Help troubleshooting severe high speed wobble

[redir]

All my bikes high speed wobble except for my Moots and I always chalked that up to it being a compact frame. My bikes are big frames and I think big frame bikes have a tendency to wobble. Yours is not big so I don't know but I'm also surprised that a knee to the TT doesn't stop it, that leads me to believe it's something mechanical and not inherent to the frame.

[David Kirk]

Originally Posted by MikeD

David Kirk, can you comment on the influence of fork trail on shimmy?

Back in the Serotta days we'd have a person get a bike and they would complain that it wobbled so we would have them send the entire assembled bike back to us to test ride just as he owner rode it. It was extremely rare for the bike that wobbled under the customer would so so under me unless I provoked it. It was during this time that I learned I could make any bike wobble if I wanted using a rather foolish method.

We really didn't fully understand the issue at the time and we tried all kind of things to try to fix the problem. One of the obvious ones was to change the fork rake to give more or less trail to see how that influenced things. This was back when all the forks were steel so I would re-rake the fork and then retest. Simple stuff.

This was a very long time ago and the details are fuzzy but I don't recall ever "fixing" a bike by changing the rake/trail. It would wobble just the same (either on it's own or with provocation) but .... it would be more prone to wobbling at either higher or lower speeds. More rake (less trail) meant it would be more prone to acting up at lower speeds and visa versa. So the take-away was that the trail wasn't the cause or the fix but it did influence the speed at which it was most prone to wobble.




One last thing that might be hard to explain but here goes......if I ride a bike and induce wobble one of two things will happen -

- the wobble will set up and last for a short period of time before it damps itself out and reverts to being stable. This is a good thing.

- the wobble will set up and gradually build in amplitude becoming progressively more violent with each cycle until I stop it or it throws me to the pavement. This is not good.

I suspect that the results of the modeling that Ed talked about above is behind this. If the resonant frequency is just right between the top and down tubes it will progress and try to throw you off....if not it will damp itself out and end the cycling.

dave

[Brian Smith]

Originally Posted by David Kirk

(message snipped)
Back in the Serotta days we'd have a person get a bike and they would complain that it wobbled so we would have them send the entire assembled bike back to us to test ride just as he owner rode it. It was extremely rare for the bike that wobbled under the customer would so so under me unless I provoked it. It was during this time that I learned I could make any bike wobble if I wanted using a rather foolish method.

We really didn't fully understand the issue at the time and we tried all kind of things to try to fix the problem. One of the obvious ones was to change the fork rake to give more or less trail to see how that influenced things. This was back when all the forks were steel so I would re-rake the fork and then retest. Simple stuff.


One last thing that might be hard to explain but here goes......if I ride a bike and induce wobble one of two things will happen -

- the wobble will set up and last for a short period of time before it damps itself out and reverts to being stable. This is a good thing.

- the wobble will set up and gradually build in amplitude becoming progressively more violent with each cycle until I stop it or it throws me to the pavement. This is not good.


dave

This type of diagnostic activity continued well after Dave's tenure at Serotta, and more information was gathered, but with essentially similar results.

I can add a modicum of information to Dave's fine summary, in that even in the case of underdamped oscillation, the building amplitude does largely reach a maximal level, at which point the grotesque behavior ceases to become "more violent," although it is not a steady state that most riders would deem to be an acceptable system tuning outcome. The spring(s) in the system do have a positive rate, and the system reaches a point where the forces producing the displacement grow less rapidly than the effective spring rate resisting them. If the rider can tolerate the system acting this way, unfavorable though it may be, they can at least be advised that the unfavorable condition won't continue worsening until they are thrown to the ground. In practical terms, amplitudes at this point can easily exceed 3" of displacement at the top of the stem/steerer, whereas even only 1" is enough to shock most riders not anticipating the shimmy behavior.

[Bwana]

Originally Posted by David Kirk

Back in the Serotta days we'd have a person get a bike and they would complain that it wobbled so we would have them send the entire assembled bike back to us to test ride just as he owner rode it. It was extremely rare for the bike that wobbled under the customer would so so under me unless I provoked it. It was during this time that I learned I could make any bike wobble if I wanted using a rather foolish method.

We really didn't fully understand the issue at the time and we tried all kind of things to try to fix the problem. One of the obvious ones was to change the fork rake to give more or less trail to see how that influenced things. This was back when all the forks were steel so I would re-rake the fork and then retest. Simple stuff.

This was a very long time ago and the details are fuzzy but I don't recall ever "fixing" a bike by changing the rake/trail. It would wobble just the same (either on it's own or with provocation) but .... it would be more prone to wobbling at either higher or lower speeds. More rake (less trail) meant it would be more prone to acting up at lower speeds and visa versa. So the take-away was that the trail wasn't the cause or the fix but it did influence the speed at which it was most prone to wobble.




One last thing that might be hard to explain but here goes......if I ride a bike and induce wobble one of two things will happen -

- the wobble will set up and last for a short period of time before it damps itself out and reverts to being stable. This is a good thing.

- the wobble will set up and gradually build in amplitude becoming progressively more violent with each cycle until I stop it or it throws me to the pavement. This is not good.

I suspect that the results of the modeling that Ed talked about above is behind this. If the resonant frequency is just right between the top and down tubes it will progress and try to throw you off....if not it will damp itself out and end the cycling.

dave

So how exactly do you MAKE them wobble?

Maybe explaining the process will show bluesea (and others that have wobble problems) that they might be unintentionally contributing/initiating the wobble in the first place.

[David Kirk]

Originally Posted by Bwana

So how exactly do you MAKE them wobble?

Maybe explaining the process will show bluesea (and others that have wobble problems) that they might be unintentionally contributing/initiating the wobble in the first place.

No one would do this unintentionally thereby setting up a wobble that much is for sure.

A wobble is the bike twisting one way and steering the other so to get that to happen I ride no-handed at about 20 mph down a gentle grade and I take my fist and I give the rearmost part of the stem that attaches to the steerer a firm back to the side. This twists the head tube relative to the seat tube and at the same time makes the bike steer to the side because the weight of the bar/levers in front of the steerer have no urge to follow the head tube. DO NOT hit the stem forward of the steerer. This is bad.

So....the bike twists one way and steers the other. After you whack it the frame will rebound in the other direction and when it gets to the end of that rebound the stem will swing in the other direction......and now we have wobble or shimmy.

Caveat - you need to hit it pretty hard for anything to happen. A light tap won't result in much. Secondly.....if you try this and end up on the floor it's not my fault and I'll tell your lawyer to pound sand. Be careful. It can result in your regretting ever doing it.

But with all that out of the way I've never ridden a bike that won't wobble under these conditions. Some will cycle half heartedly 4-5 times and damp right out some will progressively build to the point where if you don't take corrective action (grab the bars my friend) you will certainly crash.

I hope this is if nothing else informative.

dave

[572cv]

Originally Posted by David Kirk

.......
We really didn't fully understand the issue at the time and we tried all kind of things to try to fix the problem. One of the obvious ones was to change the fork rake to give more or less trail to see how that influenced things. ........

dave

Following on this, it would seem to be consistent that if a fork one was using had say, a rake of 43, and the frame was designed for a fork with a rake of 50, that this would lead to some level of instability, yes?

[David Kirk]

Originally Posted by 572cv

Following on this, it would seem to be consistent that if a fork one was using had say, a rake of 43, and the frame was designed for a fork with a rake of 50, that this would lead to some level of instability, yes?

Hmmmm?

Trail is the number that we are looking for here and the amount of trail will largely contribute to the way the bike steers and its stability.

If the frame was designed for a 50 mm fork one would assume they did a good job (not a safe assumption btw) of combining this rake with the head angle of the frame to give a trail number of say about 55 - 60 mm for a road bike. If you then put a 43 mm fork in it that will increase the trail substantially. How far? it depends on the head angle. But if you had about 57 mm of trail with the 50 mm fork you'd have something like 68 (rough guess) and this would be too stable for most road bikes.

Too stable? How can that be? Well it becomes so hard and unintuitive to steer that the bike will only want to go straight and that is it's own type of instability.

-----------------

Your question and my response have little to nothing to do with the topic at hand - wobble. It's a separate deal.

dave

[benb]

If the same bike has a problem/no problem with two different riders then the fit must impact things.

One of my current bikes will be more wobble prone with a longer stem. Not really an issue for me as the longer stem didn't fit well but it sure was interesting, particularly with the "longer stems are always more stable" stuff that goes around. It's rock solid with stems that are unfashionably short.

Pretty great info in this thread. Seeing what Dave is doing for testing makes a ton of sense. Most of the times I've experienced temporary wobbles involve hitting something in the road just right in a way that must induce a similar force through the steering axis. I've seen it on probably every bike I've ever owned, including motorcycles. None of them have ever had a "resonant" wobble that would get worse, but I've also always known about relaxing and/or unweighting the saddle.

For my road bikes it seems like they've gotten more resistant/stable as time has gone on. I think the most wobble prone bike I had was my first one, an aluminum Trek I got in 2000.

[572cv]

Originally Posted by David Kirk

Hmmmm?

Trail is the number that we are looking for here and the amount of trail will largely contribute to the way the bike steers and its stability.

If the frame was designed for a 50 mm fork one would assume they did a good job (not a safe assumption btw) of combining this rake with the head angle of the frame to give a trail number of say about 55 - 60 mm for a road bike. If you then put a 43 mm fork in it that will increase the trail substantially. How far? it depends on the head angle. But if you had about 57 mm of trail with the 50 mm fork you'd have something like 68 (rough guess) and this would be too stable for most road bikes.

Too stable? How can that be? Well it becomes so hard and unintuitive to steer that the bike will only want to go straight and that is it's own type of instability.

-----------------

Your question and my response have little to nothing to do with the topic at hand - wobble. It's a separate deal.

dave

My question betrays a certain conflation of terminology, sorry for the confusion. Thank you for the response, which, unlike the question, is clear!

So that makes sense, and was consistent with the behavior with the old fork. I have a lot to learn/internalize on this stuff.

[oliver]

You can find out more from here once the embargo is lifted: http://repository.tudelft.nl/islando...tion=education

[oliver]

Also Cane Creek came out with an anti-wobble headset somewhat recently that e-bike manufacturers requested. However, I haven't heard anything about how well it actually works.

https://www.canecreek.com/products/headsets/viscoset

[Rusty Luggs]

Originally Posted by David Kirk

Back in the Serotta days we'd have a person get a bike and they would complain that it wobbled so we would have them send the entire assembled bike back to us to test ride just as he owner rode it. It was extremely rare for the bike that wobbled under the customer would so so under me unless I provoked it. It was during this time that I learned I could make any bike wobble if I wanted using a rather foolish method.

We really didn't fully understand the issue at the time and we tried all kind of things to try to fix the problem. One of the obvious ones was to change the fork rake to give more or less trail to see how that influenced things. This was back when all the forks were steel so I would re-rake the fork and then retest. Simple stuff.

This was a very long time ago and the details are fuzzy but I don't recall ever "fixing" a bike by changing the rake/trail. It would wobble just the same (either on it's own or with provocation) but .... it would be more prone to wobbling at either higher or lower speeds. More rake (less trail) meant it would be more prone to acting up at lower speeds and visa versa. So the take-away was that the trail wasn't the cause or the fix but it did influence the speed at which it was most prone to wobble.




One last thing that might be hard to explain but here goes......if I ride a bike and induce wobble one of two things will happen -

- the wobble will set up and last for a short period of time before it damps itself out and reverts to being stable. This is a good thing.

- the wobble will set up and gradually build in amplitude becoming progressively more violent with each cycle until I stop it or it throws me to the pavement. This is not good.

I suspect that the results of the modeling that Ed talked about above is behind this. If the resonant frequency is just right between the top and down tubes it will progress and try to throw you off....if not it will damp itself out and end the cycling.

dave

I'm not so sure that total stiffness / relative stiffness top/down tubes really encompasses why a given bike will wobble for one rider and not for another. (Wobble in the sense of increasing magnitude oscillation, it is pretty much a given that it will oscillate"at all" if deflected since there is a moment of inertia around steering axis, centering torque provided by trail, and little damping, hence "any bike will wobble")

I suspect it may include weight distribution/front wheel load, and/or, the rider's impact on damping. If you watch motorcycle road racing, you will often see wobble during acceleration off a corner when the front end is getting light (even having steering dampers).

Damping-wise, the mechanical parts of the bike that contribute are friction of headset, and the damping effects of the tire at road surface (friction/hysteresis). I think the lowering the tire pressure suggestion is a good one that is often overlooked. Impacts both damping and pneumatic trail effect of contact patch.

Also, some years ago I did see a video demonstration of motorcycle wobble being halted by the rider getting out of the saddle.

[David Kirk]

Originally Posted by benb

If the same bike has a problem/no problem with two different riders then the fit must impact things.

One of my current bikes will be more wobble prone with a longer stem. Not really an issue for me as the longer stem didn't fit well but it sure was interesting, particularly with the "longer stems are always more stable" stuff that goes around. It's rock solid with stems that are unfashionably short.

Pretty great info in this thread. Seeing what Dave is doing for testing makes a ton of sense. Most of the times I've experienced temporary wobbles involve hitting something in the road just right in a way that must induce a similar force through the steering axis. I've seen it on probably every bike I've ever owned, including motorcycles. None of them have ever had a "resonant" wobble that would get worse, but I've also always known about relaxing and/or unweighting the saddle.

For my road bikes it seems like they've gotten more resistant/stable as time has gone on. I think the most wobble prone bike I had was my first one, an aluminum Trek I got in 2000.

Yes fit has an effect as do other things.

If you place the rider in a very upright position is moves weight off the front wheel and the trail has less weight pushing down on it making the bike more prone to wobble. This is one of the reasons why wobble so often hits when riding no handed or sitting very upright.

The other thing that has a huge contributing factor is what the rider does. Does the rider sit upright with an arch in their back with locked elbows and a white knuckle grip? If so the chance of wobble goes way up. Are they relaxed and soft on the bike? Then chances of trouble are pretty low.

Typically people say a given bike has a problem and in some cases that is the deal....in many cases the bike is merely prone to wobble and the one rider will have real issues and another will find it perfect.

dave

[David Kirk]

Originally Posted by Rusty Luggs

I'm not so sure that total stiffness / relative stiffness top/down tubes really encompasses why a given bike will wobble for one rider and not for another. (Wobble in the sense of increasing magnitude oscillation, it is pretty much a given that it will oscillate"at all" if deflected since there is a moment of inertia around steering axis, centering torque provided by trail, and little damping, hence "any bike will wobble")

I suspect it may include weight distribution/front wheel load, and/or, the rider's impact on damping. If you watch motorcycle road racing, you will often see wobble during acceleration off a corner when the front end is getting light (even having steering dampers).

Damping-wise, the mechanical parts of the bike that contribute are friction of headset, and the damping effects of the tire at road surface (friction/hysteresis). I think the lowering the tire pressure suggestion is a good one that is often overlooked. Impacts both damping and pneumatic trail effect of contact patch.

Also, some years ago I did see a video demonstration of motorcycle wobble being halted by the rider getting out of the saddle.

I agree - the rider fit and actions are one of the biggest deals. If he shifts his weight up and back (upslope stem, lots of spacers...etc) and rides with a death grip they are taking weight off the front wheel and not helping damp the system.

dave

[benb]

Isn't the death grip + locked out arms with lots of weight on them going to be even worse? (Within reason)

It seems like either case can be an issue.. with motos it's definitely the weight coming off the wheel as the bike tries to lift the wheel.. you get the shake when the front wheel gets light at the same time it hit something a little off center. You can kind of feel that on "comfort bikes" that have the bars jacked up to the sky and no weight on the front wheel too.

But flip that around and put too much weight on the front wheel your arms are going to tense up and you're not going to be relaxed enough to be any kind of damper.

I can really feel the negative effect on handling if I'm off balance to the front and using my arms to hold myself up. And little changes seem to have a big effect.