shimmy on downhill at speed

A Muzi writes:

I say "interesting phenomenon" because I am a medium 56cm kind of
guy. Shimmy is usually associated with straight light big bikes (whose
headsets are adjusted) . If I were a 6'3" shimmy-plagued rider I
wouldn't be so flippant and I would call it the most important issue
of the moment.

Oddly enough I am a 6'3" guy and- thankfully- I am not "shimmy
plagued." I'm not sure I can tell you why.

I ride a 61 cm Ritchey Road Classic which will shimmy hands-off at
about 22 mph but has never shimmied hands-on. It also tends to shimmy
if coasting hands-off but rarely if pedaling hands-off. I have a 60
cm Rivendell All-Rounder which isn't prone to shimmy (26 inch wheels
with skinny high pressure road tires). My 60 cm Gunnar Crosshairs
developed a shimmy when I mounted a small Nitto rack on the fork for
doing brevets; taking it off eliminated the shimmy. My commuter/track
bike, a 60 cm Windsor Profesional, doesn't seem to shimmy ever. My
winter beater, a 63 cm Katakura Silk Road, doesn't seem to shimmy but
then the frame builder appears to have burgled a plumber to get
materials.

OTOH a friend of mine who rides a 56 and is about 145 lbs had
tremendous trouble with a DeRosa aluminum bike shimmying like mad.
Replacing the front tire, for some reason, fixed it.

Now, I've noticed that if I lean back while riding no-handed, the
bikes seem more likely to shimmy. That to me suggests that shimmy can
result when damping is less than adequate. This leads me to some further
questions-

1. would modern frames with short chainstays and long top tubes,
resulting in a longer front center and proportionally more weight on
the rear wheel and less on the front, be more likely to shimmy?

2. Are shallow head tube/large fork rake (or offset, if you prefer)
geometries more likely to shimmy?

3. And finally, are skinny high pressure tires more likely to shimmy?

Tom Nakashima writes:

Sorry, but I don't see the analogy between the shopping cart and a
bicycle? The shopping cart has 4 wheels, while the bicycle has
two and if you watch a shopping cart being pushed around the
supermarket, it touches on only 3 wheels, usually the two rear and
one in the front, while the other front wheel shimmys, hands or no
hands.

Shopping carts with all four wheels on the ground can shimmy
excellently as did cars before ball joint front suspension. Solid
wheels can shimmy just as well as Spinergy wheels for instance.
The point is that FEAR is not the cause of shimmy. As I mentioned
on this topic, descending in cold weather with low energy, avoiding
shimmy on the stablest of bicycles is difficult.

When does a no suspension shopping cart touch on all fours?

They are designed that way and tweaked to do so when new. They only
tilt when crashed and then shimmy only when empty if the damage is
large. Under load most carts ride on four wheels, but that's not the
problem. As I said, automobiles before the days of ball joint front
suspension often shimmied, wheels fully loaded.

Also wouldn't fear relate to how one holds the handlebars on a
descent? I tried the no hands descent on Sand Hill Rd., either my
bike is stable and or I have absolute no fear of riding down hill
with no hands, but I couldn't get the my long wheelbase Reynolds 531
frame bike to shimmy.

As I pointed out, before I suggested you do that test, natural human
muscular response operates roughly in the shimmy frequency, so a
prefect match will sustain or enhance a shimmy. It does this without
fear. I have had to fight shimmy on cold descents after a long day's
ride by loosely holding the bars. As soon as I didn't need to use
brakes I rode no-hands to avoid the problem.

However I have gotten my bike to shimmy by straight arming the
handlebars and rocking it back and forth on a descent.

That seems reasonable.

Jobst Brandt

David L. Johnson wrote:
On Mon, 01 Mar 2004 05:37:02 +0000, jim beam wrote:


agreed, it's not a simple problem and different solutions seem to work
for different situations.

my personal opinion is that the two biggest variables are wheels &
frames, though i'm sure other factors contribute. my personal
experience has been that a rear wheel built with straight gauge 2.0mm
[comparativel inelastic] drive side spokes on a shimano hub effectively
/just/ killed it on a very shimmy-prone frame, but replacing the frame
with one built with oversize tube was much more effective and i could go
back to riding my butted spoke wheels. ymmv.


The last sentence (well, acronym thereof) is the most reasonable. Doing
_anything_ will change the system, and because of the complex nature of
the interactions, it is likely to improve things, since it will throw off
the resonant frequencies.


i will also add that my fixie, with an undished rear, butted spokes, has
proven impossible to shimmy, even with significant encouragement.


But not, I wager, too many 40+ mph downhills, which is where most shimmies
that scare people occur.

agreed, but all the bikes i've had that have shimmied, can also be
encouraged to do so at much lower speeds, hands off. an harmonic
effect. my fixie, a bianchi pista, will not. it is by far the most
stable and clumsiness-tolerant machine i've ever ridden, hands on or off.


imo,

that is a significant indicator of the effect wheel build can have.
the op mentions they're using a shimano hub - they have the best
[least bad] dish geometry available afaik.


The difference in dish between a shimano 9-speed hub and a Campagnolo one
is very small.

the difference in tension ratio between left/right spokes, according to
my calculations, are ~48% for a campy record hub and ~57% for dura-ace.
that small improvement in tension differential seemed to be sufficient
to tune the resonance between frame and wheels sufficiently apart to
help reduce an harmonic. that was my logic when i "defected" to shimano
on my shimmy-prone frame. and it worked.

confession: i did /not/ perform a controlled experiment because i also
changed the spokes for plain gauge 2.0mm as well as changing to the
shimano hub, but the rim and tire & tire pressure, as well as the rest
of the bike, were unchanged, so either the change in hub spacing, the
change in spoke gauge, or a combination of both was sufficient to
beneficially affect this bikes shimmy behavior.

On Tue, 02 Mar 2004 02:37:39 +0000, jim beam wrote:


agreed, but all the bikes i've had that have shimmied, can also be
encouraged to do so at much lower speeds, hands off.

Usually the case, but also possible on a bike that doesn't shimmy on
a downhill, and also possibly a bike that will not shimmy no-hands might
show a shimmy on a fast downhill. My track bike shimmied once, when I*was
desperately trying to sprint. It was as much due to my death-grip on the
bars as anything else.

that small improvement in tension differential seemed to be sufficient
to tune the resonance between frame and wheels sufficiently apart to
help reduce an harmonic. that was my logic when i "defected" to shimano
on my shimmy-prone frame. and it worked.

confession: i did /not/ perform a controlled experiment because i also
changed the spokes for plain gauge 2.0mm as well as changing to the
shimano hub, but the rim and tire & tire pressure, as well as the rest
of the bike, were unchanged, so either the change in hub spacing, the
change in spoke gauge, or a combination of both was sufficient to
beneficially affect this bikes shimmy behavior.

I want to emphasize that it is impossible to determine what the "cause" of
the shimmy is, just by the effect of replacing/changing one part. By this
I mean that the part replaced might be functioning well within design
specs -- nothing "wrong" about it, but still changing that thing could
effect the system enough to change the resonant frequencies. The effect
is nearly chaotic -- a very small change in parameters can have a huge
difference in when the system will resonate.

Change whatever you don't like, and it will have a fair chance of ending
your shimmy (or cause it if you don't have it yet...). But still, shimmy
is not a problem; unloading the saddle will always stop it, though people
are often afraid to try that in the midst of an episode.

--

David L. Johnson

__o | We have a record of conquest, colonization and expansion
_`\(,_ | unequaled by any people in the Nineteenth Century. We are not to
(_)/ (_) | be curbed now. --Henry Cabot Lodge, 1895

Jobst:

Quick summary, although it's been a few years so I may have forgotten
some detail: the bike could shimmy BADLY at most any speed above 13mph,
the shimmy was dependent on road conditions & wind conditions, shimmied
no handed & handed. It was a 57cm frame made with Serotta's OS
dimensioned tubing.

After a couple years of fighting the bike & spending lots of time trying
to understand the problem based on my interpretation of the
input/response of the bike, I decided that the front triangle of this
bike was not stiff enough for the size frame. Why? Because the bike
would start shimmying after hitting a crack/stone/bump that moved the
front wheel out of the original plane of travel. Side winds made the
bike totally unstable, and "pebbly" pavement would cause the front end
of the bike to wander horribly. As a test, I could ride the bike down
the road, shake the handlebars and have the toptube bend either
direction minimum or 2", easily 3-4+" when I really forced the issue. I
convinced Serotta that the bike was defective, explained the tests that
I had done, explained what I wanted in a new frame and why, and they
built the frame that way. I asked them to make the front triangle out
of their "Oversize" OS dimensioned tubing that was usually reserved for
frame sizes 58cm and larger. This is the main change that was done.
Two other changes I asked for were for reasons generally unrelated to
the shimmy but may have made a difference - I'm not sure. The head
angle of the new frame is 73degrees instead of 73.5degrees, and the
toptube is 1/2cm shorter than their standard frame.

So this is a brief history for you. The bike is a much better bike now
with the changes & I have had no shimmying with the new frame at any
speed. All the components were transferred directly from the old frame
to the new frame. I'm still curiuos what Serotta did to their recent
frames as the original person had said. BTW I personally know one other
person with a Serotta TI with shimmying problems just like mine.

MOO,
Matt

wrote:
Matt Locker writes:


I would love to hear more about the Serotta issue as I am a person
who got a replacement frame from Serotta to fix a shimmy. They
made a few changes to my new frame at my request and it is now a
far superior bike to the original. I'm curious what their
"official" changes were.


How about revealing your secrets. What was changed and how and
when did the previous bicycle shimmy.


This was discussed here a few years ago, I believe. I'm just
curious what v... knows about the changes.


So just the same, for those of us with faint memories, could you
recall what was changed on your bicycle? I've seen so much of this
that I don't recall what you wrote "a few years ago".

Jobst Brandt

..

Change whatever you don't like, and it will have a fair chance of ending
your shimmy (or cause it if you don't have it yet...). But still, shimmy
is not a problem; unloading the saddle will always stop it, though people
are often afraid to try that in the midst of an episode.

I can't even count how many times I've had my ass in the air, and my
knees locked on the top tube. To me, shimmys are just part of riding.

David L. Johnson wrote:
On Tue, 02 Mar 2004 02:37:39 +0000, jim beam wrote:


agreed, but all the bikes i've had that have shimmied, can also be
encouraged to do so at much lower speeds, hands off.


Usually the case, but also possible on a bike that doesn't shimmy on
a downhill, and also possibly a bike that will not shimmy no-hands might
show a shimmy on a fast downhill. My track bike shimmied once, when I was
desperately trying to sprint. It was as much due to my death-grip on the
bars as anything else.


that small improvement in tension differential seemed to be sufficient
to tune the resonance between frame and wheels sufficiently apart to
help reduce an harmonic. that was my logic when i "defected" to shimano
on my shimmy-prone frame. and it worked.

confession: i did /not/ perform a controlled experiment because i also
changed the spokes for plain gauge 2.0mm as well as changing to the
shimano hub, but the rim and tire & tire pressure, as well as the rest
of the bike, were unchanged, so either the change in hub spacing, the
change in spoke gauge, or a combination of both was sufficient to
beneficially affect this bikes shimmy behavior.


I want to emphasize that it is impossible to determine what the "cause" of
the shimmy is, just by the effect of replacing/changing one part. By this
I mean that the part replaced might be functioning well within design
specs -- nothing "wrong" about it, but still changing that thing could
effect the system enough to change the resonant frequencies. The effect
is nearly chaotic -- a very small change in parameters can have a huge
difference in when the system will resonate.

where's the engineering logic in this? if you're considering a problem,
consider the contributory elements! flange spacing and spoke elasticity
are contributors! they may not be the whole story - frame torsion being
another obvious avenue for consideration, but at least try to get to the
root of the problem! you seem to be saying "it could be magic" which
gets neither of us anywhere.


Change whatever you don't like, and it will have a fair chance of ending
your shimmy (or cause it if you don't have it yet...). But still, shimmy
is not a problem; unloading the saddle will always stop it, though people
are often afraid to try that in the midst of an episode.


or they may not have time if the onset is large and sudden enough. we
don't agree on this! my personal experience is that i got thrown by a
shimmying bike a few years ago. i sustained serious hip injuries that
kept me from riding for a full year. if that's not a problem, i don't
know what is. shimmy is serious dangerous stuff. different people
experience different levels of it and tend to trivialize something if
their personal experience is not as bad as the other guy.

On Wed, 03 Mar 2004 16:53:12 +0000, jim beam wrote:

I want to emphasize that it is impossible to determine what the "cause" of
the shimmy is, just by the effect of replacing/changing one part. By this
I mean that the part replaced might be functioning well within design
specs -- nothing "wrong" about it, but still changing that thing could
effect the system enough to change the resonant frequencies. The effect
is nearly chaotic -- a very small change in parameters can have a huge
difference in when the system will resonate.

where's the engineering logic in this? if you're considering a problem,
consider the contributory elements!

Every element of the system is contributory, and changing any one will
change the system. That does not mean that any one element is the "cause"
of the problem.

flange spacing and spoke elasticity
are contributors!

So is saddle mass, tire color....

they may not be the whole story - frame torsion being
another obvious avenue for consideration, but at least try to get to the
root of the problem! you seem to be saying "it could be magic" which
gets neither of us anywhere.

People tend to assign causality, understandably, if they change something
and after that the problem is fixed. That does not mean that the thing
they replaced is causing the problem, though, in a complex system.

or they may not have time if the onset is large and sudden enough. we
don't agree on this! my personal experience is that i got thrown by a
shimmying bike a few years ago. i sustained serious hip injuries that
kept me from riding for a full year.

I'm sorry to hear you were hurt. This is the first I had heard of someone
actually having an accident from shimmy. Often people talk about
"almost" falling, but rarely more than that.


if that's not a problem, i don't
know what is.

I am not suggesting that shimmy does not exist. I have experienced it
myself. Most riders who downhill fast, with a larger
probability for taller riders, experience some shimmy in some conditions.
Some bike&rider combinations are more likely than others, but there is a
huge element of chance to the totality of conditions that produce it.

There are definite engineering solutions that could eliminate nearly all
shimmy. The one requirement is that damping forces in the direction of
motion are small. Since this motion is oscillatory rotation around the
center of mass (the rider) what you would need would be a rather large
damper (shock absorber) cantilevered out on one side, or both, of the head
tube, with reinforcing anti-sway bars attached to the rear axle.

The alternative to such a device would be to unload the saddle when shimmy
begins. It will stop. Alternately, press with one leg against the top
tube (providing that damper, by the way).

--

David L. Johnson

__o | To announce that there must be no criticism of the president, or
_`\(,_ | that we are to stand by the president right or wrong, is not only
(_)/ (_) | unpatriotic and servile, but is morally treasonable to the
American public. --Theodore Roosevelt

or they may not have time if the onset is large and sudden enough. we
don't agree on this! my personal experience is that i got thrown by a
shimmying bike a few years ago. i sustained serious hip injuries that
kept me from riding for a full year.

I'm sorry to hear you were hurt. This is the first I had heard of someone
actually having an accident from shimmy. Often people talk about
"almost" falling, but rarely more than that.

I know a woman who went down from shimmy on the Bicycle Tour of
Colorado back in the late 1990s. I believe that ended her tour. She
was riding a Rodriguez bike, 650C wheels. She is small. And had only
been riding seriously for less than a year. The most contributing
factor of all for going down. She was not aware of the shimmy
phenomenon and was not aware of the simple solution of clamping the
top tube between your legs. Or getting out of the saddle. Or loosely
holding the handlebar. Or going slower.

Shopping Cart vs. Bicycle for downhill shimmy competition.

http://www.antiochshrine.com/parades...rtDCP00100.jpg

-tom