Ceramic Speed Pulleys
We've been around cycling long enough to remember ads for Bullseye pulleys. The gist of their ads was that if you used their pulleys in a 100mi road race, you'd be ahead of where you would have been without them. Something like 75 feet ahead, if we remember the old ads correctly.
While Bullseye pulleys are still around, ceramic-bearing pulleys are the next great leap forward. We decided to test out Ceramic Speed pulleys, the most storied of the ceramic pulleys on the market. Ceramic Speed is a Danish company that makes ceramic bearings for a variety of applications. We've heard that a number of Pro Tour racers have installed their bearings not only in their derailleurs, but hubs and bottom brackets as well. Most of these riders are under contract to ride derailleurs from Campy and Shimano so Ceramic Speed can't tell us who those people are. Danish…hmmn. There is a certain demanding Danish director of a Danish-registered pro team…but since 2006, CSC sponsor FSA has been selling ceramics, albeit working some of that time with Ceramic Speed. Ceramic Speed is not embarrassed to tout Dane Michael Rasmussen, also an equipment geek, as an enthusiastic user of their bearings. .
Ceramic Speed says they do more than just press in their hybrid ceramic/metal cartridges into BBB or Tacx pulley wheels. They start with exceptionally round silicon/nitride balls and put them in a low-friction retainer and exceptionally-smooth races. They say that steel balls are the weak link in a bearing because dirt will score the steel ball and that starts the process of breaking down the ball. The ceramic balls are so hard; it crushes the dirt without scratching. They also claim the stock lube needs to be thinned a bit before the pulleys run as smooth as they can, and that occasional re-lubing with a thin lube is most of the service you need to do. To see how to achieve their recommended twice-yearly service, check out this document.
Because we want to keep up with those who lie about their whereabouts to foil WADA, we pulled the Campagnolo pulleys out of our Record rear derailleur and installed the Ceramic Speed pulleys. Campy pulleys spin on brass bushings, sometimes called sleeve bearings. They've been using this technology forever. The technology has worked pretty well over the years, though it has probably been refined a good bit. One advantage the bushing design has is that it works in all environments for a long time. The drag of frozen ball bearings could be pretty great, but this design should never freeze, though it might not be the smoothest on the market and needs periodic regreasing. .
When Bullseye's Roger Durham tested bearings back in the late 70s, sleeve bearings were de rigueur in derailleurs. In fact, Durham's claims were based on testing sleeve bearing pulleys. He took a pulley, cut a groove through the teeth for a string to be set in the pulley teeth. He put the pulley on a lathe, weighted each end of the string so the pulley would have the same tension as a derailleur in action with a chain wrapped around the pulleys and spun the lathe at 200rpm. He put a postal scale underneath the weighted end, which was lower to determine drag. There were ads with different ways of explaining his data. One is that over 100miles, the saved energy could lift 100lbs 120ft. Another is that a 150lb rider could do an extra 75-foot climb with the energy saved, and so on. His pulleys were another example of “free” speed. Install them and you'll go faster for the same effort.
Time has moved on. But pulleys haven't completely. Bushings are still being used in a large portion of derailleurs on the market. The exceptions are Shimano Dura Ace, where both the upper and lower pulleys are sealed, Ultegra, where the lower pulley is sealed, SRAM's Red rear derailleur has ceramic pulleys, while their Force and Rival rear mechs have standard sealed bearing pulleys. Campagnolo still relies on the sleeve bearing.
We asked Campagnolo USA about putting in Ceramic Speed pulleys. They warned us that swapping out Campy parts for after-market items voids the warranty if there's a failure that could be attributed to the aftermarket part. They also said, “(we) haven't changed yet because it's a tried-and-true design that seems to last a long time and are loath to change. Weight is similar or lighter than anything with a ball bearing.”
Our used Campy pulleys weigh in at 19g for the set. Our new Ceramic Speed weighed in at 20g. Point taken. The Campy jockey wheels are also easy to find after-market and cheap to source for Campy when building new.
But that 1g savings should be obliterated by any savings in efficiency. And since the Campy design was eclipsed for efficiency about 30 years ago, it is entirely possible that the newer designs are better still.
One design issue to overcome is the upper “floating” pulley. Most derailleurs that are designed work with indexed shifters come with an upper pulley that has a small amount of sideways pay to account for less than perfect shifting, cable stretch, swapping out wheels, cassettes, etc. Most sealed upper pulleys don't float. The Ceramic Speed upper pulley we have measures approximately 1.5mm wide (we don't have more precise calipers) while the Campy upper pulley measures 2mm. It's hard to believe this is enough to make up for no float, but we could be wrong. And, with a month of swapping wheels behind us, we do seem to be wrong about that. We had been concerned that if the non-floating pulley wasn't centered under a cog, the chain would have added friction rubbing against the pulley wheel. We can't seem to find this friction, either.
There are a few different ways to test how much drag is reduced by switching pulleys. The easiest, and one that can be performed by anybody is the backspin test. Spin the cranks backward and see how they fly. By employing this test, we see there is less drag after the Ceramic Speed pulleys were installed. The cranks spin back for longer and the spin makes a higher pitch noise as well. A second way is doing repeated runs in a controlled environment carefully measuring with a power meter. We did not do this, partially because we've read other people's results, which confirm what we've long expected. The power savings, even at low power, should fall within the margin of error of a power meter. Even at +/- 1.5%, there's a 3w range at 100w and these might save 1w in ideal conditions.
We've read FSA's lit, which claims that the bearing maker SKF tested ceramic-bearing pulleys and had the Danish magazine Cykel Magasinet repeated the test. Both found that Dura Ace pulleys consume .78w at 500rpm, ceramic pulleys consume .06w. 500rpm was chosen because that's what a pulley might spin if you're pedaling in the big ring.
Since DA pulleys have sealed ball bearings and Campy wheels have sleeve bearings, there's a chance Campy pulleys have more friction resistance, making the energy savings greater. DA pulleys have 11 teeth, while Campagnolo's have 10, the difference means that for a given cadence, Campy pulleys spin even faster, which could mean more drag.
Engineers and those who like to play at engineering -- at least those we know that aren't affiliated with a company selling ceramic pulley bearings -- have been skeptical of the claims made. “Unvalidated” is one of the nicer comments we heard. A less kind comment is “a waste of money in bicycles.” Here's more detail from one of the engineers we spoke to; “(t)he difference is probably not measurable under normal conditions. Just because a bearing feels 'smoother' when twirling it with your fingertips, don't assume that that means it runs with less friction under load. The only way to tell for sure is under instrumented, controlled conditions with all else being identical…Remember when you're twirling a sealed bearing, that you're feeling the seal drag more than anything else. In addition, remember the environment that these have to operate in, which a bushing may end up being better than either rolling element at the end of a rainy race.”
Fair enough. Twirling doesn't count. Backpedaling doesn't count. The lube and seals might be more of an issue. The benefits can't be measured with a bicycle-mounted power meter. And the numbers generated might have come from a test where the parameters were designed to maximize the advantage ceramic pulleys have. It's almost enough to make us want to put 'em on eBay and find the next sucker out there. We're not ready to go back to Campy pulleys, but probably something a bit less flash, like Tacx sealed-bearing derailleur pulleys.
But before we did, we thought we'd visit the Analytic Cycling site to see how much difference the advertised advantage can make.
We went to the page that calculates speed on a given power. We picked 250w (it's the default) assuming 75kg bike-and-rider weight on a flat asphalt road and moderate frontal area and sea-level air density. That 75kg rider putting out 250w will travel at 11.23 meters/second and cover 40.428km in an hour. We decided to see what 250.5w, less than the tested difference between stock DA and stock Ceramic Speed, could do. 250.5w rides 11.24 m/s with the same rider and conditions. 11.24 m/s equals 40.464km in an hour, an extra 36m. In a four-hour ride, that adds up to 144m, over a one-and-a-half football field advantage. Not insignificant.
But few of us 75kg bike and rider types are putting out 250w for four hours. 150w might be more accurate. Using the same parameters as above, here is how the model calculates the difference. 150w is 9.29 m/s is 33.444km an hour or 133.776km in four hours. 150.5w is 9.30 m/s is 33.480km an hour or 133.92km in four hours. Still 144m ahead. And a greater relative improvement! Yikes.
So there's the rub. Ceramic Speed makes some crazy-expensive pulleys that, if they have an advantage at all, is probably too small to measure anywhere but in a lab. But there could be impressive real-world savings despite the low drag in standard-equipment pulleys to begin with. Looking at the Analytic Cycling models, we can understand why pro riders might be doing the secret pulley-swap. Even if we were the lowest domestique on the worst Pro Tour team, we'd get these pulleys on the tiny chance they'd offer a tiny respite from ignominy. As long as we have these jockey wheels in our rear derailleur, we want to believe the advantage is real, though we'll just as surely hope the pulleys are merely expensive hype when we don't.