Speed Wobble

Discuss light weight issues concerning road bikes & parts.

Moderator: robbosmans

fruitfly
Posts: 120
Joined: Sun Feb 09, 2020 1:40 pm
Location: Wet coast

by fruitfly

There are some excellent comments on the Parlee RZ7 thread on this subject, but here I wish to describe my experiences so far, even if it only provides amusement for others.

I built up an ARC 8 Escapee frameset. On its first few rides it was fine on all my usual descents with its supplied 15mm offset seatpost, a Berk Lupina saddle, and my usual fit for saddle height, setback measurd from the saddle nose, and distance to hoods.

When the Lupina rails broke at the saddle clamp, I swapped on an SMP full carbon, and noticed that descending was a bit more skittish compared to my S-Work Tarmac SL5. Because I had to have the saddle on the nose end of the rails rather than centred to get my usual fit, I swapped in a Darimo seatpost with 25mm offset. This made the feeling that it was about to wobble happen more frequently. Then I swapped in some new 55mm depth wheels, and that was the last straw-definite scary wobble on every rough descent, which where I live is most of them.

Cue the point and laugh time: during all this I completely overlooked that I had moved my cleats back 2.4cm from my metatarsal joint in a trial of mid-foot cleat position, but most of my experience was on the trainer. Perhaps because this requirs lowering the saddle, and I didn't change the fore-aft position of the saddle, I didn't consider it.

So I started reading, and there seems to be wide divergence of opinion on causes. I had plenty of opportunities in the past few days to try clamping the top tube with knees, relaxing my arms and hands, changing position, and getting off the saddle to uncouple resonance to my body from the frame. Nada.

So I have done 3 experiments, 2 on purpose and once by accident. First I swapped wheels on the ARC 8 with my winter bike. All three of my wheelsets wobble on the ARC 8, and none wobble on the Domane, so it's not wheels, spokes, tires. Second I moved the saddle forward a cm, but this didn't make a difference. The accident was my saddle slipped back 2cm on yesterday's ride and the wobble went from unpleasant but controllable to very scary.

My guess is that the underlying issues is the fore-aft weight distribution. The ARC 8 has very short chainstays (402mm), and is correspondingly rather long from BB to front axle. The SMP low spot is much further back than all my previous saddle where the low spot is about the middle (Power, Lupina, Toupe). I think the combo of saddle, more setback, and mid-foot cleat position all lead to my weight being too far back.

So my next experiment is to put back the 15mm offset post, move the saddle so that the low point is the same setback as my normal fit, and see what happens. I will report after tomorrow's ride.
2020 ARC8 Escapee
2017 Trek Domane
2017 S-Works Tarmac SL5 (sold)
2018 Giant Defy (sold)
2013 Cinelli Stratos (sold)

by Weenie


dmetzinger
Posts: 47
Joined: Mon Jul 22, 2019 5:46 pm

by dmetzinger

Whatever you do, be careful......just had coffee this morning with my buddy that got speed wobble at 55 mph on his Aeroad w/ 62mm DT's and went ass over bars and broke his femur. I watched it all happen - thankfully he is recovering well and is in good spirits. He now realizes going that fast on an aero bike in a cross wind isn't advisable unless you are a pro and get paid to take those risks.
Seven Axiom SG - Campy Chorus 11
Colnago C50 - Campy Chorus 12
Pinarello Dogma F10 rim - Campy SR EPS 11
BMC Timemachine Two - Shimano 105
Open U.P. - Campy Record Hydro
BMC Fourstroke 01 Three

TobinHatesYou
Posts: 7498
Joined: Mon Jul 24, 2017 12:02 pm

by TobinHatesYou

Speed wobble is almost exclusively caused by main triangle flex, most commonly from the HT area.

More weight over the rear of the bike should actually reduce wobble because the wobble is caused by tiny steering inputs from that main triangle flex. If you shift your weight forward, you are putting more weight on the front wheel and amplifying the result of that steering input and also moving the center of mass over the flexy main triangle.

I'm afraid if you're noticing more wobble from a saddle change or cleat position change, that's likely down to mistaken correlation/causation. Other variables have changed which you just haven't accounted for.

As previously suggested, the quickest way to stop wobble is to brace your TT with your knees/thighs to stiffen the main triangle. If this doesn't fix it, then I'm wondering if the phenomenon you're experiencing is something else and not actually speed wobbles.
Last edited by TobinHatesYou on Wed Mar 31, 2021 11:48 pm, edited 1 time in total.

fruitfly
Posts: 120
Joined: Sun Feb 09, 2020 1:40 pm
Location: Wet coast

by fruitfly

dmetzinger: good advice indeed! One of the reasons I stopped racing apart from the need to get a career going, was that mentally I switched from "If I don't go fast down this hill I am going to lose" to "If I crash here I am going to die". I did over 200,000m of climbing last year, so did about the same amount of descending, but the Tarmac SL5 was fantastic on descents, so while I thought I was sedate, my friends thought I was a maniac. So it is a bit of a shock to be nervous on a bike. I am aware at 66 that I don't bounce any more.

A guy on a club ride a few weeks ago was going up a suburban street and someone pulled out of her driveway without looking, and he ended up underneath with three broken cervical vertebrae and unable to feel his arms and legs. That fickle finger of fate is everywhere, more so now that reflexes are going going gone.....
2020 ARC8 Escapee
2017 Trek Domane
2017 S-Works Tarmac SL5 (sold)
2018 Giant Defy (sold)
2013 Cinelli Stratos (sold)

fruitfly
Posts: 120
Joined: Sun Feb 09, 2020 1:40 pm
Location: Wet coast

by fruitfly

TBH: There is an intereting paper from an Italian group who put sensors on every part of the bike and then measured before, during onset, and while wobbling, and it is more complex than you suggest. The frame was designed to maximize torsional stiffness first, and then lightness, aero, and comfort, followed, but the requirement was that improvement in these areas could not affect torsional stiffness by more than 5%. One of the major divergences in opinion seems to be whether it is better to damp the frame or lessen damping during wobble. Supporters of the latter suggest that because the human body resonates at a similar frequency to the bike, it is mistake to increase coupling. Instinctively, I want to damp!

I will do the experiment and let you know. I am fine with beautiful theories (including my own) being slain by ugly facts.
2020 ARC8 Escapee
2017 Trek Domane
2017 S-Works Tarmac SL5 (sold)
2018 Giant Defy (sold)
2013 Cinelli Stratos (sold)

TobinHatesYou
Posts: 7498
Joined: Mon Jul 24, 2017 12:02 pm

by TobinHatesYou

The human body is the opposite of stiff. It is 70% water by mass. It is a wonderful damper and can buffer a ton of energy.

Case in point, many modern skyscrapers use huge water tanks on the rooftops as ballast specifically to reduce swaying in high winds and earthquakes.

fruitfly
Posts: 120
Joined: Sun Feb 09, 2020 1:40 pm
Location: Wet coast

by fruitfly

I am not competent to comment or decide, but note the controversy between those who claim humans are good at damping frames, and those who claim that the resonant frequency of humans is similar to that of bike frames. For an ignorant person like me, it is not obvious what to do with advice that says : "damp with your knees" vs "Immediately minimize contact with the saddle, don't touch the frame, and relax the hands". So the question is: what's the resonance frequency of a bike frame vs water?
2020 ARC8 Escapee
2017 Trek Domane
2017 S-Works Tarmac SL5 (sold)
2018 Giant Defy (sold)
2013 Cinelli Stratos (sold)

User avatar
wheelsONfire
Posts: 4121
Joined: Mon Jul 07, 2014 8:15 am
Location: NorthEU

by wheelsONfire

Bikes:

Ax Lightness Vial EVO Race (2018.12.21)
viewtopic.php?f=10&t=156137
Paduano Racing Fidia (kind of shelved)
Open *UP* (2016.04.14)


Ex bike; Vial EVO D

User avatar
wheelsONfire
Posts: 4121
Joined: Mon Jul 07, 2014 8:15 am
Location: NorthEU

by wheelsONfire

TobinHatesYou wrote:
Wed Mar 31, 2021 11:55 pm
The human body is the opposite of stiff. It is 70% water by mass. It is a wonderful damper and can buffer a ton of energy.

Case in point, many modern skyscrapers use huge water tanks on the rooftops as ballast specifically to reduce swaying in high winds and earthquakes.
Why do i get the feeling this just will not happen to a Madone?
Bikes:

Ax Lightness Vial EVO Race (2018.12.21)
viewtopic.php?f=10&t=156137
Paduano Racing Fidia (kind of shelved)
Open *UP* (2016.04.14)


Ex bike; Vial EVO D

MoPho
Posts: 690
Joined: Sun Jul 10, 2011 7:48 pm
Location: NorCal

by MoPho

I had a Moots that I was getting really bad and recurring speed wobbles. It would happen at various speeds, especially over 30mph, and a couple of times at over 40mph (that'll wake you up). I spent over a year chasing the problem, changed the fit (stem length and saddle position), the wheels, the fork rake and nothing really fixed it. I am a pretty fast technical descender and I had no confidence in the bike even cornering at lower speeds, I was getting dropped on descents because I would have to ride the brakes all the way down. After the above Cycling tips article came out last year I immediately sold the Moots and later bought a Tarmac SL6 which handles amazingly, should have done sooner. At least with the Covid demand I almost made my money back on the Moots.

gurk700
Posts: 693
Joined: Fri Jan 20, 2017 7:40 pm

by gurk700

Speed Wobbles can be cause by several different parts of the bike however, don't forget human error.
If I've learned anything from racing motos and crashing several times, use your core and have minimal weight on your hands.
Bumps, wind etc will cause your front wheel to step out of line in varying degrees and every human's reflex/tendency by nature is to correct it which makes it worse.

tjvirden
Posts: 108
Joined: Sun Nov 22, 2020 9:21 pm

by tjvirden

TobinHatesYou wrote:
Wed Mar 31, 2021 11:31 pm
Speed wobble is almost exclusively caused by main triangle flex, most commonly from the HT area.

More weight over the rear of the bike should actually reduce wobble because the wobble is caused by tiny steering inputs from that main triangle flex. If you shift your weight forward, you are putting more weight on the front wheel and amplifying the result of that steering input and also moving the center of mass over the flexy main triangle.

I'm afraid if you're noticing more wobble from a saddle change or cleat position change, that's likely down to mistaken correlation/causation. Other variables have changed which you just haven't accounted for.

As previously suggested, the quickest way to stop wobble is to brace your TT with your knees/thighs to stiffen the main triangle. If this doesn't fix it, then I'm wondering if the phenomenon you're experiencing is something else and not actually speed wobbles.
I certainly agree with main triangle flex being key to 'speed' wobble - the top & down tubes store energy that can sustain the oscillation even with no further rider input, but I don't go with less weight on the front necessarily helping reduce it.

I find that having more rider weight forward tends to reduce the wobble, provided that I don't grip the bars tightly. I suspect that the front tire contact patch has a significant part to play, both in resisting the steering motion and acting as a minor damper (hysteretic loss in the rubber) and more weight forward should enhance both effects. Lower front tire pressure, within reason, should also reduce the wobble severity for essentially the same reasons, but I'm not sure it's relevant as pressure tends to be set for other considerations; I haven't actually properly tested varying pressure anyway.

Tandems might be a useful reference point - I think that if your idea is correct, then tandems should wobble very badly. They're both torsionally flexible and have almost all the mass nicely distributed between the contact patches. However, so far as I can tell, very few people have reported any significant problem.

Karvalo
Posts: 1596
Joined: Fri Aug 10, 2018 6:40 pm

by Karvalo

fruitfly wrote:
Thu Apr 01, 2021 2:17 am
I am not competent to comment or decide, but note the controversy between those who claim humans are good at damping frames, and those who claim that the resonant frequency of humans is similar to that of bike frames.
Noted. I have also recently noted the controversy between those who think the earth is flat and those who think it is spherical.

tjvirden
Posts: 108
Joined: Sun Nov 22, 2020 9:21 pm

by tjvirden

fruitfly wrote:
Thu Apr 01, 2021 2:17 am
I am not competent to comment or decide, but note the controversy between those who claim humans are good at damping frames, and those who claim that the resonant frequency of humans is similar to that of bike frames. For an ignorant person like me, it is not obvious what to do with advice that says : "damp with your knees" vs "Immediately minimize contact with the saddle, don't touch the frame, and relax the hands". So the question is: what's the resonance frequency of a bike frame vs water?
Cyclingtips article is helpful, but have a read of this too: https://www.sheldonbrown.com/brandt/shimmy.html

I'm not convinced there is any controversy the way you've framed it - confusion perhaps. Shivering while coasting has been discussed/known about as a starter and/or enhancer of shimmy since the dim and distant past. Damping the oscillation (shimmy/speed wobble, whatever name is used) using a variety of means (legs against top tube, changing grip on bars etc etc) is no doubt just as ancient. It's much less of a problem now as frames typically have much higher torsional stiffness than steel frames of yesteryear.

"what's the resonance frequency of a bike frame vs water" If you mean resonant frequency of shimmy of the assembled bicycle with rider, then resonant frequency of water (that's cooking food in a microwave oven) is of no relevance at all.

Continue experimenting until you solve it - the number of variables is enormous so one person's suggestion may or may not work for you.

by Weenie


rides4beer
Posts: 586
Joined: Wed Nov 28, 2018 5:27 am
Location: SC

by rides4beer

I've done 50+mph on four different bikes, multiple wheelsets, never had any kind of wobble/shimmy. Am I just lucky? Is it because I'm 84kg? Is it because I automatically tuck my knees in for better aero (not really gripping the toptube, but def touching)? It's always in the back of my mind, after hearing lots of stories, but thankfully haven't experienced it.

Post Reply