frames with 7/4/7 25.4mm top tube

[unterhausen]

I am building myself a randonneuse with Columbus SL. It comes with an 8/6/8 top tube. I'm weighing the idea of ordering a Kaisei 7/4/7 top tube as an experiment, since some people really like the way that makes the bike feel.

Anyone have a bike like that who can give their impressions? My experience has been that I'm not real sensitive to frame tubing, but I have never gone with a 1" tube that thin before. My current road bike has 1" 9/6/9 and my other road bike has 28.6mm 7/4/7, which is somewhat stiffer than 25.4mm.

[ColonelJLloyd]

The 1" 8/6/8 TT is already going to be less stiff than the 1.125" 7/4/7 TT on your other road bike, right?

[hampco]

I had a frame with a 1" top tube, 56cm c-c, probably 7/4/7. It had the most delightful shimmy on descents and I couldn't get rid of it fast enough.

[false_Aest]

Hampco,

Can you identify if the shimmy was a direct result of the TT?

In other words, if the frame was remade with a 28.6TT and the geo + fork remained the same, do you think the shimmy would disappear?

[8aaron8]

I have built a number of frames with 1" 7/4/7 top tubes, most have been for myself as I am quite light. On road frames without carrying weight I absolutely love the way they ride, of course the frames other tubes make a difference in it's ride qualities. I do have a randonneur bike with the same TT configuration and will say that I can get that bike to shimmy at speed, though I can safely stop it by coming off the saddle slightly or pinching the TT with my legs. With the added weight of fenders, a rack, front bag, longer chainstays and larger tires I will say this frame flexes much more than my road bikes, which for the most part is fine, but sometimes I do feel when I push hard much of my effort is lost in the frame flex and unfortunately I'm not feeling it spring back to me. However, with that said I'm not pushing myself as hard on the randonneur as I do my road bikes so I am 95% elated with it's ride. There is so much to be said on this, but on a finishing note, when building for myself, a 5'10" 130lbs rider, Columbus SL has been my favorite all around tubeset for a frame, more so than my bike with 7/4/7 1" TT, for me it provides the perfect blend of spring and stiffness.

[timto]

I'm with Hampco - While I do believe in flexible frames there is 'too' flexible when shimmy is involved. Some are ok with shimmy but I had a bike with it and decided that it was unacceptable for me. I've since landed on the os size at 160 lbs and my particular riding style which involves the occasional bar grabbing big ring stomp. The super skinny stuff is nice when tempo on flats or noodling around at cruising pace but I didn't like the movement in the high torque out of saddle scenario nor the occasional shimmy.

[unterhausen]

well, for fear of getting into another pointless internet argument, I have seen really good evidence that shimmy is not related to frame stiffness. Or rather, you have to lower stiffness a lot in order for it to influence shimmy. Far less stiff than this frame will be. Maybe I'll do a modal analysis on the frame after I build it.

I'm not that tall anyway, this will be a 55cm frame.

[Ken Robb]

If a randoneuse bike is meant to carry loads why would it make sense to try use such light tubing that might not be stiff enough to handle well when loaded. I understand that very light tubes might make sense for a race bike. Besides the question of how it might handle isn't this a bike that would be "parked" while loaded during trips? Leaned against a tree or even on a kickstand I dropped mine several times and only got a few scratches. Really light tubes would probably have dented.

[chazzwazzer]

The Rawland Nordavinden was one such frame. I owned one (56cm) and rode someone else's (58cm). They are also very low trail frames and I'd guess the flex fork kept some shimmy away.

The only time I recall having shimmy was riding the 58 for the first time at very low speeds (like walking speed) with a front load. I can't recall if it had a needle bearing headset. Never had the issue on my 56. I imagine the shorter top tube provided less leverage. I really enjoyed that bike, very comfortable, very easy to pedal and fast feeling. I sold it because I started training for track sprinting and the frame would wind up strangely under hard starts (duh). I also recall riding across wet grating on a bridge and realizing the front end and back ends were flexing in different directions! I'm no stiffness disciple but it has a place, sometimes :)

I would argue that a thin/light TT absolutely has an effect on shimmy, and for "planing" purposes is the whole point: to allow the head tube and seat tube to have some lateral independence. It feels really good when it works.

[Mark McM]

well, for fear of getting into another pointless internet argument, I have seen really good evidence that shimmy is not related to frame stiffness. Or rather, you have to lower stiffness a lot in order for it to influence shimmy. Far less stiff than this frame will be. Maybe I'll do a modal analysis on the frame after I build it.

What is the good evidence that shimmy is not related to frame stiffness?

I've seen very specific cases where a reduction in frame stiffness was accompanied by an introduction of a shimmy that hadn't been present previously.

[David Kirk]

If you run the tube diameter and wall numbers in a calculator I think you'll find that the difference in stiffness between your thick choice and your thin choice are surprisingly small.

The tube diameter is of course the biggest factor when comparing tubes in either bending or torsion and wall has a much smaller effect.


dave

[unterhausen]

If a randoneuse bike is meant to carry loads why would it make sense to try use such light tubing that might not be stiff enough to handle well when loaded.
Randonneuring is far closer to racing than it is to touring. It's touring for people that aren't smart enough to stop for the night. You are racing the clock, if nothing else. I am not building anything like a touring bike. My description of a randonneuse is a racing bike with lights, fenders, and a light rack for a front bag. It's a bit of a stretch, but that's my starting point. I know some people like to use touring bikes for randonneuring. I think those people are either crazy or a lot stronger than me.

As far as shimmy goes, I know this is pretty much pointless to expound upon, but the literature doesn't support frame stiffness as a cause for shimmy on a usable frame. And the lowest modal frequency we have found on a bike is 10Hz, which is above any shimmy I have ever seen. I don't know how you reduce stiffness on a frame, but it sounds like a bad idea. Unless it's not the same frame. In which case the stiffness wasn't reduced.

Just in case it will reach someone that is on the fence about this, any bending involved in shimmy will be relieved by motion in the headset. Kind of hard to bend something that has a hinge in the middle.

I have owned 3 bikes that would shimmy. It was always accompanied by rotation of the fork in sync with the shimmy. One of these bikes was fixed by aligning it. I haven't checked the other two. One is in constant use because a little shimmy doesn't bother me too much as long as I know it's going to happen. And I don't expect this new bike to shimmy no matter what I do with the top tube.

[hampco]

Hampco,

Can you identify if the shimmy was a direct result of the TT?

In other words, if the frame was remade with a 28.6TT and the geo + fork remained the same, do you think the shimmy would disappear?

As has been pointed out, it may sometimes be tough to isolate a single factor when trying to nail down a front-end shimmy. I'm quite confident that in the case of the frame I mention above either stouter tubes or larger tubes with a similar wall thickness would have resolved the issue.

In the twenty years that I've been selling and designing bicycles there were a few early iterations that were reported to have some front-end shimmy. I usually either replaced the frame or the top tube and head tube and the issue resolved itself. I know what tubes to use and what sizes to avoid - but it's partly trial-and-error, partly my own experience along with that of the folks I work with.

A good rule of thumb is that a 10% increase in wall thickness will give a 10% increase in torsional stiffness but a 10% increase in tube diameter will yield a 20% increase in torsional rigidity. Ymmv.

[Mark McM]

As far as shimmy goes, I know this is pretty much pointless to expound upon, but the literature doesn't support frame stiffness as a cause for shimmy on a usable frame. And the lowest modal frequency we have found on a bike is 10Hz, which is above any shimmy I have ever seen. I don't know how you reduce stiffness on a frame, but it sounds like a bad idea. Unless it's not the same frame. In which case the stiffness wasn't reduced.

Resonant frequency analysis is not necessarily helpful in the case of bicycle shimmy, because shimmy is not strictly a resonant phenomenon. Instead, it is an example of a Hopf Bifurcation. There was letter in a Velonews Technical FAQ (http://www.velonews.com/2013/11/bikes-and-tech/technical-faq/technical-faq-bifurcation-and-high-speed-shimmy_309601) a few years that goes into a few details about this. Below is an excerpt, with a few key points pertinent to this discussion highlighted:



Dear Lennard,

A linear analysis leading to resonance is appropriate for any system where there is an oscillator that is being forced at a special frequency — the resonance frequency — and when this happens, the amplitude can simply build to infinity. This is not what happens in bicycle instability for two reasons: first, there is no periodic forcing that causes the high-speed wobble (in fact, it can happen on a smooth road); and second, there is not a phenomenon that shows a characteristic building of amplitude.

Instead, the high-speed wobble is a critical phenomena, which is typical of bifurcations and bifurcation theory in general. Below the critical parameter value, you see one thing, in this case a stable equilibrium characteristic of a smooth ride, and slightly above the critical parameter, the smooth ride is no longer stable (but it still exists as an equilibrium, but an unstable equilibrium, just as standing a stick upright is an equilibrium but unstable because if it tips even slightly away from the exact equilibrium, it quickly drifts away), but the now unstable equilibrium gives way to a stable periodic orbit, which is the wobble. And as the parameter increases, the amplitude of the wobble can increase to some larger but fixed amplitude.

Also, Hopf-born limit cycles are self-exciting if you like to see it that way, as opposed to resonance that requires an external forcing to excite the resonance frequency. (For resonance, think of a bridge with a special characteristic frequency and if soldiers march over it moving their feet just at that frequency, then you have resonance; this happened when I worked at the Naval Academy when some of the midshipmen forgot to break step and they actually did crack a walking bridge!)

There are many types of bifurcations, and they explain all sorts of critical phenomena on nature, and this type, the Hopf bifurcation, explains the onset of oscillations in all sorts of natural systems, from population dynamics, to chemical reactions, to airplane wing flutter, to stability of steerers in bicycles, trucks on trains (leading to derailment) to landing gear wobble on airplanes.

To many engineers not in the know, these oscillations and their onset are often mistaken for resonance, but it requires a nonlinearity in a certain way. This is often guessed by the critical onset of the phenomena as a parameter is adjusted.

For the bicycle scenario, the parameter in the engineering phase could be increasing the stiffness build of the bike — a stiffer bike will be more resistant. The several design parameters could all be “nondimensionalized” to a single, non-dimensional parameter as is necessary to explain with Hopf since it is a one-parameter phenomena (this concept is called co-dimension-1 bifurcation). Or once the bike is designed and built, then the parameter is speed. Each bike has a critical speed at which it will cross the Hopf bifurcation value, and then steady state becomes unstable and wobble becomes the stable state. This, by the way, is why it is very, very dangerous to try those 150 mph downhill bicycle attempts!

The letter goes on to give some information of other examples of Hopf Bifurcation instability, such as aerodynamic flutter in airplane wings, and the very famous collapse of the Tacoma Narrows bridge during a high wind event.

[unterhausen]

It's pretty complicated and that letter writer gets it wrong. Faulting engineers for not understanding non-linear dynamics when he gets simple facts about linear dynamics wrong in the first sentence. Yes, bicycle steering geometry is non-linear. Bicycle structural dynamics is well modeled by linear analysis. It's too complicated for a letter to Lennard Zinn, that's obvious.

I thought it was interesting, but the letter writer has obviously not looked at the literature on this subject. I detect enough education that I feel like that person probably knows better. Or maybe they only got a undergrad degree in math and never had his head (metaphorically) ripped off for talking nonsense like that.

[dddd]

Shimmy is one thing, but the overall handling also is affected by flex in the chassis, so the weight of the rider is something that would have to be considered in order to make a decision as to the use of lighter-gage tubing.

I ride with others who weigh 1-1/2 times what I weigh at the same height, so we are always on different pages when talking about flex characteristics or about reliability. In particular, I am often alarmed to see how fast that they tear up wheels that would last indefinitely under my own usage. Broken Ksyriums right and left, and just last week a broken De Rosa carbon downtube. He bought the "SL" model, lol.

[Mark McM]

It's pretty complicated and that letter writer gets it wrong. Faulting engineers for not understanding non-linear dynamics when he gets simple facts about linear dynamics wrong in the first sentence. Yes, bicycle steering geometry is non-linear. Bicycle structural dynamics is well modeled by linear analysis. It's too complicated for a letter to Lennard Zinn, that's obvious.

I thought it was interesting, but the letter writer has obviously not looked at the literature on this subject. I detect enough education that I feel like that person probably knows better. Or maybe they only got a undergrad degree in math and never had his head (metaphorically) ripped off for talking nonsense like that.

And just what does the letter writer get wrong in the first sentence?

Before making assumptions about the writer of that letter, you should probably look into his background. The writer, Dr. Erik M. Bolit, has more than "undergrad degree in math", he is a professor of mathematics, and holds the W. Jon Harrington chair at Clarkson University. And I'm pretty sure he has looked at the literature on the subject (bifurcated dynamic systems), since he is also an Associate Editor of the "International Journal of Bifurcations and Chaos" - in fact, he has written some of the literature on the subject.

Other than just not liking the conclusions, what can you provide to counter the assertion that bicycle shimmy is an example of a Hopf Bifurcation? Also, you have yet to present any evidence that frame stiffness doesn't play a role in shimmy.

[false_Aest]

im so out of my element my ashes need to be poured out of a coffee can.

im hoping ya'll can dumb some of that letter to Zinn down for me.

[unterhausen]

so he should know better. He is making an argument from authority without actually looking at the specifics. The notion that engineers don't understand non-linear systems is patently ridiculous. I was studying non-linear differential equations when he was in short pants, and I'm an engineer. I know what he meant. That doesn't mean he is right.

Linear systems can have damping. Any system found in nature, including bicycles and bicycle structural dynamics will feature some damping. So even they will decidedly _not_ go to infinity. That blows his credibility right out of the water right away. Plus, there is plenty of energy floating around in the frequency band required to excite a resonance, even on a smooth road, so that's just wrong too. And it's clear he never looked at bike dynamics. But the guys that do bike dynamics consider the frame a rigid body, so that doesn't help us much. For good reason, really, but it doesn't answer this controversy.

You can consider bicycle dynamics to be two loosely coupled dynamic systems. The frame is well modeled as a high order linear dynamic system. Thus an eigenfunction/resonant frequency analysis is very much appropriate. The steering geometry is a fairly well-behaved non-linear system. These two dynamic systems are coupled in that the weight of the rider and the forces from steering push on the structure of the frame. It's not all of a sudden going to adopt some weird non-linear behavior because the input forces are being generated by non-linear dynamic system. We know how this works, it's not magic.

Now, if the stiffness of the frame is such that it is low enough that it is influenced by the limit cycle of the steering geometry in shimmy, then it will affect the limit cycle in some way. My assertion is that this is not the cause of shimmy, because the shimmy is a limit cycle that is dominated by the properties of the steering geometry. You certainly wouldn't see the gross deflection of the steering axis in shimmy due to a frame resonance. However, the lowest resonant frequency of most frames is too high to really contribute anyway.

Ok, adding this on how the stiffness of the frame can play significantly in shimmy. If the rotational stiffness of the frame about the head tube is very low, any deflection of the bike from vertical will cause a steering input. The steering geometry of a bike is stable in small deflections, so the wheel will turn, then it will go to the other side, turn back, etc. Shimmy. Note that I said "very low." This has not been my experience with my bikes that shimmy. All of them have been quite stiff frames. I'm relatively short.

I'm glossing over some of the issues. The point is it really is unlikely that structural elements of the frame contribute significantly to shimmy. The truth is in some details that take a significant amount of work to establish. The literature says no, but hope springs eternal. I am not asserting this with 100% certainty, but the letter writer fails to take some fairly obvious facts into account and doesn't recognize the nature of the system he is making conjectures about.

I think if the letter writer had experience with the engineering approach to large order dynamic systems, he would not have mentioned structural stiffness in the context of shimmy.

Maybe I'll write a paper and send it to his journal.

[ripvanrando]

Also, Hopf-born limit cycles are self-exciting if you like to see it that way, as opposed to resonance that requires an external forcing to excite the resonance frequency.

I stayed up all night watching my bike on the trainer but it never self-excited.

On the road I can INDUCE a periodic oscillation by slamming or banging the handle bar-easily done. I also notice that when I'm shivering on a cold descent, most bikes will shimmy for me (except the one Zinn designed for me).

Resonance phenomenon to me but been almost 4 decades since this dumb ass engineer had graduate level control systems.

[unterhausen]

I don't think there are any undergrad engineering programs that really prepare you to analyze a nonlinear system at all. We usually linearize systems and then decide if it was a reasonable thing to do. Nonlinear control is definitely an advanced topic, not studied by all that many graduate engineers. Turns out that my Ph.D. advisor sent a lot of students off to the math department, so I happen to know a lot of engineers that have studied nonlinear dynamics. I had done it on my own as an undergrad.



Other than just not liking the conclusions, what can you provide to counter the assertion that bicycle shimmy is an example of a Hopf Bifurcation? Also, you have yet to present any evidence that frame stiffness doesn't play a role in shimmy.
Ignoring the faults in his letter, I am not disagreeing with his assessment of the dynamics of the phenomenon at all. Where does the Hopf Bifurcation come from? The steering geometry, which he doesn't address. He doesn't have any basis to say anything other than structural stiffness might be a cause or contributing factor, but it really depends on the construction of a particular bike frame. Whereas he leaps to that cause in a way that leads people to believe it is the most likely. There are plenty of very stiff bike frames that shimmy. There are many other mechanisms that are more likely than structural stiffness, misalignment being most likely in my personal experience.

His argument is akin to the apparently apocryphal story that Descartes once won an argument by saying that a squared plus b squared equals c squared, therefore god exists. Yes, it's a nonlinear system with a limit cycle, yes, most engineers have no idea what that is, no, it doesn't necessarily follow that structural stiffness is the cause.

[ripvanrando]

UnterHausen......please advise. I have a bike prone to shimmy but is not terribly disturbing. Others who have the same bike claim that theirs does not shimmy.

Is it possible that I have a "defective" carbon fork, meaning it is not aligned properly and being carbon can't be cold set?? This is at least my suspicion. Thinking of buying a cheap fork and installing and testing.

BTW.....doing PBP under 56 hours is for crazy people. Good luck next summer.

[choke]

There was a good thread on the subject of wobble last year with a lot of input from Dave Kirk. https://forums.thepaceline.net/showthread.php?t=203949

Though if you don't want to read the entire thread, this post by DK is the most interesting of the bunch... https://forums.thepaceline.net/showpost.php?p=2163709&postcount=26A few things about wobble -

- all bikes can speed wobble. Every single bike ever made can be made to do it. Zero exceptions. I'm sure there are a lot of folks reading this and thinking "BS - my bike won't"....but I'm here to tell you that I can get on it and make it wobble every time. Fly me to your house and I'll be happy to prove it

- big bikes are more likely to wobble. The longer main tubes will be softer in torsion than the same tubes in a shorter frame and therefore they are more prone to wobble.

- bikes with smaller main tubes are more likely to wobble because the tubes are not as stiff in torsion and it's easier for the cycle to set up.

- longer stems contribute to the likelihood of wobble. A long stem moves the weight of the bars and levers further away from the center of steering rotation and makes the bike more likely to steer to the side as the wobble cycles from one side to the other.

- heavy levers make a bike more likely to wobble. Back in the day, when STi was new we suddenly saw a rash of bikes of all brands wobble. It was the added weight of the levers so far from the center of rotation and make it more likely to steer.

- alignment is almost never the cause of wobble.....in fact the closer the alignment is to perfect the more likely it is to wobble. It sounds crazy until you think of it. If the bike is so out of line that it pulls hard to the left all the time it's very hard to get a cycle to set up where it will want to go right on its own. So rather than going right-left-right-left it just goes left. One can test this by purposely misaligning a bike by cocking the wheels in the frame or fork. Cock the wheels in there so it pulls hard to one side and it won't wobble. Not exactly a fix but certainly interesting.

[Mark McM]

Ignoring the faults in his letter, I am not disagreeing with his assessment of the dynamics of the phenomenon at all. Where does the Hopf Bifurcation come from? The steering geometry, which he doesn't address. He doesn't have any basis to say anything other than structural stiffness might be a cause or contributing factor, but it really depends on the construction of a particular bike frame. Whereas he leaps to that cause in a way that leads people to believe it is the most likely. There are plenty of very stiff bike frames that shimmy. There are many other mechanisms that are more likely than structural stiffness, misalignment being most likely in my personal experience.

I'm still having trouble with what is wrong with what Erik Bolit has written. His mentioning of resonant oscillations tending to increase to infinity at the critical frequency isn't a disregard of the role of damping - I read it as showing that shimmy behavior does not resemble a resonant vibration. In a resonant vibration, the amplitude of oscillation increases as the frequency of the forcing function approaches the critical frequency, and diminishes as it moves away from the criticial frequency. This is not the behavior of bicycle shimmy - it doesn't slowly build as speed builds, and then slowly diminish as the bike speeds up or slows down from the critical frequency. Instead, it tends to initiate suddenly, and not diminish until the speed has changed drastically. This is more in line with a bifurcated system. In addition, there is no obvious forcing function to cause a harmonic oscillation in the steering - forcing functions from the wheel turning will be primarily in the plane of the wheel, an not the out-of-plane force needed to cause a steering shimmy. (I've ridden wheels with enough mass imbalance to cause the wheels to practically hop off the ground, but it didn't cause the steering to shimmy).

It's hard to say how much time Dr. Bolit has spent examining bicycle shimmy (although he obviously has enough interest in riding to send a letter to a bicycling magazine). However, he is undoubtably aware of the work done on automobile steering shimmy, were it has largely been demonstrated self-excited shimmy (which can occur on smooth roads with perfectly round and balanced wheels) is a result of a hopf bifurcation.

As far as your "stiff" frames that shimmy: The first obvious thing to say is that frames don't shimmy - bikes do. There are so many other variables at play, that it certainly possible that a bike with a relatively stiff frame may still shimmy. But that doesn't mean that frame stiffness doesn't play an important role in shimmy, or that changing the stiffness of your frames won't change its shimmy behavior. Years of experience has shown that bikes with less stiff frames are more prone to shimmy. In a many cases, making a frame stiffer has resulted in the elimination of shimmy (or at least raised the critical speed outside the range or normal operation).

Now, It might argued that it is difficult if not impossible to increase frame stiffness without changing anything else, so how can we be sure that it was the frame stiffness increase that eliminated shimmy? Well, we can't be absolutely certain in all cases, but there have been some cases where a bike that previously never shimmied suddenly developed a shimmy after experiencing a decrease in frame stiffness. I'm speaking about frames which have developed fatigue cracks in their main frame tubes. When a fatigue crack is still small, a bike may still be perfectly rideable. The crack does not change geometry or mass distribution of the frame, and obviously it does not affect the other bike components - its only immediate affect is to reduce the stiffness of the tube affected, and subsequently the total stiffness of the frame. I've seen several cases where cracked frames have developed shimmies that had not been present previously.

[adampaiva]

Unterhausen - just curious, are you in the acoustics program at PSU ?