference is for such a situation.
Fechner laid the groundwork, but most examples
for the JND are for a 100 gram weight as felt on
a fingertip, not for relative movement on a crank
giving way under 50,000 grams under a foot encased
in sock and shoe.
I did find a table suggesting, however, that the
sensitivity ratio changes between the popular
fingertip test and weightlifting:
(Browse down to the red-blue table in the middle.)
A crude and possibly misleading example might be
whether we would notice a creaky floor giving way
that much (3 mm) under a carpet as we walk barefoot
across it while wearing ear-muffs.
While a crank must flex, given enough force, it
seems likely that anyone detecting a degree or two
of bend while pedalling is actually experiencing
experimenter's effect, which would vanish in double-blind
testing.
Carl Fogel
Benjamin Weiner
> jlscott3 wrote:
> > The equation for torsional stiffness of a cylindrical shaft is
> > given as: J*G/L
> > where r1 = outer radius r2 = inner radius G = shear modulus of the
> > material (80KN/mm^2 for steel) L = length of the shaft
> > I just so happen to have a Suntour Superbe Pro BB spindle sitting on my
> > desk, with an outer radius of ~8mm and inner radius of ~3.5mm, and
> > length of ~111mm. Plugging those numbers into the equation yields a
> > stiffness of 77.97 Nm/deg, or 7.956Kg-m/deg here on earth. That's
> > equivalent to 56.5 ft-lb/deg.
> I didn't check your numbers, so I'll assume your math is correct. The ID
> is only important if it is hollow all the way through. Is it?
> > So you'd have to apply 46.8 Kg of force (458.6 N) to a 170mm crank to
> > twist this BB spindle 1 degree - actually a bit more since the effective
> > length of the BB would be shorter than the overall length. Seems to me
> > that a person of 75Kg or more might be able to perceive the twist
> > induced by standing on the pedals.
> If you shorten the length a bit use a solid shaft, you might get
> slightly higher numbers. But the basic conclusion would be the same: the
> average rider should be able to get a degree or two of torsion out of
> the BB. One degree of torsion gives 1/360*2*pi*170mm = 3 mm of arc
> length. Note that the left side will demonstrate this flex, but loads on
> the right side don't apply any torque to the BB spindle.
1) The BB spindle is supported at the bearings so it is closer
to use the length from bearing to crank rather than the overall
spindle length (you certainly don't want to use more than half the
spindle length). Then L ~ 30 mm not 111 mm, yielding a stiffness
of J*G/L = 290 N m/deg.
2) Assume the force is applied at pedal center 75mm outboard of
the crank-BB junction. If the crank is at 3 o'clock then the
bending moment is ~170mm but you multiply the angle by only 75mm
to get the amount of sag at the pedal. If the crank is at
6 o'clock the bending moment is 75mm and you multiply by 170mm to
get the sag. IOW, you don't get to multiply by 170mm twice.
Standing on the crank with 50 kg weight yields a deflection angle
of (490 N * 0.170 m / 290 N m /deg) = 0.29 deg and a sag at the
pedal of 0.4 mm. That's more believable than 3 mm. Of course,
I didn't take into account any bending of the cranks, but since
they're much larger than the spindle I bet that it is small.
Note that titanium has a modulus of rigidity about 42 KN/mm^2
or half that of steel, so a square taper Ti BB spindle would
flex twice as much. That might be noticeable.
Ian G Batten
Carl Fogel <carl...@comcast.net> wrote:
> Applying my long-suffering dial calipers, I
> determined that a 3mm squish would be roughly
> equal to wearing two ordinary pairs of socks
> instead of one.
I'm not a serious cyclist. I'm not a mechanical engineer. I'm not a
biomechanical engineer. I'm fascinated that my original question has
started such in-depth discussion.
That said, wouldn't deflecting a fairly stiff piece of metal by three
millimetres, and then allowing it to rebound later, absord energy that
could otherwise be put into the wheels? Does the deflection distance
actually matter? Isn't it the issue the energy required to deflect the
system and the energy required to hold the system in the deflected state
the issue? I would guess (see disclaimer above) that when the
deflection is released the energy doesn't go anywhere useful.
ian
Brian Watson
news:40289681$1@darkstar...
It's been a while since I studied physics, but I think there are a few
problems below. Torsion is twisting so...
> 1) The BB spindle is supported at the bearings so it is closer
> to use the length from bearing to crank rather than the overall
> spindle length (you certainly don't want to use more than half the
> spindle length). Then L ~ 30 mm not 111 mm, yielding a stiffness
> of J*G/L = 290 N m/deg.
wouldn't it make sense to use the full spindle length (or the distance
between the crank arm centres) as the length as originally suggested. It is
after all the chain, via the chainwheel and crank arm at the other end of
the spindle, which provides the resistance that allows a rider pressing down
on a pedal to create torsion in the spindle. Joke about the quality of your
BB bearings withheld :-)
> 2) Assume the force is applied at pedal center 75mm outboard of
> the crank-BB junction. If the crank is at 3 o'clock then the
> bending moment is ~170mm but you multiply the angle by only 75mm
> to get the amount of sag at the pedal. If the crank is at
> 6 o'clock the bending moment is 75mm and you multiply by 170mm to
> get the sag. IOW, you don't get to multiply by 170mm twice.
I don't think this makes sense either. The length of the pedal shaft will
have not impact on the torsion created in the spindle. The only realistic
calculation assumes the crank arm is a 3 o'clock position. If it is at
other than 3 o'clock then the bending moment (your word, not mine - remember
it's a long time since I studied physics) would need to be calculated as the
horizontal distance from the centre of the spindle to the centre of the
pedal shaft. i.e.. perpendicular to the force (= the riders mass * the
acceleration due to *gravity*) being applied.
> Standing on the crank with 50 kg weight yields a deflection angle
> of (490 N * 0.170 m / 290 N m /deg) = 0.29 deg and a sag at the
> pedal of 0.4 mm. That's more believable than 3 mm. Of course,
> I didn't take into account any bending of the cranks, but since
> they're much larger than the spindle I bet that it is small.
>
> Note that titanium has a modulus of rigidity about 42 KN/mm^2
> or half that of steel, so a square taper Ti BB spindle would
> flex twice as much. That might be noticeable.
Either way, the numbers are small.
I believe that the human body has an incredible ability to perceive even
minute changes of this nature - even if we can't consciously feel the
change. As an example, I recently changed my crank arms from 170mm to
172.5mm. (I did not change for the length - it was a side effect.) After
the change the bike feels different. A tiny 2.5mm! Now I can't really say
that the crank feels longer and I am sure that if someone conducted a blind
test I would not be able to pick the longer one - but it does feel
different!
Brian.
dvt
> "Benjamin Weiner" <b...@mambo.ucolick.org> wrote in message
> news:40289681$1@darkstar...
>>to use the length from bearing to crank rather than the overall
>>spindle length (you certainly don't want to use more than half the
>>spindle length). Then L ~ 30 mm not 111 mm, yielding a stiffness
>>of J*G/L = 290 N m/deg.
> wouldn't it make sense to use the full spindle length (or the distance
> between the crank arm centres) as the length as originally suggested. It is
> after all the chain, via the chainwheel and crank arm at the other end of
> the spindle, which provides the resistance that allows a rider pressing down
> on a pedal to create torsion in the spindle. Joke about the quality of your
> BB bearings withheld :-)
I think we have a misunderstanding here. I'm with Brian on this one.
Maybe Benjamin can point out where we are wrong. The stiffness being
calculated, as I understand it, is the torsional stiffness of the
spindle. The torque is being applied at the left crank interface and the
resisting force is applied at the right crank interface (via the
chainrings). I *think* Benjamin is thinking of some other type of
stiffness, perhaps the bending stiffness of the spindle.
>>2) Assume the force is applied at pedal center 75mm outboard of
>>the crank-BB junction. If the crank is at 3 o'clock then the
>>bending moment is ~170mm but you multiply the angle by only 75mm
>>to get the amount of sag at the pedal. If the crank is at
>>6 o'clock the bending moment is 75mm and you multiply by 170mm to
>>get the sag. IOW, you don't get to multiply by 170mm twice.
I didn't understand this part. If you're talking about the spindle
bending rather than twisting, I think it makes sense. But I think the
fundamental equation of J*G/L and the resulting numbers are not applicable.
> Either way, the numbers are small.
>
> I believe that the human body has an incredible ability to perceive even
> minute changes of this nature - even if we can't consciously feel the
> change. As an example, I recently changed my crank arms from 170mm to
> 172.5mm. (I did not change for the length - it was a side effect.) After
> the change the bike feels different. A tiny 2.5mm! Now I can't really say
> that the crank feels longer and I am sure that if someone conducted a blind
> test I would not be able to pick the longer one - but it does feel
> different!
I rode a bike for a few weeks that had a 165 mm crank on one side and
170 mm on the other. I noticed it when I first hopped on the bike, but
after a few minutes, I completely forgot about it. So unlike you, I'm
not very sensitive to these things. I've never noticed crank flex except
as manifested by derailer rub, and that derailer rub was probably caused
by frame flex.
dvt
> Originally posted by Dvt
>>One degree of torsion gives 1/360*2*pi*170mm = 3 mm of arc length.
>>Note that the left side will demonstrate this flex, but loads on the
>>right side don't apply any torque to the BB spindle.
> Hmmm...ya lost me there. Are you saying that the chain prevents right-
> side torque transmission to the BB spindle?
Yes. There is torque applied to lift the left pedal through the upward
part of the stroke, but it's small enough to pretend that it doesn't
exist (see the Powercranks thread that's going on right now.)
dvt
> That said, wouldn't deflecting a fairly stiff piece of metal by three
> millimetres, and then allowing it to rebound later, absord energy that
> could otherwise be put into the wheels?
A very small amount of energy. Aluminum, when deformed elastically, has
a loss factor of something like 0.01%. Steel has a slightly higher loss
factor (Cremer-Heckl-Ungar, Structure-Borne Sound). This means that
nearly all of the energy you put into that deflection is recovered.
> Does the deflection distance actually matter?
Not unless you go beyond the elastic region of the metal. If you did
that, your crank wouldn't last very long.
The reason for calculating the deflection distance is to see if the
movement was large enough to be felt. That, I think, was the original
intent of the thread.
Ian G Batten
> Ian G Batten wrote:
> > That said, wouldn't deflecting a fairly stiff piece of metal by three
> > millimetres, and then allowing it to rebound later, absord energy that
> > could otherwise be put into the wheels?
>
> A very small amount of energy. Aluminum, when deformed elastically, has
> a loss factor of something like 0.01%. Steel has a slightly higher loss
> factor (Cremer-Heckl-Ungar, Structure-Borne Sound). This means that
> nearly all of the energy you put into that deflection is recovered.
I don't doubt it. However, would it be recovered in a useful way? I'm
imagining the crank being put under tension during the downstroke,
absorbing energy, which is then released during the upstroke. Would
that release of energy do anything useful?
ian
Mark Hickey
>Brian Watson wrote:
>> "Benjamin Weiner" <b...@mambo.ucolick.org> wrote in message
>> news:40289681$1@darkstar...
>>>1) The BB spindle is supported at the bearings so it is closer
>>>to use the length from bearing to crank rather than the overall
>>>spindle length (you certainly don't want to use more than half the
>>>spindle length). Then L ~ 30 mm not 111 mm, yielding a stiffness
>>>of J*G/L = 290 N m/deg.
>
>> wouldn't it make sense to use the full spindle length (or the distance
>> between the crank arm centres) as the length as originally suggested. It is
>> after all the chain, via the chainwheel and crank arm at the other end of
>> the spindle, which provides the resistance that allows a rider pressing down
>> on a pedal to create torsion in the spindle. Joke about the quality of your
>> BB bearings withheld :-)
>
>I think we have a misunderstanding here. I'm with Brian on this one.
>Maybe Benjamin can point out where we are wrong. The stiffness being
>calculated, as I understand it, is the torsional stiffness of the
>spindle. The torque is being applied at the left crank interface and the
>resisting force is applied at the right crank interface (via the
>chainrings). I *think* Benjamin is thinking of some other type of
>stiffness, perhaps the bending stiffness of the spindle.
The original question was whether a rider could feel flex in the
bottom bracket spindle / crank. To me, the only type of flex that's
going to be felt is the sag of the spindle outboard of the BB bearing.
True enough, there may be a very small amount of torsional "wind-up"
involved when putting weight on the left pedal, but I doubt seriously
anyone would really notice the left crank "accelerating" by 1mm over
the first half the stroke, and "decelerating" by the same amount over
the second half.
jlscott3
jlscott3 wrote:
dvt> One degree of torsion gives 1/360*2*pi*170mm = 3 mm of arc length.
dvt> Note that the left side will demonstrate this flex, but loads on
dvt> the right side don't apply any torque to the BB spindle.
jls> Hmmm...ya lost me there. Are you saying that the chain prevents right-
jls> side torque transmission to the BB spindle?
dvt> Yes. There is torque applied to lift the left pedal through the
dvt> upward part of the stroke, but it's small enough to pretend that it
dvt> doesn't exist
I don't want to belabor the point - if I'm really missing something
maybe we can continue via email - but it seems to me that in a static
situation (rider standing on pedals, track-standing or coasting) the
left and right crankarms would torque the BB spindle in equal amounts
but opposite directions. Under pedaling load the chain transmits torque
to the rear hub so I can see where it would transmit less torque to the
BB spindle, and the torsion in the BB spindle would be less in that
situation as well.
dvt> (see the Powercranks thread that's going on right now.)
Hmm...that thread does not resemble my idea of a good time on Usenet :-)
JLS
--
Carl Fogel
Dear Brian,
It would be interesting to have some double-blind testing
on how well riders can actually distinguish 170 mm from
172.5 mm cranks. For some reason, the psychology departments
that care about the Just Noticeable Difference don't seem
to use this as an example--they prefer 100 gram weights on
student fingertips. The crank length example amounts to
only 1.5%. Scaled up, this is about the difference between
someone six feet tall versus someone an inch taller--you need
a scale to tell the difference unless you compare them side
by side.
Weight, length, light, sound, and other stimulus vary in
how much is needed for us to notice a difference--and with
what we use to notice it, fingertips for small weights
behaving differently than lifting large weights.
There may be an important difference between noticing
how long a crank is and how much it bends. That is, we
might be able to notice our feet going absolutely further
down and extra 2.5 mm (I'm a little dubious, since this is
still about the thickness of an extra sock), but not able
to notice the crank twisting untwisting smoothly for 3 mm
on the spindle through a half pedal cycle. After all, what
would our feet be comparing this flex to, if all cranks
twist all spindles?
Carl Fogel
dvt
> ...it seems to me that in a static
> situation (rider standing on pedals, track-standing or coasting) the
> left and right crankarms would torque the BB spindle in equal amounts
> but opposite directions.
You are correct. I neglected those situations. I was thinking only of
"normal" riding under power.
Brian Watson
news:8bbde8fc.04021...@posting.google.com...
> "Brian Watson" <not4...@yahoo.com> wrote in message
news:<2S1Wb.50259$Wa....@news-server.bigpond.net.au>...
> > "Benjamin Weiner" <b...@mambo.ucolick.org> wrote in message
> > news:40289681$1@darkstar...
> > > dvt <dvt_...@psu.edu> wrote:
> > > > jlscott3 wrote:
[...]
Carl,
I said in my comment that I suspected my "theory" would not stand up to
testing. However, based on my experience I would suggest that if two
identical bikes were built and then one had the BB spindle, or crank arms,
replaced such that the crank arm could deflect 3mm (at he pedal) more than
that of the other bike - a fair percentage of people could pick the
difference. I am not saying they would be able to attribute the difference
to crank flex or BB spindle torsion - I just think they would feel that the
bikes were different. This is just my hypothesis which I don't have the
resources (or the desire really...) to test.
I guess this is as good a place as any to also suggest that the ability to
detect differences of this type varies greatly between people. My wife can
quite happily drive our car about with a tire half flat etc. whereas I
cannot tolerate even small variations in pressure. One of my sons cannot
tolerate even a tiny grain of sand in his shoe, my other can happily walk
around with a whole river bed in there :-) I know these examples are
completely different to detecting crank flex but I humbly suggest that there
are similar things going on in the brain.
I would also like to make a small correction to my post above - I said the
bike *feels* different with the different cranks. Correction - it *felt*
different. It only took a few minutes for it to feel just the same again.
Similar to Dave's experience with the 165/170 combination.
Brian
> Carl Fogel
onefred
It's probably noticeable when you factor in everything else that's flexing.
Dave