# Bicycle Mechanics - The problem of shimmy explained

Bikeforums.net is a forum about nothing but bikes. Our community can help you find information about hard-to-find and localized information like bicycle tours, specialties like where in your area to have your recumbent bike serviced, or what are the best bicycle tires and seats for the activities you use your bike for.

View Full Version : The problem of shimmy explained

jur
07-15-06, 11:43 PM
The time has come to take the problem of shimmy out of the realm of myth and firmly into the world of science.

Background: When I started reading up on these forums a couple of years ago when I started cycling, I was amazed to find that the problem of shimmy (speed wobble, death wobble, harmonic vibration, whatever is it variously called) has not yet been explained in a rigorous mathematical way. Myths and anecdotes abound, but no solid math-based physical explanation is to be found anywhere. At best there are vague explanations involving a lot of hand-waving, but even these are either completely off target or too vague to be of real value.

I believe I have found the physical explanation to the phenomenon of shimmy. It is relatively straightforward to explain shimmy in broad math terms once you know the underlying mechanism; in rigorous details the math quickly becomes unwieldy far beyond its worth. To keep it simple, some simplifications have to be made which are perfectly acceptable but the ability to explain detailed cases mostly goes down the drain. My explanation differs markedly from what you can find on the web in the various FAQs, but I can write down equations or run a simulation that shows the effect.

So, to get forward in this problem I want to embark on a series of experiments to test various hypotheses about the cause of particular instances. For example, there is lots of talk about frame stiffness and about wheel stiffness. I want to put hypotheses like these to the test. I have figured out exactly how to test it. All I now need is time and equipment. If I am to obtain all the equipment myself, then it will obviously take a bit longer; however, if there are bike/frame builders or other interested parties who want to participate, specifically wrt experience and equipment, let me know. I am in Melbourne, Australia. My email address for expressions of interest is juried.noxiousweed@yahoo.com. Just kill the noxious weed. I will provide a more direct email address to interested parties via that portal. If you know of potential interested parties that do not frequent these forums, please feel free to pass the word.

To recap, I already have the explanation down in broad math terms; I need to back that up with experiments for 2 reasons: One, to validate the details of explanation, and Two, to gain enough detailed understanding to provide answers to the question, "How can I cure my bike of shimmy?"

I am not associated with any academical insitution nor employed in the bicycle industry. I am just a curious individual who by happenstance encountered enough related details in my various jobs to realise what the explanation for shimmy is, and to have enough ability to write down that explanation in mathematical terms.

This will be a fairly long term project. One of the end results may be a technical paper, perhaps with a co-author if there are interested parties like graduate students. Hopefully the most important end result will be a widespread detailed understanding of the problem together with specific approaches of how to fix problems.

What I need off these forums are anecdotes of people who have encountered shimmy and have found various ways to counter it or who have performed repeatable experiments. One good example is by DannoXYZ who gives definite data about the effect of different wheels here (http://69.16.211.161/showthread.php?t=190787&page=2&#39). All these can be brought into the comprehensive explanation. What I don't need are posts about second-hand cases like "my friend's cousin's cat's fleas once had shimmy, very scary", because I may want to pose questions to gain further details, and first-hand info is more reliable.

Specifically there are two different cases of shimmy, one with riding hands OFF the handle bars and the other with hands ON the bars. It is the second case which I am focussing on, the first case I already understand in fairly fine detail. However, if you have anecdotes about case #1 in which you had repeatable experimental evidence on how you eliminated it by changing parts on the bike, please tell.

07-16-06, 01:33 AM
i only experienced shimmy one time and it was on the whippiest frame that i have ever owned, a specialized epic carbon.

i had just crested brockway summit from the truckee side on about mile 90 of the 110 mile lake tahoe century and i was attempting to hit 50 m.p.h. when at about 48 m.p.h. my frame started shimmying and i thought i would die.

scared the hell out of me and i am convinced it was the whippy frame. the fork was a kestrel and i have used an EMS fork on my last half dozen road bikes with no problems.

soon after i dumped the epic and bought a kestrel 200 SCi. in retrospect the epic was the worst road bike that i have owned.

Carusoswi
07-16-06, 06:32 AM
Five years ago, my only bike was a Schwinn Le Tour (new in 1973). At the same spot on the descent down a hill I rode daily, I experienced this sort of shimmy. Normally, having crested the climb side of this hill, my speed going into the descent is down around 7 mph. As I began to pick up speed, this shimmying would come on gradually increasing in intensity until violent enough that, as the previous poster mentioned, I thought a crash would be my only way out of the situation. I applied the brakes lightly, increasing brake pressure until I could bring the bike to a stop. Its shimmying continued as the bike slowed and did not stop until the bike and I were nearly motionless.

This happened to me a half dozen times, only on that one hill.

I cannot correlate these experiences with wind direction or velocity, and there is nothing unique about the finish or condition of that road.

The bike had been meticulously maintained, so there is no particular mechanical problem to which I would attribute this shimmying.

I was totally stumped by these experiences (totally terrified by them as well).

I would be very interested in learning the results of your study. Good luck, and I hope this little story is useful to you.

Caruso

Stacey
07-16-06, 06:56 AM
From my time spent with a motorcycle road racing program, I've found there are many causes for front wheel shimmy, the major one being steering geometry, closely followed by front to rear weight biasing. Flex, be it frame or fork is another factor. Also something as simple as tire model design... Yes, we had one bike that would develop a horendous headshake at speed, with all other factors examined/eliminated we changed the make and model of front tire used on the bike (On the same cast alloy wheel) and amazingly the tank-slapping proclivity disapeared.

Treefox
07-16-06, 07:42 AM
What's shimmy??

Sorry if this is a terribly basic question, but I've never heard of it.

Stacey
07-16-06, 07:54 AM
It's a side to side oscillation in the front wheel and handlebars that usually occurs at a higher rate of speed. When it gets bad enough it's almost impossible to recover from and usually terminates in a painful and unexpected dismount from the bike. On motorcycles, when it gets bad enough, they call it a "tank slapper"

Watch this video (http://www.flyingsnail.com/Sprung/tankslapper.html) for an example

Carusoswi
07-16-06, 08:04 AM
Wow, Stacey. I wonder how that rider made out - it looked like a terrible crash.
On my Schwinn, the vibration was much faster, but the "amplitude" for lack of a better term was not as high - and my path, while sloped, was straight (fortunate for me).
Caruso

Stacey
07-16-06, 08:08 AM
Probably faired pretty well, those guys dress for the crash.

Reason being is you have much less mass than a MC front end to oscillate, hence the higher frequency and lower amplitude. You were able to ride it out, I hope.

San Rensho
07-16-06, 04:05 PM
It's a side to side oscillation in the front wheel and handlebars that usually occurs at a higher rate of speed. When it gets bad enough it's almost impossible to recover from and usually terminates in a painful and unexpected dismount from the bike. On motorcycles, when it gets bad enough, they call it a "tank slapper"

Watch this video (http://www.flyingsnail.com/Sprung/tankslapper.html) for an example

Have you noticed that when a guy gets into a horrible tank slapper and falls off the bike before the bike falls, the bike just straightens out perfectly and keeps going as if it had a ghost rider?

I wish I had the video clips, but I remeber seing it once at Daytona, and anothe time with a guy trying to set a speed record at Bonneville. He was going over 150 mph and the bike starts shaking like a wet dog. He bails, and the bike immediatley stops shaking and just keeps going and going because the throttle stayed pinned.

Bicycles/motorcycles are self correcting and very stable, its usually rider inputs that start or worsen headshake. When I raced motorcycles, one of the things I trained myself to do was have a really light touch on the bars, just let them go where they want and they usually get more stable. Easy to say, hard to do.

Stacey
07-16-06, 05:12 PM
^^^ Very good point. Good training for that ON A MOTORCYCLE is to practice that on steel grate bridges. On bicycles, that can be disastrous as evidenced in the 89 Tour De Trump.

sfrider
07-16-06, 07:03 PM
I once had it happen when I was a kid. I rode passenger with a friend and we went down a steep hill (about 20% or so). We both went airborne and he got whacked in the face by the handlebars and the curb, and got at least a titanium cheekbone. He looked like hell for months afterwards. Me, I was just a little scratched up. Never had a tank slapper on a MC, or shimmying on a bicycle other than this one incident.

I always assumed it's caused by a warped frame with more than one point of stability. As it's loaded sufficiently, it starts oscillating between these two stability points. It could also simply be overloaded and flex itself into a warp with speed or a downhill angle, where there's more load on the front wheel. Or it could be approaching the end of its service life and fatigue may have weakened it a bit. But that's just a hypothesis.

By the way, skateboards may wobble, too.

jur
07-16-06, 07:31 PM
Stacey: Thanks for that link. The video confirms 2 of my hypotheses directly.

Stacey
07-16-06, 07:33 PM
You're welcome. Care to share?

Sporkinum
07-16-06, 08:00 PM
2 times.. Once when I was a kid, I wanted to see how fast I could go down a hill into town on my Pea Picker. It had one of those cable driven speedos that was way off due to the 16in front tire. Anyway, it said I was going 60 when I laid it down, But I would gues the reality was maybe 35 or 40. Some guy that raced motorcycles saw it happening, knew what was going to happen, stopped, and took me to the hospital. Had severe road rash.

The other time was 30 year later and I was riding across the US. Bike this time was a Schwinn Passage. Was going down a mountain in Oregon got up somewhere over 45 and it started oscillating. I was in a full tuck and tried the brakes. That seemed to make it worse, so I slowly sat up and when the bike slowed the wobbles went away. Only thing odd on the Passage was a flexstem.

jur
07-17-06, 10:22 PM
You're welcome. Care to share?
Of course. See http://en.wikipedia.org/wiki/Speed_wobble

Stacey
07-18-06, 06:44 AM
Excellent piece. Thank you.

rmfnla
07-18-06, 08:45 AM
I think one contributing factor is the amount of mass in the handlebar/stem area and that mass's position in relation to the front axle.

I've noticed changes in shimmy from simple alterations such as stem length or angle, and not always in ways I might have predicted.

San Rensho
07-18-06, 08:54 AM
Of course. See http://en.wikipedia.org/wiki/Speed_wobble

Cool, a physics and mathematical explanation for the point I was getting at before, death grip=death wobble.

rmfnla
07-20-06, 09:29 PM
Cool, a physics and mathematical explanation for the point I was getting at before, death grip=death wobble.

I have one bike that shimmys when I ride no-hands; explain that, Mr. Science.

(I used to have a 3Rensho; shouldn't have let that one get away...)

San Rensho
07-21-06, 08:48 AM
I have one bike that shimmys when I ride no-hands; explain that, Mr. Science.

(I used to have a 3Rensho; shouldn't have let that one get away...)

The article explains it. There is shake from no handed riding and shake from death grip, but a light touch on the bars and no shakes. If you have no handed shake, you might want to go through and see if everything is tight on the entire bike, not just the headset and front wheel bearings. Sometimes something loose in the back will make the front shake.

MotoProf
04-02-07, 04:34 PM
This happened to me a half dozen times, only on that one hill.
I cannot correlate these experiences with wind direction or velocity, and there is nothing unique about the finish or condition of that road. Caruso

There might be some psychology involved here. Since rider grip/position is a variable, it is possible that once you have a bad experience you are more anxious on that hill, thus more rigid, tighter grip, etc. which can create the very situation that caused oscillation the first time, creating a "self-fulfilling prophecy".

I experienced the dreaded "death wobble" twice; 20 years apart, on two very different frames (high end steel, CF). In both cases I was already nervous when the shimmy started and in the last one, I was also cold/stiff. It would be interesting to see how often this is a factor.

BTW. re. the recent incident: I was on a long ride w/lots of hills/decents. My LBS guy, who is also a much better rider than me, was on the ride. He had worked on my bike and assured me it was me, not the bike. I reluctantly forced myself to hit a similar speed (36-38/mph) on the next decent with imporved position (farther back, knee touching top tube, light on the saddle, & loose grip on the bars) so that I wouldn't be afraid to ride again. Worked fine. No wobble. No death.

rmikkelsen
04-02-07, 04:42 PM
I noticed a bit of new shimmy on fast downhills after switching to a 120 stem from 90mm (my top tube is a bit short for me).

atomsuite
04-02-07, 04:49 PM
I've never experienced it on a bike, but I have on skateboards back when I was a kid.

I drive a manual transmission car with a real high (rpm) first gear. Usually in stop and go traffic I'm in 2nd gear but sometimes it's bad enough that I stay in first. When in first, I barely hover my foot over the gas pedal to give it a super steady amount of pedal pressure. My foot remains very steady and sure. I should mention I also have bumpy (custom sport) suspension. When hovering over the gas pedal in this way going real slow, it takes only the slightest bump in the road for my foot in relation to the pedal to change which causes the car to jump a bit. That jump (even when I purposefully don't move my foot to compensate) will continue as the car adjusts for the slight change in throttle and when it winds back down again it gets repeated because now the car slows and my stable foot hits the pedal again.

I think this is a very similar phenomenon. By the way, the only good way for me to break the shimmy is to give it more gas or use the clutch. I've learned that trying to control the shimmy by modulating the pedal directly is pointless and makes it worse.

Mel_A
04-02-07, 07:07 PM
I am 6'4" tall, was 220 at the time of the first wobble, 2 years ago. Trek 1800c, got going so fast that it went into the wobble. Thought I was going to die! Think it was about 42MPH. Everyone said the bike was to small for me. Had a Seven ID8 Ti and carbon custom built for my size, reach and stiffness. Love the bike. After 6 months of riding it, I started to descend from Timberline Lodge on Mt Hood, Oregon. Had only travelled about 1/4 mile when the wobbles took over. Headed straight for a small U-Hual coming up in a line of traffic. I could not turn, I could barely think, in fact I could not think. I am looking into the eyes of the drivers and they are enormous. Remembered the previous wobble and the antidote that I researched. Pinched my top tube with my knees as hard as I could and slowly, ever so slowly I was able to steer myself away from the traffic. Although I really didn't have time to look at speedometer, I think I was only going about 35.

xB_Nutt
04-03-07, 09:51 AM
While I have never experienced it on a bicycle, I did survive an 85 MPH+ speed wobble induced crash on a 1981 Yamaha 400 Special motorcycle in my younger days. I usually was able to keep it under control, but this particlar time, I was going down a hill and had to bail and roll to get out of it. It was at night and a friend of mine was coming the other way and said he saw my headlight flailing from side to side and knew I was in trouble. Thank goodness for full face helmets and leather jackets. The dealership where I had just bought the bike new said there was nothing they could do about it since I was exceeding the legal limit, which was 55 MPH then. I thought that was pretty lame as I could push my parents Chevette up over 85 and it was smooth as silk. So I sold the bike and have never ridden one again...

humble_biker
04-03-07, 02:36 PM
Here you go; A fairly long 50mph descent. In a full "Lemond style" tuck, IE nose and chin out over the front wheel. And the wooble starts. I had done this plenty of times before so why now? Time to get back in the saddle and slow down down I say to myself. So I do the shimmy goes away. I put on the brakes to slow down more and the shimmy comes back. I do this to the bottom and am scared ****less :eek: at that point. I coast to a stop flip my bike over and check to make sure axles are firmly seated in the drop outs, front and back. I hung it up for about a month. Eventually I start to ride again and the bike begins to shimmy around 30 or 35mph. instead of the 50+ aero. I eventually figured out the exact cause of the shimmy mechanically on this particular bike and learned alot about position in the following years. :D

byte_speed
04-03-07, 08:50 PM
My current bike (Litespeed Classic) would wobble sometimes when I first got it. It scared me pretty badly, but never got bad enough to cause a fall. After I learned to relax and forgo the death grip on the bars, the wobble disappeared. The light touch is the secret for me.