rpenmanparker 

Not exactly. I am assuming that you were forced to accept a trail that compromises every other aspect of handling except for stabilizing the bike against shimmy. You shouldn't have to give up proper handling to get safety. I understand you may not see it that way, but did the bike with the original fork handle badly at all other speeds? The problem is NOT usually about fork rake and trail.

wphamilton

Seems plausible and may be right, but I have reservations from the beginning assumptions "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."

Rider response can be a periodic forcing due to the negative feedback nature, which conforms to the solid advice of releasing your death-grip, or loosening up as I said earlier. In my few experiences there was always a trigger of some sort, and then something that seemed like a periodic forcing input. And it is not clear that there is no "characteristic building of amplitude" - my own experience is that there is?

That said, hopf limit cycles are similar enough to resonance for our purposes, does it really make a difference? Other than academic.

Shimagnolo

No, the trail is in the normal range, and the bike handles just fine.
The trail is the product of headtube-angle and rake.
1. I measured headtube-angle and rake.
2. Calculated existing trail.
3. Referred to Moulton's articles for desired trail.
4. Calculated rake to get the desired trail.
5. Ordered fork.

Shimmy is all about trail:
As trail increases, stability increases, and the bike is more reluctant to turn.
As trail decreases, stability decreases, and the bike is quicker to turn, and also more prone to shimmy.

on the path

I regularly descend at 40+ mph and have descended in the high 50s. Never had a shimmy like described here. If I had a bike that did, I'd fix the problem permanently or ditch the bike.

rpenmanparker 

So your frame was defective and didn't match the nominal head tube angle specification? The head tube angle did not match the fork rake to give an appropriare trail? I would not have fixed the problem for the manufacturer with a rather bizarre 37 mm rake fork for a road bike. I would have gone back to the maker for a resolution. To me that would mean a new frame that would correspond to a 43 or 45 mm fork as advertised. The point is that stock frame and fork combinations shouldn't require an afrermarket fix.

Shimagnolo

I already stated it was a secondhand frame.

I'll let Lennard Zinn explain it to you:

Unfortunately, on road bikes, we don’t get a bigger wheel-size option, and, to add insult to injury, we even have rocket scientists at the UCI telling us that the maximum front center dimension we can race is 65cm. Tell that to 6’9” guy who needs a 69cm top tube. How are you going to get to 65cm from the bottom bracket to the front hub if you have parallel seat- and head-tube angles (which so many 5’9” riders find to be quite stable and nice handling) when you also have to add on the fork rake? Any small or medium-sized bike has a front center dimension longer than the top tube, not shorter! The only way you’re going to get to UCI legality if you’re 6’6” or more is to make the head angle way steeper than the seat angle. And you can promptly kiss your good handling goodbye; hello oversteering!

Source: Technical FAQ with Lennard Zinn: Why tall bikes often handle poorly | VeloNews.com

My Litespeed was custom built for a rider on the 2002 Colorado Cyclist team.
It is 64cm from crank center to top of seat tube.

rpenmanparker 

So your frame has a head tube angle. And there is a fork rake that would combine with that HTA to give an absolutely standard trail known to provide good handling. Did your frame come with such an HTA/rake conbination...at least nominally? Why was the HTA such that you think you needed an ultra low rake fork?

American Euchre

My routes are pretty flat. However, it's possible to descend at 36+ mph on certain sections.

I MIGHT have had an instance of high speed shimmy on one occasion. I was descending at 35+ mph and the front end got squirrelly on me for whatever reason.

Fortunately, I haven't been able to duplicate it. It could have been a random gust of wind, or maybe the road surface was sketchy at a particular spot with sand or oil or who knows what.

Can you isolate certain variables that could lead to sketchy handling? Descents where winds blow in variable directions? Extremely sketchy road surfaces? Patches of sand? It COULD be the environment you're riding in, not the bike.

If this is happening consistently on nice road surfaces, when wind is not much of a factor, might as well check out frame alignment at a shop.

Shimagnolo

The reason was in my post that you quoted.

What I bought was a frame with a headset, fork, handlebar, stem, seatpost, and saddle.
IIRC the guy I bought it from was the 2nd owner, (it was 4yrs old at the time).
It came with a fork with a 45mm rake.
I assume that was the original fork, but I don't know for sure.

"Shimmy is usually caused by not having enough trail."
Source: Dave Moulton's Blog - Dave Moulton's Bike Blog - High Speed Shimmy.

"As with any design aspect, more is not necessarily better; for a road bike with a 73 degree head angle the optimum trail seems to be around 2 to 2 ½ inches (5 to 6.3cm.)"
Source: Dave Moulton's Blog - Dave Moulton's Bike Blog - Trail, fork rake, and a little bit of history

Digital angle gauge: Digital Angle Finder / Angle Gauge

Trail calculator: Bicycle Trail Calculator | yojimg.net

Forks available in range of custom rakes: Components - Seven Cycles

woodcraft

Tennis shops sell adhesive backed lead tape, for tweaking racquets.

However, those in the know say that neither valve stem weight, or balancing

bike wheels has any appreciable effect.

VNA

Keep the upper body relaxed, arms bent, hand grip not too tight use your hips to guide and let the bike do its thing!
Be very surprised for a defect on such bike, unless a bent wheel or fork?

L134 

Agreed, and butt off the saddle. It is the rider not the bike.

FBinNY 

No matter how perfect your bike seems to be, there are always imperfections like tiny wheel wobble (possibly at the tire rather than the rim. At high speed these tiny inputs that you'd never otherwise notice get amplified through the magic of harmonics.

You can't do anything about the imperfections except to get them as good as possible, but you can change the harmonics, or increase the dampening forces.

1- try changing the harmonics by placing one knee tightly against the top tube (or both if possible) and pressing into it to change the frame's harmonics, o dampen the vibration (or both).

2- increase steering damping force by reducing front tire pressure or increasing handlebar weight, or bracing the bars by getting low with your hands in the drops and pressing both sides forward so your arms and shoulders act as an A-frame.

3- (this takes a bit of skill) cant the bike slightly to the side while keeping your body over the ground contact line (ie. balanced). The steering forces on a canted bike are very different, and it'll pull in one direction which you can counter with some steering input.

I've also found that avoiding tires with raised centers, or grooves close to the center helps. Anything that alters the contact patch on both sides and isn't absolutely perfect, and it never is, can cause greater steering inputs for the harmonics to work their magic on.

One of my older steel frame bikes tended to shimmy when coasting at about 30mph. But it wouldn't shimmy when I pedaled, probably because I was messing up the harmonics. So my solution was to pedal the bike up beyond 30mph, and try coasting at 35+, at which speed the shimmy disappeared.

So, my message is that while there may not be a cure, the shimmy disease can usually be managed.

jsigone

try different tires and pressure yet?

only bike that I had shimmy with was a CX bike going 50+. had to put the knee on the top tube to control it

Fiery

It's really not that simple. At the very least you have to take into account wheelbase and weight distribution. Moulton focuses a lot on trail, but you can bet he takes care of making the front center and chainstay length what they need to be when designing a bike. And then there's Colnago for example, who seemingly don't worry about trail at all, but are all about having the wheels in just the right place for each frame size - do you really think they have no idea what they are doing? There is a reason why Cervelo guys abandoned fixed head angle for all frame sizes after they started working with pro cyclists, even though it was a seemingly good way to make sure trail is perfect on all their bikes.

mapeiboy

I have yet to experience this problem over the years with my Colnago EP .

Sy Reene

Interesting. Just curious what your trail was before/after the fork change? From your math, did you determine a universal truth to what would be too little amount of trail? It sounds like you increased your trail, correct?

Sy Reene

This isn't Lightspeed's data, but take a look at the geometry chart for eg. the Merlin Extralight or Cyrene.. typical road geometries back in early 2000s anyway.
https://www.bikyle.com/MerlinGeom.htm
As you'll note, as frame sizes get larger, the head tube angles increase (72.5 up to 74 degrees), and in compensation, the fork rakes decrease (45mm to 43mm to 40mm). One can imagine the OPs 'custom' secondhand 64cm bike ideally should have stayed on this progression with an even greater than 74 degree HTA, but it sounds like(?) that maybe the bike was originally provided with a stock fork and there weren't maybe <40mm forks back then? EDIT: I see that OP later mentions original fork was 45mm

bakes1

I wish I could get my 40 year old steel Motobecane to shimmy even a tiny bit on descents.
I love that living on the edge feeling.
I guess that's what I deserve though for not tinkering with my bike incessantly and obsessing over every gram. Someday maybe...

cyclezen

Back to the OP's Q...

I had an OFF-THE-SHELF bike (shall remain nameless) which I found would consistently head-shake/shimmy on fast descents. Always in a straight descent, never when slightly or heavily leaned (like twisties or corners). Happened at about 47+ mph with the 'stock' low profile wheels, happened at 43+ with a set of Higher Profile Eastons. But always on a straight fast descent.
SO yes, wheels make a difference - have you tried the descent on a different set of wheels? result?
My bike, different wheels changed the speed at which it happened, but it always happened at some point above 45 mph ish.
At that speed, I was always scared ****less as to whether I would bring it under control or just crash. Never got easier...
Bike just spooked me - I sold it cheap and told the buyer of my concerns. He was OK with that, considering the price he paid. I was happy also... Ended up with my first tarmac - never ans issue.

As mentioned by others, the amount of trail in the bike/wheel system has a big effect. I had a custom frame back in the 70's which was built for the quick handling needs of crit racing. At 45+ mph, straight running, it would always headshake. But it was designed with minimal trail, to be a quick handing bike in tight conditions, and was mostly ridden in the 25 to 42 ish mph range. And only for crit type racing.

Let us know of any further outcomes on your part - this kind of info is always welcomed.

Camilo

This is helped for me on my bike that has some high speed wobble. Today I descended at 45 mph nice and solid by concentrating on putting more weight on the handlebars. I usually get up off the seat, and I think I was getting really far back subconsciously, as it helped me get lower. Previously I often got a little shimmy going faster than about 40 mph. I'm going to try this again this weekend to see if it really helps consistently.

My other bike is rock solid up to 50 mph no matter what I do so I might have developed bad habits, or maybe that bike just has diffrent balance so the same body position is balanced a little more forward.

This may indeed be true, but there's a LOT of things you can try: weight balance, pressing a knee against the top tube, etc.

SherpaBob

Sy Reene

hmm.. were there ice cubes in the water bottles?

Garfield Cat

Clamping knees to top tube: the clamping force from the knees does not need to be great. Just keep the touch constant. Equally important is the foot pressure on the pedals in a horizontal position. The downward pressure on the pedals is greater than the knee lock pressure.

Usually on fast descents, the ground will be uneven and cause a bit of "jump" and that's why the pressure on the pedals will be a natural position to take. It holds the rider in place during those uneven parts of the descent.

I would imagine the pro riders who take fast descents with the trunk of the body hovering over the top tube are also creating a similar effect, avoiding shimmy.

GailT

It would be interesting to know the frame size for people who have or have not had problems with shimmy. I ride a 25 in/64cm frame and was concerned about shimmy, so after reading Dave Moulton's blog on trail and shimmy, I had Waterford decrease the fork rake and increase the trail slightly on a new bike I ordered earlier this year. I have not tried it on a high speed descent yet (nothing much over 35 mph), but I plan to in the future, and I was willing to sacrifice some responsiveness in steering to avoid a death wobble. I always stand on the pedals and put my knees against the top bar when descending, seems like a good aero position and a good habit to avoid shimmy.