Shimmy -- caused by play in hub?

[destry]

Does anyone know whether the dreaded shimmy -- that is, when a bike goes into a sort of controlled wobble at high speed -- can be caused by having too much play in the rear hub?

It happened to me for the first time a week ago, on a frame I've ridden for several years without incident. Bike was unloaded, speed about 35 mph. I had noticed recently that there is more play in the rear hub than there used to be and wondered if this could be behind it.

Speculations and anecdotes welcome!

[cycle cat]

Service the hub and see if it goes away.
I don't really recall shimmy ever being caused by a loose hub.
I may be wrong though.

I've never experienced it but I ride rather over built bikes.
I would have said it's caused by the frame flexing somewhere.
Brucey will be along in a minute!

[destry]

I have fixed the hub.,. But the shimmy only happened once, on quite a rough bit of road. so if it doesn't happen again, that might not mean anything...

Of course, what I really wnat is to feel that I can reasonably blame the wheel (easy to fix) rather than the frame (impossible to fix). I'm never going to know for sure, of course -- it's a matter of feeling confident in one's kit as one sets off down that lovely long hill!

The frame is a 64cm titanium Lemond, by the way, and a very sweet ride indeed.

[Brucey]

fundamentally most bikes will oscillate quite happily in a shimmy, given a suitable opportunity.

Stuff that helps stop a shimmy includes;

-being in the wrong speed range for your shimmy
-damping (this is mostly you sat on the bike and holding the bars)
- not having other 'sprung masses' on the bike (like weight on floppy pannier racks)
- having a stiff frame and wheels
- having wheels/tyres that are perfectly true and and balanced
-riding on nice smooth roads

stuff that provokes a shimmy is the converse of the above. Anything that shakes the bike at a regular frequency that is speed related can potentially 'drive' a shimmy.

Loose hub bearings will allow the bike to flap about in a way that it wouldn't normally be able to, so I don't think it will help; however I suspect there will be something else driving it as well, so it might be a good idea to check everything carefully.

The worst shimmy will happen on any given bike/load at a certain speed. If you know what this speed is, you can be alert for it. To assess a bike for shimmy, I usually ride no hands down a gentle hill where I will accelerate slowly from 15mph up to about 25mph; most bikes will start to shimmy somewhere in this range. If all is well, just grasping the handlebars will quell the shimmy, but if it is being driven very hard you may struggle to control the bike at all.

Large steel frames in thin walled tubing can shimmy more than others, presumably because they are relatively flexible. Some folk report that the use of a roller bearing headset acts as a steering damper, and that this will sometimes quell a shimmy-prone frameset.

People who have had a shimmy on a bicycle often find it alarming. Having had a tanksalapping shimmy at about 90mph on a motorbike, anything I've felt on a bicycle doesn't seem too bad by comparison.

cheers

[teh]

Having feared for my life several times in the past due to "violent death wobble", I'm always interested to read tales of "shimmy". Invariably, it occurs to tall people, going down hill, over rough roads, picking up speed.

If you want to avoid shimmy, then the last bike you want to be on is a "64cm titanium Lemond". Let me guess - 74degree head angle, 4.5cm fork rake, just like every other bike "designed" for a tall person by a short person.

My recipe for shimmy avoidance is: slack head-tube angle, low fork rake, short stem (i.e. long frame), stiff frame and forks. If you can fit cranks in proportion to your leg length, do so. You will sit lower, which will help a lot.

Shimmy is a design flaw.

[[XAP]Bob]

Ive seen a short person get thrown by a bike shimmying into a speed wobble. They spent the next few hours in hospital (rather than on an expedition).

Rough roads are a great way to induce oscillations in all sorts of things....

[Brucey]

it is my understanding that any bike can shimmy under the right (wrong) circumstances; it is almost a necessary consequence of having a castor attached to, well, anything. But stuff on the list (and maybe some other things I forgot) all makes a difference to how bad/how easily provoked a shimmy is. Incidentally a true shimmy like this will occur with no intervention from the rider, and even on a good bike will happen at the critical speed when riding 'no-hands'. Shimmies of this sort typically shake the bike about three or four times a second.

Also note that there is a sub-class of 'death weave' which can precede an accident, which I'm not sure is so much a classic shimmy as a 'feedback problem'. Essentially the rider seems to be trying to control the bike through the bars, but the forces they apply appear to become 180 degrees out of phase with the thing they are trying to control. These motions are typically slower, maybe one or two cycles per second, and larger amplitude, having the bars turn about twenty degrees or more. They typically don't go more than a few cycles before there is an accident.

cheers

[teh]

My wife tells me she experienced shimmy on her Raleigh Budgie, going down Overnorton Hill when she was a child. So, yes you can get hands-on-speed-wobble on any bike, but when it's just adult rider and bike (i.e. no luggage), it's a tall rider phenomenon.

The first time I had hands-on-death wobble, I thought the forks were going to snap. There is certainly rider involvement, but I'm not sure feedback is the right word. You have no option but to hold on, braking makes matters worse, and anyone who thinks touching the top tube with your knee makes any difference, has never experienced it.

[destry]

Thanks for all your thoughts. It looks to me like I am lucky to have cycled for so long before experiencing a shimmy! I'm treating it as a rite of passage.

Interesting thoughts about frame design, though the bike you describe is going to have very long top and down tubes, hence less stiff (all else being equal) and with a very long wheelbase, which has its own problems... What's the theory here, though? Low trail equals less chance of a shimmy? Why so?

I do have another frame which is 27 inches and has a short bracing tube between the seat and down tubes, 3 or 4 inches above the bb. Unfortunately, it isn't nearly as comfortable or fun to ride as the Lemond.

[531colin]

This is very interesting.
I'm 5' 10" and not much more than 11 stone, so not a prime candidate for shimmy. (as "cycle cat" said up thread, I'm basically riding bikes that are overbuilt for my weight....mostly not deliberately, thats just how they get made)
I bought one bike new in about 1965 that shimmied when I had my hands on the bars, always freewheeling downhill, the bars oscillated at about the rate "Brucey" mentions.......

..... Shimmies of this sort typically shake the bike about three or four times a second...........

..(I'm not much good at estimating stuff like that, but clapping my hands fifteen or twenty times in 5 seconds felt about right.)
Just shifting my weight back in the saddle stopped it immediately.....I set the saddle back a bit and rode that bike for about forty more years with no return of the shimmy.
I also get a shimmy, about the same frequency, riding no-hands with a bar bag, putting a hand to the bars stops it straight away.

But I was startled to read;

............. To assess a bike for shimmy, I usually ride no hands down a gentle hill where I will accelerate slowly from 15mph up to about 25mph; most bikes will start to shimmy somewhere in this range. If all is well, just grasping the handlebars will quell the shimmy.........................

Can I just get that straight in my head? Most bikes (or most racing bikes?) will shimmy somewhere between 15 and 25 mph riding no-hands....does that have to be freewheeling? (ie is pedalling enough of a "disturbance" to stop shimmy from starting, in these circumstances?)...and is that "most bikes that are prone to shimmy can be made to shimmy by doing this" or the rather startling "most bikes shimmy" ?

I like to freewheel no-hands on occasion, perhaps I need to be careful!

@teh....can you put some numbers to your thoughts for avoiding shimmy?
For a given type of bike, what head angle and fork offset etc. would you be looking for?

I have never had a slower (one or two cycles/sec. ) weave.....I don't think I want one!

(I had a speed wobble on a motor bike at 70-ish, 2 up and a tent...it seemed to be a low frequency, sashaying sort of motion.... or maybe that was due to my life flashing before my eyes.....)

[[XAP]Bob]

(I had a speed wobble on a motor bike at 70-ish, 2 up and a tent...it seemed to be a low frequency, sashaying sort of motion.... or maybe that was due to my life flashing before my eyes.....)

I'd ask someone following, who wasn't related, to estimate time. adrenalin does wonderful things....

[Brucey]

incidentally the low speed -pre crash instability I described is very similar to what happens in moto GP when rider 'catches' a highside but doesn't save it. The handlebars are hitting the lockstops very often; the process is brief and violent. I'm not sure if it is the same thing as a 'normal shimmy' -but just writ different, or not.

It has been my observation that most of the lightweight bikes I have ever owned have shimmied at a critical speed between about 15-25mph. The main difference between them is the ease with which they can be provoked, and the enthusiasm with which they then shimmy. It seems that the more elastic the frame and wheels are, the greater the maximum amplitude of the shimmy can be. I think that stiffer bikes may shimmy at higher speeds, too.

My own bikes (typically frame sizes ~21" to 23" for horizontal top tubes) are set up fairly carefully and generally pass through the critical shimmy speed with no trouble when my hands are on the bars; I take my hands off the bars and stop pedalling when I test, because that way I can best allow a shimmy to develop and best feel what is happening. It does make a difference if you are pedalling or not but I am not sure what systematic difference to expect.

I can be riding at or close to the critical speed and nothing will happen on some bikes; in this case I can usually provoke a shimmy to start by giving the handlebars a slap.

Generally speaking an undamped resonant system will show a very large amplitude peak response over a very narrow frequency range; this is what my 'no hands' test is meant to be like. When holding the handlebars, there is more damping; the response of the resonant system is a usually a broader range of frequencies, but a reduced peak amplitude, i.e. there will be a broader range of speeds over which a shimmy might occur but a reduced amplitude. This means for my purposes the test isn't very sensitive.

What I don't try and do is the same test whilst fully laden; it is just plain scary because the amplitude and energy of the shimmy can be much greater and I would be worried about controlling it.

Allowing a shimmy to develop on an unladen bike under gentle acceleration is intrinsically 'safe' to some extent in that it costs energy to maintain the shimmy state, and the driving force for increasing the shimmy amplitude is the same small one that would otherwise be increasing the bike's speed; the net result is that the bike can hold a fixed speed where it would normally accelerate a little, all whilst shimmying at moderate ampltude. One hand back on the bars quells the shimmy and the bike will then accelerate normally. A steeper hill will simply produce a more violent shimmy and this would be dangerous. Even allowing a moderate shimmy to be sustained puts terrible stresses through the bike so this is not something to make a habit of.

Any shimmy that develops as the bike decelerates is intrinsically more dangerous, because the kinetic energy of the bike can be converted into shimmy energy. If as the bike slows, the shimmy gets closer to peak resonance, the amplitude can develop very strongly and very suddenly.

Some folk would argue that to have a shimmy at all is bad design. Not so. It will happen at some speed, it is just a question of at what speed and how badly. If you know what to expect and when to expect it i think it is no bad thing.

Bikes with out-of round, out of balance, very flexible wheels can shimmy very badly indeed and this is bad design, or perhap bad maintenance, anyway.

cheers

[mig]

only had shimmy once on a columbus max steel framed bike with deep rimmed alu wheels(so plenty of inherent stiffness) and yes it was picking up speed on a rapid downhill. only happened the once even after riding the same set up over the same roads many further times.

very disconcerting though!

[531colin]

I'm still playing with this....still havn't managed to get a shimmy out of my bike.
Been out 3 times, just locally.
First time out, the hills I selected weren't steep enough, I wasn't getting anything like 25mph.
Second time out, I picked a steeper hill, unfortunately I had quite a sidewind and there are trees lining the road, with the odd gap in the trees....I had to make the odd grab on the bars.
Third time out to-day, different hill again, unfortunately into the teeth of the wind and I probably didn't get to 25mph.

I'm finding it difficult to get the ideal road....it needs to be steep enough to freewheel no-hands at 25mph, so I want good visibility, not so steep as to be suicidal, and a decent width and not much traffic would be good....got some rather bewildered looks from drivers to-day.

Riding no-hands on a tourer requires quite a large angle of lean to steer the bike....much more than on a race bike.
I think I must be used to riding no-hands at around 15mph...I was surprised the last couple of days to find that as I went faster, the angle of lean needed to steer the bike was much greater, the thing was quite reluctant to deviate from straight ahead......is this the general experience?
I think its a very standard 72deg head, 45mm offset.

[Mick F]

I've never ever ever had a bike with shimmy.
Speed?
Roads?
No Hands?
One Hand?
Feet Off?
Never ever ever.

In fact I'd never heard of it until I joined this forum.