I’ve been looking into bike frames and trying to figure out how different geometries effect handling characteristics, and there is something I don’t get.
From what I’ve read, most bicycyles have 35-50mm rake (fork offset) because it is supposed to make the bike more stable. I don’t know how this works, but that’s what I read.
I’ve also read that most bikes are designed to have trail (the horizontal distance from where the steering axis intersects the ground to where the front wheel touches the ground) because increased trail improves bike stability. This make sense because it works in a similar way to the wheels on a shopping trolley. The wheel trails behind the steering axis so they always point in the right direction.
Here’s the thing I don’t get: If more rake creates less trail, wouldn’t rake actually make the bike less stable?
This image helps to demonstrate the relationship between trail and rake. As you can see, a fork with rake has less trail than a straight bladed fork would. So wouldn’t the straight fork therefore be more stable?
There are many factors at play, and they all affect each other, so this is not a simple subject to explain or understand.
There’s also pneumatic vs geometric trail to consider.
But simply, low trail will be more stable at low speed, less affected by cross winds, easier to ride with front loads. High trail will be nimbler & more stable at high speeds (stayer bikes have negative rake).
Search the 650B or framebuilders mailing lists, rec.bicycles.tech newsgroup etc for discussion on this topic.
so how does head tube angle factor into the handling?
a keirin bike with a steep steering angle and an old pathracer with shallow angles both have ‘long’ trail, but have quite different handling. what’s the technical explanation?
i’ve heard of some bikes tending towards the ‘death wobbles’ at high speed on the track - is that caused by a lack of trail?
There are many factors (wheel base , trail, rake . frame flex etc.) that come into play that effect handling and trial and rake is only part of the story.A good frame builder juggles all of these to achieve a certain outcome
Shimmy aka "death wobbles " is another story and involves synchronised feedback by the rider to an oscillating system. Basically it means something in the bike is too loose AND you are too tight. It can be stopped by loosing your grip on the handlebars or adding dampening mass to the bike frame by clapsing your knees on to the top tube. (another arguement for a horizontal top tube)
If you want to enter the rabbit hole, read the archives of the Kogswell Owners Group. There are so many factors at play with shimmy, it can be incredibly difficult to isolate and cure. An easy fix that sometimes works is a needle bearing headset.
that article is brilliant, exactly what i’ve been looking for. especially table 5.
i’d be interested to see the 1887 Rudge with the 90 degree headtube angle!
Thanks, that’s a very interesting article - it pretty much answers my question. It’s also pretty funny picturing that guy testing all those crazy experimental bikes with extreme or reverse rake
This bit is especially interesting:
A very stable control system responds sluggishly to perturbation, whereas one nearer to instability is more responsive; modern bicycle design has emphasised nimbleness and maneuverability. Best of all, URBIII comes out much more stable than any commercial bike. The result explains both its wonderful self-righting properties and also why it is difficult to ride - it is too stable to be steered.
So rake basically “destabilises” a bike to make it easier to steer. Ease of steering and stability are, in a sense, inversely related.
Responsiveness and stability are inversely related, but ease of steering is a sweetspot thing because I wouldn’t call unstable twitchy steering easy.
FYI modern fighter jets crash and burn without the electronic stability systems. They are designed to be unstable for responsiveness.
Crazy sports bikes also tend to the unstable side and compensate with steering dampers to tame the headshake (a bit is fun though).
