New Cervelo Model
am...@physics.utoronto.ca
http://www.cyclingnews.com/photos/2006/feb06/hetvolk06/index.php?id=live/DSC02027
from what I've heard the seatstays contribute very little to the
structure. the chainstays and seatube are strong enough that the bike
would be ridable without seatstays.
-Amit
Mike Reed
stays?
Luke
Reed <waterr...@gmail.com> wrote:
> What about braking force? I guess that works out to a tension on the
> stays?
If the seatstays are truly superfluous, why doesn't Cervelo mount the
rear brake on the seattube and rid itself of the vestiges?
Luke
Ryan Cousineau
Luke <lucasi...@rogers.com> wrote:
That is a question best directed to the UCI. In short: the rules require
a diamond-frame bicycle (with a bunch of dimensional restrictions that
are too boring to explain). The seat stays are required for that reason.
--
Ryan Cousineau rcou...@sfu.ca http://www.wiredcola.com/
"I don't want kids who are thinking about going into mathematics
to think that they have to take drugs to succeed." -Paul Erdos
Scott
I had a conversation w/ a local yet prominent framebuilder in CO,
regarding the use of carbon stays on otherwise non-carbon bikes. He
said that the reason no one expressed concerns over the potential for
bad glue jobs is that there's almost no stress on the seatstays and
therefore the junction doesn't have to be that strong.
It should follow that w/ sufficiently strong chainstays, the seatstays
are even less necessary if not completely superfluous.
Michael Press
<1141072940.6...@j33g2000cwa.googlegroups.com>,
am...@physics.utoronto.ca wrote:
Where have you heard this?
Consider an axis through the rear axle, directed fore-aft,
and parallel to the ground. Torque at the axle about this
axis is normally resisted by seat stays. Remove the seat
stays and that torque must be resisted by the joint at the
chain stays and the bottom bracket.
More generally, the two chain stays, two seat stays, seat
tube, and rear axle form a tetrahedron. In a tetrahedron
with rigid members the joints could be pinned without
compromising the strength of the structure, making the
tetrahedron an efficient spacial structure. Analogously,
a triangle of rigid members can be pinned at the joints
and yet be very strong in its plane. Contrast this with a
pinned square; it will flop around when subject to stress.
In the real world the rigid members actually flex;
therefore the joints must be built to resist this flex. So
we put a bridge between the chain stays near the bottom
bracket, and a bridge between the seat stays near the seat
tube.
--
Michael Press
Luke
Michael Press <ja...@abc.net> wrote:
How do you account for designs such those of Softrides[1]? I'm no
proponent of this approach; chainstays of dimensions equal to those of
motorcycle swingarms strike me as incredibly ugly. Is this a case of
simply over-engineering the chainstays and BB junction to handle the
torsional loads?
I'm unsure from the photo whether Cervelo's design utilizes CF
chainstays; perhaps advances in the application of that material may be
able to provide strength comparable to that Softride achieves with its
gargantuan aluminum stays.
Luke
b...@mambo.ucolick.org
> am...@physics.utoronto.ca wrote:
> > from what I've heard the seatstays contribute very little to the
> > structure. the chainstays and seatube are strong enough that the bike
> > would be ridable without seatstays.
> Consider an axis through the rear axle, directed fore-aft,
> and parallel to the ground. Torque at the axle about this
> axis is normally resisted by seat stays. Remove the seat
> stays and that torque must be resisted by the joint at the
> chain stays and the bottom bracket.
>
> More generally, the two chain stays, two seat stays, seat
> tube, and rear axle form a tetrahedron. In a tetrahedron
> with rigid members the joints could be pinned without
> compromising the strength of the structure, making the
> tetrahedron an efficient spacial structure. Analogously,
> a triangle of rigid members can be pinned at the joints
> and yet be very strong in its plane. Contrast this with a
> pinned square; it will flop around when subject to stress.
>
> In the real world the rigid members actually flex;
> therefore the joints must be built to resist this flex. So
> we put a bridge between the chain stays near the bottom
> bracket, and a bridge between the seat stays near the seat
> tube.
Amit said that (he heard that) the bike would be rideable without
the seatstays. This is probably true. However, I believe it
would last longer with the seatstays. W/o seatstays, the chainstays
will flex more at the BB joint and fatigue quicker. I'm thinking
of torque about the crank axle that is applied when the bike
goes over a bump - that is what the seatstays resist. Torque
about the axis that runs from rear hub to the BB shell is applied
by pedaling force/chain tension, not by the rear wheel, and causes
torsion in the BB shell, independent of the seatstay arrangement.
Seatstays don't bear a lot of stress and this is why, even on a
classic bike, they can be made of pretty skinny lightweight tubing.
The seatstay and chainstay bridges are not there to resist pedaling
induced flex in the rear triangle. They are there to provide a
convenient place to attach fenders. The seatstay bridge is also
a useful place to put a rear caliper brake, and it does help reduce
flex of the stays with a rear cantilever brake. The bridges are
usually made of small diameter tubing that is not very stiff in bending
(much less so than the chainstays) and would not add much to the
stiffness of the rear triangle. Bicycles have been constructed without
seatstay or chainstay bridges, with no ill effects. Other than the
difficulty of attaching fenders (and getting the rear tire stuck
between the chainstays when removing the wheel).
Dan Connelly
From which it follows that carbon seatstays are essentially a gimmick, anyway.
Dan
Michael Press
Luke <lucasi...@rogers.com> wrote:
Yes. Without seat stays there can be enough torque to
twist the chain stays sufficiently to induce
`auto-shifting.'
> I'm unsure from the photo whether Cervelo's design utilizes CF
> chainstays; perhaps advances in the application of that material may be
> able to provide strength comparable to that Softride achieves with its
> gargantuan aluminum stays.
The seat stays and seat stay joints do not need to be as
strong as the chain stays and chain stay joints. Still,
they must be strong enough to make the structure a
tetrahedron.
The bicycle frame is an elegant design. But there is
always someone to improve it. {:^)
Custom frame builders typically take the approach that a
good musician takes to Mozart: be skilled enough to
implement the design, then get out of the way.
--
Michael Press
Michael Press
<1141166145....@e56g2000cwe.googlegroups.com>,
"b...@mambo.ucolick.org" <b...@mambo.ucolick.org> wrote:
First of all, for this analysis the model of a rear
triangle is insufficient. We need to consider the space
structure for what it is--a tetrahedron. The various
forces have components in three dimensions.
We have had several reports here of fatigue failure at the
chain stay bridge; enough to suppose that there could be
large stresses at this joint sufficient to require its
inclusion.
Finite element analysis shows that the bottom bracket and
bottom bracket joints are the most critical for reducing
frame flex.
I am puzzled how you think that the stresses here are so
small as to render the chain bridge superfluous. Could a
frame builder chime in here?
> They are there to provide a
> convenient place to attach fenders. The seatstay bridge is also
> a useful place to put a rear caliper brake, and it does help reduce
> flex of the stays with a rear cantilever brake. The bridges are
> usually made of small diameter tubing that is not very stiff in bending
> (much less so than the chainstays) and would not add much to the
> stiffness of the rear triangle. Bicycles have been constructed without
> seatstay or chainstay bridges, with no ill effects. Other than the
> difficulty of attaching fenders (and getting the rear tire stuck
> between the chainstays when removing the wheel).
--
Michael Press
Andy Morris
>
> First of all, for this analysis the model of a rear
> triangle is insufficient. We need to consider the space
> structure for what it is--a tetrahedron. The various
> forces have components in three dimensions.
>
> We have had several reports here of fatigue failure at the
> chain stay bridge; enough to suppose that there could be
> large stresses at this joint sufficient to require its
> inclusion.
>
> Finite element analysis shows that the bottom bracket and
> bottom bracket joints are the most critical for reducing
> frame flex.
>
> I am puzzled how you think that the stresses here are so
> small as to render the chain bridge superfluous. Could a
> frame builder chime in here?
>
>
Possibly the chain stay bridge caused the failure by providing a stress
raiser.
--
Andy Morris
AndyAtJinkasDotFreeserve.Co.UK
Love this:
Put an end to Outlook Express's messy quotes
http://home.in.tum.de/~jain/software/oe-quotefix/
b...@mambo.ucolick.org
> "b...@mambo.ucolick.org" <b...@mambo.ucolick.org> wrote:
> > The seatstay and chainstay bridges are not there to resist pedaling
> > induced flex in the rear triangle.
>
> First of all, for this analysis the model of a rear
> triangle is insufficient. We need to consider the space
> structure for what it is--a tetrahedron. The various
> forces have components in three dimensions.
I used "rear triangle" colloquially, in the sense of back part of
a bike frame.
> We have had several reports here of fatigue failure at the
> chain stay bridge; enough to suppose that there could be
> large stresses at this joint sufficient to require its
> inclusion.
I don't see how welding a ~10mm diameter piece of metal
between two ~22mm diameter tubes that are both attached
to a rather large BB shell <50mm away is going to add any
significant stiffness or strength to the structure. It is
probably more likely to cause a heat affected zone at or
near the braze or weld between bridge and stay. More metal
isn't automatically better. It has to be applied in the right
places (like a Bontrager gusset).
> Finite element analysis shows that the bottom bracket and
> bottom bracket joints are the most critical for reducing
> frame flex.
>
> I am puzzled how you think that the stresses here are so
> small as to render the chain bridge superfluous. Could a
> frame builder chime in here?
I don't think the stresses are small - that is one of the highest
stress places on a frame. Rather, I think the bridges used on
conventional road bikes are too small to have any meaningful
effect on the structure as-used (with a rear axle making the
rear tetrahedron rigid), and will not deter the stresses that
cause fatigue failure.
Here is a report by Jim Adney on deflection testing:
Andrew Muzi's and Jobst's opinions:
<http://groups.google.com/group/rec.bicycles.tech/msg/bd9ab8f258a59026>
Googling "chainstay bridge" in this group brings up a number of
opinions. I was never convinced by any of the arguments in favor
of the chainstay bridge as a significant structural element.
jlee
Michael Press wrote:
> In article
> <1141072940.6...@j33g2000cwa.googlegroups.com>,
> am...@physics.utoronto.ca wrote:
>
> > This is the new Cervelo model :
> >
> > http://www.cyclingnews.com/photos/2006/feb06/hetvolk06/index.php?id=live/DSC02027
> >
> > from what I've heard the seatstays contribute very little to the
> > structure. the chainstays and seatube are strong enough that the bike
> > would be ridable without seatstays.
>
> Where have you heard this?
>
> Consider an axis through the rear axle, directed fore-aft,
> and parallel to the ground. Torque at the axle about this
> axis is normally resisted by seat stays. Remove the seat
> stays and that torque must be resisted by the joint at the
> chain stays and the bottom bracket.
How/why would such a torque be created? I don't think that is a
realistic load/stress.
The rear of a bicycle is essentially a truss. There is downward load on
the seat tube which is held up by the rear dropout. The beauty of this
and the reason the seatstays can be so thin or even poorly joined (for
CF) is that the seatstays are mostly in compression, while the
seatstays are mainly in tension (compression during pedaling, perhaps).
It may be easier for people to imagine an "L" bracket used to support a
shelf on the wall. By both statics and by simple intuition, a closed
bracket (triangle) would be far stronger than an open L bracket. The
third member thus acts in compression to hold the vertical load of the
shelf, while the top member (just under the shelf) is in tension (to
keep the shelf from pulling away from the wall - which is what happens
when poorly secured) as well as some bending moment (that means it has
to bend) if the third member were to say compress due to the load or an
impact.
Likewise, imagine the seatube as the wall and the shelf as the
chainstays. The load is at the rear dropout upwards on the frame
(downwards on the hub/wheels), so you just have to turn the bike upside
down to parallel the diagram of the shelf on the wall. The seatstays
thus has to be designed with just enough strength to withstand the
compression forces without buckling. The criteria though is not simply
the rider weight, but the maximum impact you can imagine the bike
sustaining. The problem with the photo is that if those are thin-walled
tubes, I'm not sure how long they would survive in the real world or
potholes, cobbles, sewer-caps, or wheelies, if you're Floyd. Longevity,
however, is usually not the priority for a sponsored racing
professional.
You can, of course, have bicycles without seatstays, just as there are
open L brackets. Those brackets usually have ribbing in them to
strengthen them. Likewise, those aero bikes like some Zipps or Lotuses
that don't have seatstays have similarly strengthened BB/chainstays.
They're well suited for controlled conditions like track events but I
wouldn't want anyone to to be hitting any rough patches during a
descent on those...
Scott
You're right. He admitted that, too. Went on to joke about how it was
such a huge profit maker for him, as it made it easier and less
expensive to build the frame and yet he could sell it for more.
amit....@gmail.com
b...@mambo.ucolick.org wrote:
I did see this bike (the R3) in person this weekend and the cervelo
employee said the same thing i had heard, the frame without the
seatstays is very strong and stiff.
if you follow this link you can see a test of the bayonne (the yet to
be released lightweight version of the R3) , and the wispy seatstays
are mentioned :
http://www.cervelo.com/bikes.aspx?bike=Bayonne2006
> Googling "chainstay bridge" in this group brings up a number of
> opinions. I was never convinced by any of the arguments in favor
> of the chainstay bridge as a significant structural element.
if you follow this link
http://www.cervelo.com/section.aspx?m=Engineering
you can check out the "8 grams of engineering article", and it explains
the design of the soloist chainstay bridge. i know that the early
soloists had no bridge and last year's model had one. if this article
is to be believed, the implementation of a chainstay bridge isn't so
simple and the bridge had to be carefully designed.
b...@mambo.ucolick.org
> if you follow this link
>
> http://www.cervelo.com/section.aspx?m=Engineering
>
> you can check out the "8 grams of engineering article", and it explains
> the design of the soloist chainstay bridge. i know that the early
> soloists had no bridge and last year's model had one. if this article
> is to be believed, the implementation of a chainstay bridge isn't so
> simple and the bridge had to be carefully designed.
That's an interesting article. Thanks. It's rare to see a
bike company actually talk about that kind of technical detail.
It seems that some of the failed designs failed to improve
fatigue life because they put welds in vulnerable places or
because adding metal redistributed the stress to vulnerable
places, or both. More metal isn't necessarily better.
The EFBe test uses high loads to fail a bike before
100,000 cycles, because real bikes last more pedaling
cycles than that. This has been debated before in rbt,
but presumably a chainstay bridge that is worse on the
EFBe test is worse in the real world too.
The actual bridge they used is pretty big compared to the
chainstays so it has a real effect on the stiffness of the
assembly. In classical steel construction the bridge is
a pretty wimpy tube so I think it's fairly cosmetic.
As proof that I shuld be riding instead (hey, it's dark out)
here are some pictures of chainstay bridge areas
on frames, and three different approaches:
http://www.ucolick.org/~bjw/misc/rbt/cstays.html
I didn't have access to any stay-free bikes, but they do exist.
amit....@gmail.com
b...@mambo.ucolick.org wrote:
> That's an interesting article. Thanks. It's rare to see a
> bike company actually talk about that kind of technical detail.
> It seems that some of the failed designs failed to improve
> fatigue life because they put welds in vulnerable places or
> because adding metal redistributed the stress to vulnerable
> places, or both. More metal isn't necessarily better.
yes, and the article mentions they had failures at the bridge in their
prototype soloists so that's why (i suppose) it was left out on the
first production models.
> The actual bridge they used is pretty big compared to the
> chainstays so it has a real effect on the stiffness of the
> assembly. In classical steel construction the bridge is
> a pretty wimpy tube so I think it's fairly cosmetic.
the effect isn't measured in (just) stiffness but also in fatigue
resistance, at least with respect to the EFBe test.
> As proof that I shuld be riding instead (hey, it's dark out)
> here are some pictures of chainstay bridge areas
> on frames, and three different approaches:
>
> http://www.ucolick.org/~bjw/misc/rbt/cstays.html
yes, but a bridge arrangment that looks good might actually be bad, as
that article shows and the combination of the bridge and gusset was the
worst.
> I didn't have access to any stay-free bikes, but they do exist.
if you mean chainstay bridge-free, cannondale is a common example.