Words by Adam Newman, illustrations by Andrew Roberts
A generation ago, there were only one common type of bottom brackets. They were square taper. And they were good.
Yes, there were some variations, and people were trying hard to forget about cottered cranks, but the interface from crankarm to axle remained largely unchanged for decades. But as bicycle frames evolved beyond lugged steel, product designers realized they had more room to work with, better manufacturing methods and, most of all, more performance demands from the marketplace. And things began to change.
In the past few years, it has been the constant evolutions of the bottom bracket “standard” that has made the term an inside joke in the cycling community. There’s PF this and 386 that. What does it all mean? Well, attention spans are only so long, so this isn’t meant to be a definitive guide, but a quick primer on bottom brackets and their design.
Push the pedals and the crankarms spin, right? But for things to spin they need bearings. In this case, they’re little balls between one surface and another that let them move against each other. This allows the crankarm assembly to rotate, pulling the chain and propelling you forward.
So where do we put those bearings? On traditional loose ball bottom brackets, the bearings are independent and hand-packed with grease and spin in a cup that was threaded into the frame. (1) Then came the sealed cartridge bottom brackets. These had the bearings inside a sealed cylinder. When they eventually wore out, you threw the whole thing out and got a new one. (2) A big leap came when engineers wanted to use bigger axles to increase performance. But there was no room in the frame for the big axle and the bearings. So they moved the bearings outside the shell and attached the axle directly to the crankarms. (3) Finally, as carbon fiber frames became common, engineers realized gluing inserts into carbon frames was expensive and kind of lame, so the carbon was molded such that the bearings sat inside it and held in place with a tight fit. (4)
The interface with crankarm
Ah yes, the square taper. Nearly all high-quality bikes used this type of mount for decades. Square peg goes in square hole and everyone was happy. (5) Well, maybe not everyone, because as bikes got more capable and riders were pushing them harder and harder, it wasn’t enough. To create a stronger interface, splines were formed in the bottom axle that mated up with the crankarm. Examples of this design include Octalink and ISIS. (6)
As the persistent pursuit of performance was pursued, engineers wanted to make the axle between the crankarms bigger. But there wasn’t room inside the bottom bracket shell for properly sized bearings. The solution: move them outside the shell, where they could retain their proper size and spec. The axle was now permanently attached to drive side crank arm instead of being a part of the bottom bracket. This is the two-piece bottom bracket designed known as Hollowgram (Shimano), GXP (SRAM) or Ultra-Torque (Campagnolo).
The bottom bracket shell
This is where most of the consternation comes in. The shell is the part of the frame where the axle sits, between the crankarms. Despite the hullabaloo you hear that there used to be a common interface, it really isn’t true. Even in days of yore, there were still English, Italian, French and even Swiss bottom bracket shells, each with different threading. None were cross compatible. The English or BSA was the most common used on road bikes, with a 68 mm wide shell and reverse threading on the drive side. (1 and 2)
When the external bottom bracket was introduced, the bearings migrated from inside the shell to the outside, but were still threaded into the same old shells. (3) You could mount these on your old bike! Rejoice! Then that whole carbon craze came along and the press-fit bottom bracket came into vogue. (4) When this happened, product designers were finally like “Wait, we can change it again?” so they went to town designing frames with all sorts of new measurements. PF30, BB30, BB86, EVO386? There are many different “standards” but they are largely the same in their execution: the bearings sit directly in a channel molded into the frame, and since carbon fiber can be shaped in nearly any way, bearings can be larger and farther apart.