Strength of frame without chainstay bridge
Russell Seaton
"loaded" touring frame without a chainstay bridge. The frame is
welded Kinesis 7005 Superlight butted aluminum tubing, according to
the sticker. Chainstays are 17" long. Cantilever brakes will be
used. 130mm rear OLD spacing.
I know mounting a rear fender will require some ingenuity. But with
Blackburn P-clips at the rear dropouts and the seatstay bridge, I
figure the fender should be fairly well secured. Wire between the
fender and chainstays behind the bottom bracket will probably be
enough to hold the fender in place.
Zog The Undeniable
Effectively the chainstays are longer, so the rear end is more flexible
side-to-side (not a good thing). It's occasionally done on racing
frames with really short rear ends, but I'm puzzled as to why it was
done on a touring bike.
daveornee
I wonder if they skipped the bridge due to welding it past the thick
part of the butting. I have a ti touring bicycle with 17.5" chain stays.
The bridge makes the back end much stiffer. OLD is set at 132.5 mm and
putting in a 135 mm OLD wheel takes a little "grunting". I would think
that there is a seatstay bridgeon your bicycle, but you may still want
to use a "Brake Booster". Fender mounting is always a challenge to make
them secure, neat, and provide just the right clearance around the
entire tire. I wish I could show some manufacturers what a fendered
bicycle looks like with big touring tires. ... then put some quality
racks on. Builders like Bruce Gordon and Waterford's Adventure Cycle
1900 seem to have this figured out... and some others do to,,, but many
don't yet have a clue. Maybe a couple of properly sized "P-clips" with a
90 degree twist would secure the front part of your fender. Otherwise
some zip-ties may do the job. If you use zip-ties, get the best quality
ones you can find, and take some spares with you.
--
Peter Cole
news:e90052be.04062...@posting.google.com...
> How does not having a chainstay bridge affect the frame?
It doesn't.
carl...@comcast.net
Dear Peter,
Russell may have been hoping for an explanation, too.
As I understand things, the rear section of a bicycle is a
tetrahedron (two triangles sharing a two common sides, the
seat tube and the axle). It's a darned clever arrangement in
that it's about as strong as possible.
Like the brake-bridge between the seat-stays, the chain-stay
bridge may not add any strength in terms of resisting
twisting of the rear structure.
If you google for jobst and tetrahedron, you should find
some better explanations like this:
or
Carl Fogel
John Thompson
> I wonder if they skipped the bridge due to welding it past the thick
> part of the butting.
AFAIK, chainstays aren't butted. At least they're not on the Reynolds
531/753 Isiwata, Tange, and Columbus steel tube sets I've worked with in
the past.
--
-John (jo...@os2.dhs.org)
Peter Cole
> On Sat, 26 Jun 2004 18:01:08 GMT, "Peter Cole"
> <peter_cole_n...@comcast.net> wrote:
>
> >"Russell Seaton" <russell...@yahoo.com> wrote >
> >
> >It doesn't.
> >
>
> Dear Peter,
>
> Russell may have been hoping for an explanation, too.
>
> As I understand things, the rear section of a bicycle is a
> tetrahedron (two triangles sharing a two common sides, the
> seat tube and the axle). It's a darned clever arrangement in
> that it's about as strong as possible.
>
> Like the brake-bridge between the seat-stays, the chain-stay
> bridge may not add any strength in terms of resisting
> twisting of the rear structure.
>
> If you google for jobst and tetrahedron, you should find
> some better explanations like this:
>
Carl,
You're making things unnecessarily complicated. There's no need to drag in
tetrahedrons or brake bridges. Ignoring all that (safely) for the question at
hand, the chainstay bridge is simply too close to the BB (which is performing
the same support function) to make any difference, even if there were a load
present (which there isn't). It gives you a place to mount fenders.
carl...@comcast.net
<peter_cole_n...@comcast.net> wrote:
Dear Peter,
I'm not sure, but you may be mistaken. If I understand the
theory, it doesn't matter where you add a bridge--it doesn't
increase the rigidity of the tetrahedron.\
That is, it doesn't matter how close the bridge is to the
end. if you took the wheel out and welded a bridge halfway
up from the axle between the chain-stays or the seat-stays,
it wouldn't improve rigidity in terms of bicycling loads.
The tubes suffer only compression and tension in this
design, neither of which is causing the seat-stays or the
chain-stays to move sideways in a bending movement.
Or so I hear.
Carl Fogel
jobst....@stanfordalumni.org
> How does not having a chainstay bridge affect the frame? I have a
> "loaded" touring frame without a chainstay bridge. The frame is
> welded Kinesis 7005 Superlight butted aluminum tubing, according to
> the sticker. Chainstays are 17" long. Cantilever brakes will be
> used. 130mm rear OLD spacing.
The chainstay bridge has two purposes, attaching a mud guard and
preventing the rear tire from getting stuck in the narrows when pushed
forward from horizontal dropouts. I recall watching a rider try to
remove a rear wheel by opening the QR and knocking the wheel forward
with the hand only to have it jam between the chainstays with no
bridge. Yanking it out of engagement put it back in the dropouts that
had the typical gouged retentive grooves from the axle nuts. This
went back and forth a couple of times before the wheel came out with
more careful manipulation.
Depending on frame size and geometry, the bridge doesn't even serve
that purpose. It is never structurally necessary, the rear triangle
actually being a tetrahedron, the stiffest solid body available.
> I know mounting a rear fender will require some ingenuity. But with
> Blackburn P-clips at the rear dropouts and the seatstay bridge, I
> figure the fender should be fairly well secured. Wire between the
> fender and chainstays behind the bottom bracket will probably be
> enough to hold the fender in place.
I am not visualizing what you mean.
Jobst Brandt
jobst....@stanfordalumni.org
Peter Cole
news:tknrd05vqu2lv7q27...@4ax.com...
I think you're missing my point entirely -- why don't we give this a pass?
Kinky Cowboy
wrote:
>Russell Seaton writes:
>
>> How does not having a chainstay bridge affect the frame? I have a
>> "loaded" touring frame without a chainstay bridge.
>
>Depending on frame size and geometry, the bridge doesn't even serve
>that purpose. It is never structurally necessary, the rear triangle
>actually being a tetrahedron, the stiffest solid body available.
>
>Jobst Brandt
>jobst....@stanfordalumni.org
We've been here before. The rear triangle isn't a tetrahedron, because
if it was the chainstays would meet in the centre of the BB shell. The
BB shell, chainstays and rear axle form a trapezium, which is deformed
by axial loads from the rear wheel. Bracing this trapezium at some
point distant from, but parallel to, the BB axis will cause this
deforming tendency to be resisted by a bending moment in the
chainstays, whereas without the bridge the entire load must be
resisted by the BB/Chainstay junction attemting to change it's angle.
For this reason, bridgeless designs usually rely on reinforcement of
this junction, such as gussets, extended BB shell or, in the case of
lugged steel designs, longer and heavier points on the BB shell.
Admittedly, the seatstays (also braced, by the brake bridge) also
resist this load, but anybody who thinks the chainstay bridge serves
no structural function should have a look at the swingarm of the
nearest high performance motorcycle.
It is possible to properly design a bicycle without a chainstay
bridge, and I do not presume to know which approach is ultimately
"better", by whatever test you happen to mean by better; this will
vary according to geometry and material choice, but in general if a
designer has chosen to use a bridge, it's likely to be there for
structural reasons; it certainly isn't required for tyre restraint or
mudguard mounting on my track or dirt-jump bikes, both of which enjoy
the structural benefit of substantial chainstay bridges (and, for that
matter, "brake" bridges, even though one is braked by the sprocket and
the other by a disc)
Kinky Cowboy*
*Batteries not included
May contain traces of nuts
Your milage may vary
carl...@comcast.net
Dear Kinky,
I think that I see your point about the chain-stay bridge
being different that the seat-stay bridge because the
chain-stays don't converge to a point but instead remain a
little separated (roughly a tire-width or 0.8 basset noses).
Where I'm not following you (yet) is how this modified
tetrahedron (trapezium?) is distorting. That is, what parts
will bend less which way with the bridges?
(I realize that this kind of thing is hard to describe in
words, so I appreciate any explanation that you attempt.)
I'm not sure about the motorcycle analogy. Aren't most
modern street bikes a trailing fork back to the axle, with a
heavy reinforcement as you describe, but with a only little
triangle whose top goes to a monoshock and whose
"seat-stays" arrive about half-way down the "chain-stays"
instead of at the axle?
To be truly similar, wouldn't a bicycle frame have to dangle
its axle on a foot-long pair of horizontal trailing struts?
I'm not sure at all about this, modern street motorcycles
being purely theoretical to me, so take a deep breath before
pointing out any gaping holes in my question. I'm really
just asking if the axles are mounted well outside the stiff
triangular frame on a motorcycle, but almost inside it on a
bicycle.
Thanks,
Carl Fogel
Benjamin Weiner
> It is possible to properly design a bicycle without a chainstay
> bridge, and I do not presume to know which approach is ultimately
> "better", by whatever test you happen to mean by better; this will
> vary according to geometry and material choice, but in general if a
> designer has chosen to use a bridge, it's likely to be there for
> structural reasons; it certainly isn't required for tyre restraint or
> mudguard mounting on my track or dirt-jump bikes, both of which enjoy
> the structural benefit of substantial chainstay bridges (and, for that
> matter, "brake" bridges, even though one is braked by the sprocket and
> the other by a disc)
"substantial chainstay bridges"
There's the rub. On most road bikes (and quite a few MTBs)
the chainstay bridge is a weeny little piece of metal, much
smaller in diameter than the chainstays, much less the BB shell.
It can't possibly add much to the stiffness of the structure.
Of course, if you used a bridge as large in diameter as the
chainstays, it could add stiffness, but there you're effectively
shortening the chainstays, substituting the bridge for the BB
shell. I think at least some CS and SS bridges are there because
people expect to see them.
Kinky Cowboy
>Dear Kinky,
>
>I think that I see your point about the chain-stay bridge
>being different that the seat-stay bridge because the
>chain-stays don't converge to a point but instead remain a
>little separated (roughly a tire-width or 0.8 basset noses).
>
>Where I'm not following you (yet) is how this modified
>tetrahedron (trapezium?) is distorting. That is, what parts
>will bend less which way with the bridges?
>
>(I realize that this kind of thing is hard to describe in
>words, so I appreciate any explanation that you attempt.)
Very difficult without a drawing, but without a bridge, pushing the
hub axle sideways will cause the angle between the chainstay and BB to
change. With the bridge, this still happens, but in addtion, the
angles between the bridge and the stays also have to change, and the
bridge either has to change length or the stays have to bend. There is
also some neding of the dropouts in both cases. All other things being
equal, the same force wil result in less deflection with the bridge in
place.
>I'm not sure about the motorcycle analogy. Aren't most
>modern street bikes a trailing fork back to the axle, with a
>heavy reinforcement as you describe, but with a only little
>triangle whose top goes to a monoshock and whose
>"seat-stays" arrive about half-way down the "chain-stays"
>instead of at the axle?
True, a motorcycle has to do without seatstays, and the chainstays are
much further apart, making the structure less like a triangle than a
bicycle
>To be truly similar, wouldn't a bicycle frame have to dangle
>its axle on a foot-long pair of horizontal trailing struts?
As many suspension designs do
>I'm not sure at all about this, modern street motorcycles
>being purely theoretical to me, so take a deep breath before
>pointing out any gaping holes in my question. I'm really
>just asking if the axles are mounted well outside the stiff
>triangular frame on a motorcycle, but almost inside it on a
>bicycle.
>
>Thanks,
>
>Carl Fogel
Kinky Cowboy*
Kinky Cowboy
wrote:
Because the bridge is short, and the main load on it is
compressive/tensile (I need to do the drawing to work out which) it
doesn't need to be as fat as the stays. Plenty of cross section will
help, but it's largely irrelevant whether it's a small thick walled
tube or a large thin walled one, because it's so short and not subject
to a great deal of bending load. In fact, my old TT bike had a solid
bridge of diablo form tapering to about 6mm diameter in the centre.
I have no doubt that the "effectively shortening the chainstays"
comment has as much bearing on the matter as any lattice girder
effect, but some metal has to be involved in this shortening, whether
it's a bridge or, as is seen on many aluminium MTB frames, a huge CNC
carved block making up the BB shell and the first few inches of
chainstay. Of course, if you build the stays so short that there is no
room for a bridge, you have to rely on the inherent stiffness of the
unbraced structure, and plenty of bikes have been made perfectly
rideable this way.
The debate is really about which solution is best in a particular
application; the bridged structure probably weighs less, on a
lightweight steel or titanium frame, than using thicker walled stays
to acheive the same effect, it's cheap to build and provides a
convenient mudguard mounting point for touring bikes. You'd likely do
something different on a true monocoque carbon fibre track frame.
Phil Brown
>AFAIK, chainstays aren't butted. At least they're not on the Reynolds
>531/753 Isiwata, Tange, and Columbus steel tube sets I've worked with in
>
>the past.
The walls are tapered on good quality stays.
Phil brown
jobst....@stanfordalumni.org
>>> How does not having a chainstay bridge affect the frame? I have a
>>> "loaded" touring frame without a chainstay bridge.
>> Depending on frame size and geometry, the bridge doesn't even serve
>> that purpose. It is never structurally necessary, the rear triangle
>> actually being a tetrahedron, the stiffest solid body available.
> We've been here before. The rear triangle isn't a tetrahedron,
> because if it was the chainstays would meet in the centre of the BB
> shell.
The virtual intersection of these tubes is guaranteed by the massive
BB shell reinforced by BB cups that make this an equivalent of a
truncated tetrahedron. Even a well meaning constructor of such a
figure using bicycle tubes would not extent the tubes to a sharp point
intersection as a mathematical line model would. Stop splitting
hairs.
> The BB shell, chainstays and rear axle form a trapezium, which is
> deformed by axial loads from the rear wheel.
So? Even a perfect tetrahedron would be deformed. This is a case of
"lies of the second kind" to support a semantic rather than a
mechanical problem. Metals are, after all, an elastic medium. It is
not the truncated tetrahedron that is deforming as you imply but don't
say, but rather the elements of the tetrahedron.
> Bracing this trapezium at some point distant from, but parallel to,
> the BB axis will cause this deforming tendency to be resisted by a
> bending moment in the chainstays, whereas without the bridge the
> entire load must be resisted by the BB/Chainstay junction attempting
> to change it's angle. For this reason, bridgeless designs usually
> rely on reinforcement of this junction, such as gussets, extended BB
> shell or, in the case of lugged steel designs, longer and heavier
> points on the BB shell. Admittedly, the seatstays (also braced, by
> the brake bridge) also resist this load, but anybody who thinks the
> chainstay bridge serves no structural function should have a look at
> the swingarm of the nearest high performance motorcycle.
You might as well take the lugs of a Rene Herse frame as proof of
stresses at certain junctions when in fact much of this is artistic
license or in response to an unrelated frame failure. Similarly, you
could cite tying and soldering spokes at their crossings.
> It is possible to properly design a bicycle without a chainstay
> bridge, and I do not presume to know which approach is ultimately
> "better", by whatever test you happen to mean by better; this will
> vary according to geometry and material choice, but in general if a
> designer has chosen to use a bridge, it's likely to be there for
> structural reasons; it certainly isn't required for tyre restraint
> or mudguard mounting on my track or dirt-jump bikes, both of which
> enjoy the structural benefit of substantial chainstay bridges (and,
> for that matter, "brake" bridges, even though one is braked by the
> sprocket and the other by a disc)
Well that's a great disclaimer after all that dogmatic explanation of
the "rear triangle".
Jobst Brandt
jobst....@stanfordalumni.org
carl...@comcast.net
Brown) wrote:
Dear Phil,
I'm fascinated, but want more details. I always assumed that
the tubing was just straight, uniform pipe, but you're
saying otherwise.
Is the pipe tapered in terms of its wall becoming thinner at
each end, or does the wall stay the same thickness, while
the diameter of the whole tube narrows?
Whatever the answer is, why does it taper? Easier assembly?
Strength?
Carl Fogel
carl...@comcast.net
<us...@domain.com> wrote:
Dear Kinky,
Okay, now I'm thinking of the bottom-bracket, chain-stays,
and axle as a long, two-rung ladder lying flat on the floor,
with a narrow rung at one end and a wide rung at the bottom.
If the narrow end of the ladder is nailed to the floor,
pushing sideways on the ladder's wide end will deform the
whole rickety trapezoidal arrangement.
But a third rung near the narrow end will stiffen the
ladder.
Is this roughly the idea?
Carl Fogel
Phil Brown
>each end, or does the wall stay the same thickness, while
>the diameter of the whole tube narrows?
>
>Whatever the answer is, why does it taper? Easier assembly?
>Strength?
The tube walls are tapered before it's swaged into its final shape.
Phil Brown
jim beam
builders recommendation to use a chain stay bridge to mitigate fatigue.
makes sense when you think about it - this section is not a perfect
triangle - its apex terminates at the most highly stressed and flexed
part of the frame.
jim beam
tapered for weight. the dropout end doesn't get twisted as much as the
bb end, so it doesn't need to be as strong, ergo tapered tube is the way
to go. just like the shaft of a golf club. the handle end has to
withstand the leverage of the user accelerating the club and is
therefore wider - the head end just has to be strong enough to stop the
end flying off so is narrower [and thinner wall too].
so, high end chain stays are tapered /and/ butted to be thinner in the
middle. scroll through this page for examples.
http://dedacciai.com/prodotti/sat.htm
as an aside, it fascinates me that certain "high end" [i.e. _expensive_]
titanium frame builders haven't figured out the value of tapered tubes
yet. i know that tapered tubes are hard to make, especially having the
thin wall at the narrow end, but for the price on some of these frames,
to /not/ have tapered tubes is pretty outrageous.
jim beam
by some tube manufacturers to mitigate fatigue, and i've seen a number
of weld failures at the bb when it's not present. [chain stays can fail
at the bridge too btw, but it's less common from what i've seen.]
A Muzi
> How does not having a chainstay bridge affect the frame? I have a
> "loaded" touring frame without a chainstay bridge. The frame is
> welded Kinesis 7005 Superlight butted aluminum tubing, according to
> the sticker. Chainstays are 17" long. Cantilever brakes will be
> used. 130mm rear OLD spacing.
>
> I know mounting a rear fender will require some ingenuity.
-snip-
That, and one's wheel would stick with normal ends during
removal. Not the problem with verticals.
There are some classic designs which are prone to chainstay
cracks at the cast bridge. . .
--
Andrew Muzi
www.yellowjersey.org
Open every day since 1 April, 1971
carl...@comcast.net
wrote:
Dear Jim,
I think that I follow your nice golf-club analogy.
Now I have to ask how much--how much weight is actually
saved by tapering the various frame tubes? Is this saving a
few grams, a few ounces, or even a pound?
Carl Fogel
A Muzi
> Brown) wrote:
>>>AFAIK, chainstays aren't butted. At least they're not on the Reynolds
>>>531/753 Isiwata, Tange, and Columbus steel tube sets I've worked with in
>>>the past.
>>The walls are tapered on good quality stays.
carl...@comcast.net wrote:
> I'm fascinated, but want more details. I always assumed that
> the tubing was just straight, uniform pipe, but you're
> saying otherwise.
>
> Is the pipe tapered in terms of its wall becoming thinner at
> each end, or does the wall stay the same thickness, while
> the diameter of the whole tube narrows?
-snip-
Cut a cheap fork or seat/chain stay from a bike such as you
recently bought for under $100.
You'll find the shape doesn't lend itself to a cheap
solution, the resulting piece being fairly thin at the fork
crown ( or at the BB in the case of a chainstay) yet
ridiculously thick at the small diameter end. One might
easily envision a pipe being worked to that shape in one pass.
A premium quality material such as Reynolds tube yeilds a
product that's of a uniform wall thickness over a wide range
of outer diameters. Reynolds calls that feature "taper gauge
tubing".
carl...@comcast.net
Dear Andrew,
Sorry, but I have to fill in the blanks that are obvious to
frame-friendly folks like you.
I think that you're saying that a cheap bicycle's
chain-stays will taper visibly, but that the too-thin walls
at the wide-diameter bottom-bracket-end turn into too-thick
walls at the narrow-diameter axle-end.
I'm guessing that this is due to a crude squeezing process
("one-pass") and that the higher-quality Reynolds tubing
tapers to a smaller outside diameter at the axle-end while
its wall stays the same thickness throughout.
Assuming that I'm following you, Is the premium Reynolds
tubing premium because it's stronger, or because it's
lighter, or what?
Carl Fogel
Kinky Cowboy
[immense snip]
>Dear Kinky,
>
>Okay, now I'm thinking of the bottom-bracket, chain-stays,
>and axle as a long, two-rung ladder lying flat on the floor,
>with a narrow rung at one end and a wide rung at the bottom.
>
>If the narrow end of the ladder is nailed to the floor,
>pushing sideways on the ladder's wide end will deform the
>whole rickety trapezoidal arrangement.
>
>But a third rung near the narrow end will stiffen the
>ladder.
>
>Is this roughly the idea?
>
>Carl Fogel
Yep!
Kinky Cowboy
wrote:
>Kinky Cowboy snipes anonymously:
>
>>>> How does not having a chainstay bridge affect the frame? I have a
>>>> "loaded" touring frame without a chainstay bridge.
>
>>> Depending on frame size and geometry, the bridge doesn't even serve
>>> that purpose. It is never structurally necessary, the rear triangle
>>> actually being a tetrahedron, the stiffest solid body available.
>
>> We've been here before. The rear triangle isn't a tetrahedron,
>> because if it was the chainstays would meet in the centre of the BB
>> shell.
>
>The virtual intersection of these tubes is guaranteed by the massive
>BB shell reinforced by BB cups that make this an equivalent of a
>truncated tetrahedron. Even a well meaning constructor of such a
>figure using bicycle tubes would not extent the tubes to a sharp point
>intersection as a mathematical line model would. Stop splitting
>hairs.
But the pin jointed model, while manifestly a gross
oversimplification, does serve to illustrate the likely effect of
adding the extra element, by making it easy to visualise which angles
or lengths need to change to accomodate the deflection
>> The BB shell, chainstays and rear axle form a trapezium, which is
>> deformed by axial loads from the rear wheel.
>
>So? Even a perfect tetrahedron would be deformed. This is a case of
>"lies of the second kind" to support a semantic rather than a
>mechanical problem. Metals are, after all, an elastic medium. It is
>not the truncated tetrahedron that is deforming as you imply but don't
>say, but rather the elements of the tetrahedron.
There is a deformation of the trapezium, in the sense of the angles
changing, as well as deformation of the elements by bending and/or
stretching
>> Bracing this trapezium at some point distant from, but parallel to,
>> the BB axis will cause this deforming tendency to be resisted by a
>> bending moment in the chainstays, whereas without the bridge the
>> entire load must be resisted by the BB/Chainstay junction attempting
>> to change it's angle. For this reason, bridgeless designs usually
>> rely on reinforcement of this junction, such as gussets, extended BB
>> shell or, in the case of lugged steel designs, longer and heavier
>> points on the BB shell. Admittedly, the seatstays (also braced, by
>> the brake bridge) also resist this load, but anybody who thinks the
>> chainstay bridge serves no structural function should have a look at
>> the swingarm of the nearest high performance motorcycle.
>
>You might as well take the lugs of a Rene Herse frame as proof of
>stresses at certain junctions when in fact much of this is artistic
>license or in response to an unrelated frame failure. Similarly, you
>could cite tying and soldering spokes at their crossings.
But I didn't; I carefully cited the similar trapezoidal structure of
the motorcycle swingarm (technically, a swinging fork), which has been
subject to a great deal of investigation as to the best method of
bracing against the large axial and torsional loads fed into it by the
rear wheel, and which also has it's direct bicycle equivalent on many
suspended frame designs.
>> It is possible to properly design a bicycle without a chainstay
>> bridge, and I do not presume to know which approach is ultimately
>> "better", by whatever test you happen to mean by better; this will
>> vary according to geometry and material choice, but in general if a
>> designer has chosen to use a bridge, it's likely to be there for
>> structural reasons; it certainly isn't required for tyre restraint
>> or mudguard mounting on my track or dirt-jump bikes, both of which
>> enjoy the structural benefit of substantial chainstay bridges (and,
>> for that matter, "brake" bridges, even though one is braked by the
>> sprocket and the other by a disc)
>
>Well that's a great disclaimer after all that dogmatic explanation of
>the "rear triangle".
>
>Jobst Brandt
>jobst....@stanfordalumni.org
Why thank you! I suspect you were indulging in the lowest form of wit,
but, ignoring your sarcasm, all bicycle design (as in any other branch
of engineering) is a series of compromises, and there are entirely
legitimate reasons why a bridgeless design might be the best
compromise for one application while a bidged design is equally
appropriate for another. While I'm being pragmatic (a quality you
might like to acquire) I am equally sure that a great many features of
bicycle design are continued because they've always been done that
way, and that as a result there are probably some redundant chainstay
bridges out there.
John Thompson
When did they start doing that? In the past, Reynolds made "taper gauge"
fork blades designed so that when the blades were tapered, you'd end up
with uniform gauge throughout the length of the blade. But the chain
stays were not done in this manner -- after tapering, the wall thickness
at the small end of the stay was substatially thicker than at the large
end. The Ishiwata, Tange, or Columbus tubes sets that I've built frames
from were all the same way with the chain stays.
But I haven't bought any tube sets for some time now, so perhaps things
have changed in the interim.
--
-John (jo...@os2.dhs.org)
John Thompson
> On 27 Jun 2004 01:14:11 GMT, philc...@aol.communged (Phil
> Brown) wrote:
>>
>>The walls are tapered on good quality stays.
> I'm fascinated, but want more details. I always assumed that
> the tubing was just straight, uniform pipe, but you're
> saying otherwise.
>
> Is the pipe tapered in terms of its wall becoming thinner at
> each end, or does the wall stay the same thickness, while
> the diameter of the whole tube narrows?
>
> Whatever the answer is, why does it taper? Easier assembly?
> Strength?
The main tubes on good quality tube sets are butted at the ends, so that
the ends of the tubes have thicker walls than the middle of the tube.
0.9mm/0.6mm/0.9mm is a common pattern for steel road bike frame tubes.
This provides more material at the ends of the tubes where the
brazing/welding is done.
Seat and chain stays are generally produced from straight-gauge (unbutted)
tubing; the tapering process where the outside diameter is reduced on one
end results in greater wall thickness at the small O.D. end.
Fork blades are typically produced from chainstays. A 22mm straight-gauge
tube is tapered to produce a chainstay, then the large O.D. end is
ovalized and the blade raked to produce a fork blade. Sometimes the
blades are not raked until after they have been built into the fork; when
you buy a tube set you can specify whether you want raked or unraked
blades.
Reynolds felt that they could create a better fork blade by using a "taper
gauge" tube to create the fork blade. Instead of using an unbutted 22mm
tube to produce the blade, Reynolds uses a 22mm tube with a gradual change
in wall thickness from one end to the other. The thin end of the tube is
tapered to a smaller O.D. with a corresponding increase in wall thickness
as a result of the tapering process. But the tapering wall thickness
means that the resulting blade will have uniform wall thickness throughout
its length.
Columbus and other tubing manufacturers have also put ribs in various
tubes to provide extra strength in high-stress areas. The steer
tubes on the original Columbus SL/SP and Tange tube sets have helical ribs
inside the butt (fork crown) end. Ishiwata used straight ribs on their
steer tubes. Newer Columbus tubes (SLX) also have helical ribs on the
insides of the main tubes and chainstays where they join the bottom
bracket.
--
-John (jo...@os2.dhs.org)
A Muzi
-snip-
> I think that you're saying that a cheap bicycle's
> chain-stays will taper visibly, but that the too-thin walls
> at the wide-diameter bottom-bracket-end turn into too-thick
> walls at the narrow-diameter axle-end.
>
> I'm guessing that this is due to a crude squeezing process
> ("one-pass") and that the higher-quality Reynolds tubing
> tapers to a smaller outside diameter at the axle-end while
> its wall stays the same thickness throughout.
>
> Assuming that I'm following you, Is the premium Reynolds
> tubing premium because it's stronger, or because it's
> lighter, or what?
Reynolds or other premium tube is :
a possibly stronger or at least a usually more suitable alloy
it's cleaner - less impurities in the steel.
More uniform both materially and mechanically
Reynolds butted main tubes are thicker at the joints, yet
thinner elsewhere.
The for blade's shape - a constant wall thickness - is
costly to make but results in about the right a mount of
flex in a fork blade without being prone to cracks at the
fork crown joint.
I knew I had a pair of these someplace.
http://www.yellowjersey.org/photosfromthepast/FORKSEC.JPG
carl...@comcast.net
Dear Andrew,
Nice pictures!
Both forks seem to be the same thickness at the crown-end,
but the upper fork's tubing is clearly thicker-walled than
the Reynolds fork down near the axle-end.
Thanks,
Carl Fogel
jim beam
i don't expect it's a lot - if you want specifics, you'd have to work
that out on an individual component basis. you could try a math
approximation for a conic but it's hard to make account for non-uniform
wall thickness.
my previous comment reflects a purist viewpoint. sure, a straight tube
may not make a lot of difference in weight, but the priciple is to
maximize material where it's needed, the bb, and minimize it where it's
not, the dropout. ignoring that in a $2k, $3k or $4k frame either
demonstrates a frightening degree of ignorance or astonishing arrogance.
A Muzi
>>>531/753 Isiwata, Tange, and Columbus steel tube sets I've worked with in
>>>the past.
> On 2004-06-27, Phil Brown <philc...@aol.communged> wrote:
>>The walls are tapered on good quality stays.
John Thompson wrote:
> When did they start doing that? In the past, Reynolds made "taper gauge"
> fork blades designed so that when the blades were tapered, you'd end up
> with uniform gauge throughout the length of the blade. But the chain
> stays were not done in this manner -- after tapering, the wall thickness
> at the small end of the stay was substatially thicker than at the large
> end. The Ishiwata, Tange, or Columbus tubes sets that I've built frames
> from were all the same way with the chain stays.
> But I haven't bought any tube sets for some time now, so perhaps things
> have changed in the interim.
The printed specs here from Reynolds, Columbus, Tange and
Ishiwata all say chainstays are delivered with constant wall
thickness end to end. But a quick measurement affirms your
observation:
http://www.yellowjersey.org/photosfromthepast/C_STAYS.JPG
Russell Seaton
> > Blackburn P-clips at the rear dropouts and the seatstay bridge, I
> > figure the fender should be fairly well secured. Wire between the
> > fender and chainstays behind the bottom bracket will probably be
> > enough to hold the fender in place.
>
>
> Jobst Brandt
> jobst....@stanfordalumni.org
The frame has vertical dropouts. So the wheel should not get stuck
between the chainstays because it will not be pushed forward. The
frame has the seatstay bridge for mounting the top of the fender in
the normal manner. And Blackburn P-clips will hold the back of the
fender near the rear dropouts. Near the bottom bracket shell the
fender has a tab/hole to bolt onto the chainstay bridge. I will use
copper electrical wire around the chainstays and into the tab/hole on
the fender to secure the fender behind the bottom bracket shell.
jobst....@stanfordalumni.org
>>> I know mounting a rear fender will require some ingenuity. But
>>> with Blackburn P-clips at the rear dropouts and the seatstay
>>> bridge, I figure the fender should be fairly well secured. Wire
>>> between the fender and chainstays behind the bottom bracket will
>>> probably be enough to hold the fender in place.
>> I am not visualizing what you mean.
> The frame has vertical dropouts. So the wheel should not get stuck
> between the chainstays because it will not be pushed forward. The
> frame has the seatstay bridge for mounting the top of the fender in
> the normal manner. And Blackburn P-clips will hold the back of the
> fender near the rear dropouts. Near the bottom bracket shell the
> fender has a tab/hole to bolt onto the chainstay bridge. I will use
> copper electrical wire around the chainstays and into the tab/hole
> on the fender to secure the fender behind the bottom bracket shell.
I mentioned the wheel getting stuck because that was one of the
benefits of older bicycles that used horizontal dropouts. Because the
tube bridge was usually there historically, it has remained on
bicycles with long chainstays, probably for belief in the arguments
presented here on their effectiveness. A thin tube in that direction
is not effective in shear or bending considering its diameter in
comparison to the tubes it is supposed to assist in such forces.
These fender mounts have been made of split tubes, ones with an open
seam down one side.
I'm sure there is a way of attaching a fender without it but not as
easily.
Jobst Brandt
jobst....@stanfordalumni.org
Benjamin Weiner
> The frame has vertical dropouts. So the wheel should not get stuck
> between the chainstays because it will not be pushed forward. The
> frame has the seatstay bridge for mounting the top of the fender in
> the normal manner. And Blackburn P-clips will hold the back of the
> fender near the rear dropouts. Near the bottom bracket shell the
> fender has a tab/hole to bolt onto the chainstay bridge. I will use
> copper electrical wire around the chainstays and into the tab/hole on
> the fender to secure the fender behind the bottom bracket shell.
Zipties around the chainstays can also work:
http://www.mindspring.com/~d.g1/ziptie.htm
The top left illustration shows a fender with extra holes
drilled in the sides near the chainstays, that can make
life easier than trying to use the original hole in the
center.