Wednesday, August 04, 2010

Chainstay Length & Italian Design

If one looks at Italian frame design, something of note is that chainstays are short, and their length varies by frame size.  While typical front centers, chainstay lengths, and trail measures vary between De Rosa, Pinarello, and Colnago - the differences aren't great.  More importantly, the way these and other design variables change with frame size is very much parallel between these three marques.

Generally, they have shorter chainstays that what I like.  This problem with short chainstays has less to do with handling, and more to do with shifting.  The cross-over angles of the chain in small/small or big/big gears are more than the systems are made to handle, whether from SRAM, Campy or Shimano. 

Before thinking about how your bike is able to shift to the extreme cross-overs, consider riding under full power and what happened to Schleck when his chain dropped.  As a rider, you can avoid the extreme combinations, or have longer chainstays (which reduce the angles).  I prefer the latter course because there are times when I just don't want to make a front shift - and most other riders seem to feel the same way.

As a conseqence, I'm not likely to copy italian geometry on my frames.  But that doesn't mean that there is nothing to be taken away from the designs of these masters.  One factor is chainstay length.

In managing front-center, the Italians (and many other builders), seat-tube and head-tube angles vary.  While this isn't quite what is happening, imagine a fixed wheelbase, and then pivoting the fork/head-tube at one end and the rear triangle at the other, around their respective wheel axles, to vary the length of the top-tube.

As the frames get bigger, the seat-tube tips further back (the angle becomes shallower).  This isn't done to keep kneecaps over the spindle - it's done to distribute rider weight, provide adequate top-tube length, and manage the front-center distance. 

In the process, the riders seat (and a significant portion of their weight) moves further backwards above the chainstay, and the seat-tube also comes closer to the rear wheel/tire.  Lets look at this a little more closely.

The distance between the seat-tube and rear wheel is a function of the seat-tube angle, bottom bracket drop, and chainstay length.  If the seat-tube leans far enough backward, it will rub on the tire.  This can be avoided by raising the bottom bracket, lengthening the chainstay, or bending the seat-tube to go around the wheel.

Some builders have choosen, in certain circumstances, to bend the seat-tube.  This works, but causes the upper portion of the seat-tube to be at a shallower angle than if the seat-tube was straight.  This can cause somewhat extreme changes in fore/aft positioning as the seat is raised/lowered.  Hence, I'm among those who consider this a suboptimal solution.

Raising the bottom bracket allows the seat-tube angle to vary without varying the angle between the chainstay and seat-tube - hence ensures clearance for the wheel.  But, it does so at the cost of changing the handling feel of a bike.  Neither higher or lower is better - but BB position is part of the overall design characteristics which determine how stable/responsive a bike will be.  And, for a production model, the goal is to maintain these characteristics across the range of sizes.  Therefore, raising the BB for larger frame sizes doesn't make much sense.  There are some other factors driving front-center distance and frame rigidity which further argue against raising the BB.

So.... to keep adequate space between the rear wheel and seat-tube, larger frames need longer chainstays.  However, there are two more considerations to argue in favor with varying chainstay length with frame size.  Otherwise frames that used longer chainstays, wouldn't need to vary the length over size.

First a simple argument: on well designed frames, wheelbase grows with frame size.  Front center also grows with frame size.  So it makes sense for the chainstay length to also grow in order to maintain appropriate front to rear balance.

Second, as one's seat comes closer to being directly over the rear axle, the ride of ones frame becomes harsher.  This has nothing to do with frame stiffness, and everything to do with the geometry of motion, related to going over a bump.  If you saddle is squarely over the rear axle, and the rear axle goes over a 1" high bump - then your saddle will rise by 1".  If the saddle is half way between the front and rear axles, then it will only rise 1/2" when the rear wheel goes over that 1" bump.  While these are extreme postions for the saddle, it shows what a big factor rear wheel versus saddle position really is in determining rider comfort.

Net net, chainstay lengths should vary over frame sizes, and the famous Italian builders demonstrate that every day in their frame designs.  Its funny, isn't it, how complicated one little measurement can be in determing the overall ride characteristics of a bicycle.

1 comment:

Anonymous said...

I agree with every word!
thanks for putting them on the web!