February 26, 2015 6 Comments
Mobility is Work
The more efficient the mobility, the more successful is the organism. Salmon and gulls rate highly but humans aren’t far behind. Our ability to move with relatively low energy expense is key to human history. Bipedal motion, sweating, a springy foot, and other physiology made it possible for indigenous hunters to simply walk their prey down over a period of days.
We demand similar efficiency from bicycles and they deliver. After all, look who made them. Bearings, of course,are key. Keeping their friction low is one of our highest priorities.
Rarely seen anymore, some early bikes used bearings with spherical races that would continue to spin even if the axle was bent. Today we try and keep axles from bending but there’s no denying this is a clever idea.
Before cartridge bearings, adjustable loose ball bearings ruled bikedom. Shimano and Campagnolo still use them in many of their high-end hubs.
A cup and cone system features angular rather than annular contact.
Why angular?
Most think it is because of regular side loads that a hub must see. True on a 3 or 4 wheel vehicle but single track side loads are relatively tiny and statistically rare. The best reason is how much easier it is to get the bearing to a sweet spot of adjustment where friction and side play are low.
Side play is important not only because of noise and squirrelly handling but the potential for harmonics. Vehicles that are light (like bikes and planes) are vulnerable to harmonic vibration. That’s why test flights, before the era of computer modeling, were so suspenseful. Many planes (and helicopters) failed because harmonic vibration that could not be dampened shook the structure to pieces. Cars and locomotives are less risk because it takes so much energy to vibrate larger masses.
We’re all familiar with bicycle speed wobbles. They rarely cause a nasty crash, but take worrying concentration to control. Bearing play will aggravate, if not promote, speed wobbles. So we want it minimized. The angular contact bearing can be adjusted whereas the annular is manufactured to a given fit and cannot. Today’s annular bearings are made so well that when located at either end of a hub, there is often no play to be detected. Small play can be made invisible by slightly side-loading a pair of annular bearings with only a small friction consequence.
None of this compares with the friction from grease and seals. Chester Kyle conducted tests that should have woken the world to this factor. In one case, he swung a 1o-pound weight on a pendulum from a hub with sealed-cartridge bearings. He then removed the seals and grease, and performed the test again. The pendulum swung 20 times longer. He estimated seals and grease were responsible for a 3 to 5-percent increase in total bike rolling resistance. Clearly our choices affect performance; bearing friction is not negligible. Since we all use similar bearings, the subject is overlooked; like complaining about air density which is pretty much a given. But bearing friction is no longer taken for granted.
Look Out Ceramics
The latest chapter in bicycle ball bearings has begun. And it is being ushered in by the widespread adoption of ceramic bearings by professional racers. These spheres of silicon nitride, not a true ceramic, are lighter, harder, smoother and rounder than the best steel bearings. Combined with high-grade chromium-steel races, they're known as hybrid bearings and are huge news throughout cycling and in other sports. Invented nearly 25 years ago by Swedish bearing giant, SKF, these units are now finding their way into bikes.
Whether the expense is warranted for your own riding is not the issue. They simply illuminate how important bearings are. A perfect object lesson.
The ceramic bearings hiding in this bottom bracket offer huge performance gains - reducing rolling resistance by up to 4 percent. That's like doping without the crime.
lnline skaters, whose wheels revolve thousands of times per minute, were some of the earliest adopters of ceramic bearings. Next came F1 (up to $250K per car), and now cycling. Ceramic bearings are everywhere in top UCI road and off-road competitions. Retrofitted to hubs, bottom brackets and pulley wheels, the bearings represent the latest performance enhancement, and unlike many injected "enhancements," they are completely legal.
The benefits are twofold: first, the actual ceramic ball is three times harder than steel. While this leads to greatly extended life spans, it also reduces friction. Believe it or not, all balls deform slightly under load. This is the major, microscopic source of friction in a bearing. Less deformation equals lower friction.
Like the proverbial truffle in cooking, however, it's a waste to use ceramic balls if the other ingredients are inferior. So the second benefit stems from the better seals, grease and retainers that harder, smoother ceramic bearings allow. For a higher price, you can use extra—low—friction seals that aren't available on industrial bearings. At the same time, weatherproof grease can be replaced with extra—low-viscosity lube. Most bearings use a retainer to separate the balls from each other. If balls bump, there's extra friction because they're rotating in opposite directions at the moment of contact. Compared to metal retainers in traditional bearings, some ceramics have low-friction retainers made from special polyamides.
Many of these non-ceramic improvements (better/fewer seals, lighter grease/oil, etc.) are available to anyone and as Chester Kyle pointed out 25 years ago, they greatly reduce friction. But when they are combined with ceramic balls, the bearing is taken to a whole new level.
The next and final chapter of this bearing excursion details how they affect wheel building. Understand what the hub is doing and don’t let bearings interfere with your building.
November 02, 2021
Interesting that you mention the Chester Kyle test, which I believe was published in Bicycling years ago. If you recall, letters were written to point out that the test was fundamentally flawed and a new test was done. The issue was starting torque (breakaway torque, can’t remember) was very high, but once the bearing was rotating it became very efficient. The pendulum test gave bad numbers because it comes to a stop at every swing and the high starting torque repeats every swing.
I believe he retested, and found much less difference.
It’s possible I may still have the issues, but the desire to find them is very low.
Also, dirt-rag virtually plagiarized the original article years ago without crediting the source, or reading the following letters and tests.
Admittedly you need the full collection, it was over many months until it was over.
November 02, 2021
Ric – Excellent work on this article. Have you seen anyone attemp to integrate the fork and hub, ie., buildi-in the fork drop outs to the hub so that the drop outs are directly over the bearings and resulting in no pre-stressed tension (deformations) of the axle? Tom
November 02, 2021
Ric – Looking at the rear wheel, by integrating the drop outs into the bearings in a similar fashion like I presented for the front wheel, it looks like it would be possible for even dishing of the wheel and to move the cassette to the other side of the drop out (cantilever the cassette). The frame and wheels would have perfect symmetry which Nature’s way. With oil lubricated bearings taking direct loads, a narrower frame symmetrical with the wheels, and a cantilevered cassette, I wonder how such a bike would ride. If it not scary performing, it would definitely have a cool look. Plus, the chain would be easier to maintain and you could replace cassette cogs without taking the wheel off. Fun to imagine. Tom
November 02, 2021
Ric – I would like to further detail some things on my previous message. The cartridge bearing inner ring (bearing could be radial) would be press fitted to the outside diameter of the hub body. The outer ring of the bearing would be the fork drop out, in other words the outer ring and 3 inch fork tip section would be one piece. There’s room to play with on the optimal diameters of the bearings and spoke flanges. The impressive foldable bikes I see these days convince me that a simple, light and reliable quick connect system on the fork could be executed. Thanks. Tom
November 02, 2021
Ric – I was thinking no axle and no quick release and no clamping force. There would be a hub with cartridge bearings pressed on at the ends with the hub spoke flanges directly adjacent. The fork tips would be integrated directly over the bearings. The fork tips would quick-connect to the fork at the low stress area, about 3 inches up the drop out end of the fork. The hollow volume in the 3-inch section of fork could be filled with oil to feed the bearings. Anyway, I’ve been having fun thinking of this and your article triggered my thought provoking. Thanks. Tom
Comments will be approved before showing up.
Dominik
November 02, 2021
Thank you for this in-depth article about ball bearings.
As you mentioned the resistance caused by grease: What type/brand of grease do you recommend when overhauling ball bearings (classic cup-cone type).
And: Can ceramic ball bearings be used in old hubs? Will this improve the performance of old hubs?
Thank you for an answer and keep up the wonderful work you do
Dominik