- News
- Reviews
- Bikes
- Accessories
- Accessories - misc
- Computer mounts
- Bags
- Bar ends
- Bike bags & cases
- Bottle cages
- Bottles
- Cameras
- Car racks
- Child seats
- Computers
- Glasses
- GPS units
- Helmets
- Lights - front
- Lights - rear
- Lights - sets
- Locks
- Mirrors
- Mudguards
- Racks
- Pumps & CO2 inflators
- Puncture kits
- Reflectives
- Smart watches
- Stands and racks
- Trailers
- Clothing
- Components
- Bar tape & grips
- Bottom brackets
- Brake & gear cables
- Brake & STI levers
- Brake pads & spares
- Brakes
- Cassettes & freewheels
- Chains
- Chainsets & chainrings
- Derailleurs - front
- Derailleurs - rear
- Forks
- Gear levers & shifters
- Groupsets
- Handlebars & extensions
- Headsets
- Hubs
- Inner tubes
- Pedals
- Quick releases & skewers
- Saddles
- Seatposts
- Stems
- Wheels
- Tyres
- Health, fitness and nutrition
- Tools and workshop
- Miscellaneous
- Tubeless valves
- Buyers Guides
- Features
- Forum
- Recommends
- Podcast
Ceramic bearings the pros and cons Nov 2018Should you buy ceramic bearings? Expert opinions polled
[This article was originally published November, 2015 and updated November, 2019]
Despite the latest advances in road bike design and technology, your bicycle still relies on the humble ball bearing to ensure it all runs smoothly. There are bearings in the wheel hubs, the headset, bottom bracket, pedals and jockey wheels, and they're commonly made from steel. There has been a lot of hype about ceramic bearings in recent years and many people claim they offer a performance upgrade. But what are ceramic bearings and what are the pros and cons?
While regular bearings are made from stainless steel, ceramic bearings are made from ceramic silicon nitride (Si2N4). Rolling resistance is the key trump card promoted by ceramic bearing fans. Because ceramic bearings are rounder with a smoother surface and more uniform size, friction is reduced and that can contribute to a less energy required to turn the cranks or spin the wheels. A ceramic bearing is also harder than steel bearings (up to 30%), which improves durability, and they also don’t rust so less maintenance should be required.
Most sealed ceramic bearings are actually hybrid ceramic bearings, which combine a steel race with ceramic ball bearings. Full ceramic bearings use ceramic races, which can be lighter and provide the lowest friction, but come at a durability cost. Unsealed ceramic bearings can be used to upgrade components that don’t use cartridge type bearings, like cup and cone hubs.
It’s in the professional peloton, a place obsessed with marginal gains, that ceramic bearings have become popular in the past couple of years. And naturally, where pros lead, amateur racers and sportive cyclists follow, keen to take advantage. Should you be following in the footsteps of the pro racers then, and upgrading your bike with ceramic bearings?
CeramicSpeed is a Danish company, founded in 2007, that specialises in supplying high-quality ceramic bearings to the cycling industry, and is a favourite of many professional cycling teams - you don’t have to look far to spot the telltale company sticker on any frame or hub fitted with the bearings. It’s seen a massive shift in the last two years with more customers keen to take up the ceramic advantage. The advantage, according to the company, is a saving of up to 9 watts with CeramicSpeed bearings in the hubs, jockey wheels and bottom brackets compared to a set of standard bearings
“The main advantages with CeramicSpeed bearings over regular bearings are two-fold,” explains CeramicSpeed Managing Director Martin Banke. “Longevity of a well-built high-quality ceramic bearing, in many cases, can be up to 10 times longer than commonly used stock bearings. The 'rule of thumb', as we like to call it when built well, and of high-quality materials, a ceramic bearing should always be able to outlast and outperform a steel bearing.
“The second advantage of ceramic bearings over stock steel bearings is their performance under load in reducing drag. Performance cyclists are performance driven and all data shows that the best performing bearings for reducing drag are ceramic bearings.”
That, in theory, should mean less energy is required to turn the wheels or cranks. Add the lower weight and improved durability, and why wouldn’t you run ceramic bearings? But are all ceramic bearings the same? Of course not, CeramicSpeed is keen to point out that not all ceramic bearings are made equal, and it tells us that a ceramic bearing built poorly with low-quality materials will deliver very poor longevity.
“A lot of bearings can spin well in the hand but that by itself is not enough,” says Banke. “It is really about the performance of a bearing under load, both drag reducing performance and lifetime performance. We are now starting to see a 'shift in the seas', a common understanding that ceramic does last and is very advantageous if of a high enough quality.”
When the benefits sound that good, you’d think all component manufacturers would be embracing ceramic bearings, right? Highly regarded British component manufacturer Hope Technology reckons the efficiency savings are simply too negligible to make them worth the increase in cost.
“We have looked at ceramic bearings in the past and talked them over with our bearing suppliers,” explains Hope’s Alan Weatherill. “They do run with less friction, which offers a significant advantage in industrial applications running at 20,000 rpm. A tiny percentage reduction in friction here can equate to a worthwhile power saving, but when you're only turning at 300 rpm, as you do on a bicycle a small percentage increase in efficiency will make a negligible change to your power output. Certainly not worth the significant increase in cost."
Leading industry wheel dynamics expert and CEO of Edco Wheels , Paul Lew, backs this up and reckons it makes ceramic bearings a poor choice for hubs and headsets, also adding that they offer no weight savings and are only beneficial in environments where high rpm (revolutions per minute) are required.
“For bottom bracket applications, the maximum sustained rpm may be 130,” explains Lew. “For wheel hub applications, the maximum rpm may be 500 – 600 revolutions per kilometer (depends on wheel/tyre diameter). The maximum rpm values in cycling are far below the typical ideal rating for ceramic bearings which is 10,000 rpm+.”
It’s clear that Alan and Paul agree that the factors that make a ceramic bearing well-suited to industrial and medical equipment applications, and high-altitude operating drone motors (Paul does a lot of work designing drones) where the rpm is high, the load is low and the operating conditions are clean, are factors that mean they’re not suited to cycling.
“Ceramic bearings are beneficial in environments not requiring grease lubrication,” says Lew. But a bicycle is expected to cope with a vast range of conditions, rain and dirt, and where maintenance schedules may be less than optimal, and the last thing you want is to ride bearings without grease. It’s this requirement to cope with the conditions common to cyclists that offset the promised lower rolling resistance of a ceramic bearing, according to Paul Lew.
“The rolling resistance of a ceramic bearing compared to an ABEC 3, 5 or 7 steel ball bearing is offset by the resistance of the grease,” he says. “In order for a ceramic ball bearing to out-perform a steel ball bearing, grease is not an option. Does this mean I should run my ceramic ball bearings dry or with light oil? Yes, but you won’t like the result in an environment where the bearings can become contaminated. If you run your bearings dry they will feel gritty and rough.”
Hope’s Alan Weatherill concurs with Paul Lew’s conclusion that ceramic bearings are not suited to the demands of cycling and says their suitability to industrial machinery doesn’t necessarily provide the performance benefit for cyclists that many people and companies claim they do.
“Another issue with using them [ceramic bearings] on bicycles is their hardness,” says Weatherill. “While this again is an advantage in many industrial applications, it's a major drawback on bikes. The shocks from hitting potholes and other road blemishes impact the hard ceramic balls into the softer steel races commonly used. This dent in the race is then felt when the bearing is rotated, giving you rough bearings.”
The less polite stance
Weatherilll and Lew are measured in their scepticism about the overall benefits of ceramic bearings. Rather more forthright is Sachin Hambini of Hambini Engineering. An aerospace engineer who has built up a cult following for his scathing criticism of the quality of manufacturing in the bike industry, especially where press-fit bottom brackets are concerned.
Here's what he has to say about ceramic bearings. Warning: adult language.
How much?
Then there is the fact that ceramic bearings aren’t cheap. CeramicSpeed’s BSA Road external bottom bracket for threaded frames costs £298. A Shimano Dura-Ace 9000 bottom bracket cost £39.99, much less if you shop around. A CeramicSpeed bearing upgrade kit for Campagnolo and Fulcrum wheels will set you back £120, and Zipp charges £194 for a CeramicSpeed bearing kit for its wheels. That makes upgrading to ceramic bearings a serious investment, fine for a professional cycling team, less so for a privateer racer.
The premium for ceramic bearings is high then, and their advantages, while looking promising in an ideal world, appear to stack up much less in the demanding environment that a bicycle is expected to perform and survive in. So should you choose ceramic bearings? We’ll let Paul Lew have the last word.
“Although the re-selling markup/ margin for ceramic bearings is significant for manufacturers such as Reynolds, and they could represent a profit-center for the brand, we choose not to offer them because they don’t improve performance, and they represent a consumer cost that we can’t justify, and that’s contrary to our value system," he concludes.
David worked on the road.cc tech team from 2012-2020. Previously he was editor of Bikemagic.com and before that staff writer at RCUK. He's a seasoned cyclist of all disciplines, from road to mountain biking, touring to cyclo-cross, he only wishes he had time to ride them all. He's mildly competitive, though he'll never admit it, and is a frequent road racer but is too lazy to do really well. He currently resides in the Cotswolds, and you can now find him over on his own YouTube channel David Arthur - Just Ride Bikes.
Latest Comments
- muhasib 9 min 41 sec ago
I wonder if the company who did the website redesign got paid before Wiggle went into administration?
- brooksby 14 min 30 sec ago
No, you're not missing anything. I've just re-read the original article and I clearly just presumed they would hand in their licence…
- hawkinspeter 25 min 10 sec ago
That still annoys me about the BBC - iplayer only keeps some programmes available for a short time, so we have to rely on people making copies to...
- hawkinspeter 41 min 4 sec ago
I just played around a little with Google Maps with directions going from Hanham/Warmley toTemple Meads and that didn't choose Crews Hole Rd. The...
- nextSibling 1 hour 15 min ago
The point of limited-run models like these isn't to make a profit. It's a marketing exercise to get the brand in the media via filler articles that...
- FionaJJ 1 hour 46 min ago
Seems like a flaw. I wonder who decided that? Surely it's in everyone's interests to come up with an agreement that a portion of the car-park be...
- chrisonabike 1 hour 51 min ago
Don't forget the hour record!...
- brooksby 2 hours 3 min ago
https://metro.co.uk/2024/11/25/uncle-picture-welcome-sign-car-crash-kill...
- pockstone 2 hours 36 min ago
See my comment on the live blog...I think it's pretty clear who the real baddie is.
Add new comment
80 comments
SKF bearings are probably the best out there. High quality, no hype, no BS.
My dad works for SKF (has done for about 40 years so slightly biased view), a lot of their bearings are actually made in the UK. I asked him about ceramic bearings for bikes and his face was a bit of a picture, apparently they test different materials quite frequently and while yes the ceramic bearings they've compared against do offer less friction than steel, the difference isn't linear, at anything under a few thousand rpm apparently it really is negligible (admittedly I don't think these were cycling specific ones but I don't think there can be that much difference as the basic design is similar)
If you've got the cash to spend then fair enough I guess, but I could upgrade all of the bearings on my bike to high grade steel ones for less than the cost of ceramic ones in one area. As much as I like free speed, I really can't justify the cost vs my performance, I'd rather put the money towards a coach if I was that bothered.
SKF also supply bearings to almost all the formula 1 teams and only a few of those are ceramics in the highest speed applications (inside engines etc I believe).
I just fitted some ceramic bearing tacx t4025 jockey wheels to my commuter, £25.50 from ebay, more for longevity, the last tacx jockey wheels were the steel bearing ones, which seized up long before the teeth had started wearing. My first set wore the teeth.
The alloy ceramics on the good bike have lasted a very long time, so I thought I might as well. Was tempted by the stainless ones, but she's worth it.
The light colour helps to show me the filth and that it needs a clean.
I have some of those. They are very good for the money, Friction Facts tested them and found them to be second only to Ceramic Speed for about a tenth the price. Had them a few years and one has got a bit gritty.
If hard bearings dent the races on bumpy terrain does that mean a frameset with an integrated headset is going to be ruined by a rough road or two?
The Rovel CLX 40 Disc wheels from my S-Works Roubiax are in the LBS this week having the Ceramic Speed bearings changed out to decent steel ones. The front ones only lasted about 5,000 miles before they felt like rocks going round.
Ceramic Speed want £400+ for a new 5 bearing set, a set of good quality steel ones is less than £100
At the moment I've left the BB and headset ones in place as they seem ok.
I build my wheels and tried a set of hybrid ceramic bearings. They lasted me about a month. My best set of wheels has SKF ABEC 5 steel bearings. They are my everyday wheels and are going for 2 years now.
I've been using SKF since around 1990, campag used to have them in their bottom brackets and even some other lesser known names like THUN. it was one of these THUN BBs 9with the same half moon interface on the cups) that is amongst the smoothest and with least resistance BBs i've ever come across. I got a semi ceramic (most Si3N4 aren't a full ceramic bearing but a coating on a steel ball) BB inclyded with an FSA K-Force light a few years ago, honestly it was garbage, it had had maybe 50-100 miles tops but so much drag compared to even the octalink BB on my daily.
There's a bloody good reason that shimano hubs do so well longevity wise and indeed the campag hubs of yesteryear, I have an ofmega hubset from the early 80s on some mavics for my 50s bike that are sweet as a nut and still running an early 90s sansin (AKA sunshine) sealed hub on an MA2 that has seen 20k
Marginal gains in some instances maybe but for most it really isn't worth it.
Yesteryear ? just dont buy inferiour fabric wheel - The Record hub still has steel bearing and races.
Campagnolo (fulcrum) is kind of interresting in this way, the more you pay for your wheel the less sound construction you'd get. But hey, marketing works.
Cost / marketing rating:
1) Shamal - ceramic bearing and aluminum spoke
2) Eurus - steel bearing and aluminum spoke
3) Zonda - steel bearing and sss spokes
Sound construction rating:
1) Zonda - steel bearing and ss spokes
2) Eurus - steel bearing and aluminum spoke
3) Shamal - ceramic bearing and aluminum spoke
reg the ceramic USB bearing yoou can upgrade them to the best steel balls on the market in 5 minuttes with the sparepart HB-RE023
The cult bearing - you'd need the record cups and cones as well: HB-RE124 & HB-RE022 (and an expensive Cyclex tool to press in the cups. 30 minuttes)
But hey we beeing bombed with advertising infomercials written by journalist with abseloutely no engernering background for 15 years som everyone assumes that a ceramic bearing is better for bicycles. Unfortunately only few manufaturs maufactures their steel bearings with the higst grade steel bearings. Campagnolo still does in their entire range (except for products with USB/Cult), Shimano does but only in their Durecae/XTR range, DT & CK does , - assume small manufctures like Hope and White does too (otherwise they die).
Compare the factory shipped bearings in a cheap Cmapagnolo wheel to a Shimano or Mavic wheel. The cheap nonseries shimano wheels are horible, the Ultegra evel are OK but does not compare to any Campy hub (or dureace).
Much like the 'science' of nutrition or tyre sealant people spin things (pun intended) any way they want. Adding to this most user info is purely emperical.
I did appreciate though that Martin Banke talks about the performance benefits under load:
"The second advantage of ceramic bearings over stock steel bearings is their performance under load in reducing drag. "
That's what I wonder about. Seeing the demos of the little bearing spinning forever appears pointless to me. I want to know what difference is there under load? Really it's not just about ceramics, but what about as bearing start to wear and you notice a bit of a grit.. or anything less than perfect. Say I hand spin the wheel and it comes to a stop sooner than when the bearings are new; what does that really mean when I'm riding? No differnece? Or 10 watts? more?
But I can't imagine this isn't easy to figure out. I see so many aerodynamic windtunnel tests; why not tests with bearings under load. All you'd have to do is take virtually anyone's bike that's been on the road for a few month and test it out. Then replace the bearings, either with new ceramic, or steel, and do the same test, and see if there really is an improvement. Compare the new steal to the new ceramic, and different manufactures, and provide some real evidence. Don't just spin a little bearing in your hand and pretend it's proof!
My guess is there is some marginal differnce. But marginal, as a racer, can be huge! And I suppose it's impossible for tests to quantify the psychological gains you get too. If my chain runs dry my bike feels slow. If I lube it up in the middle of a ride I feel much faster. I sincerely believe that a lubed chain vs a dry chain is like 10 or 20 watts faster. Is that real? Is it purely psychological? Or both?
I was checking this article out though because of another issue; brake pad drag. I went up the Combe Lane climb near London on Sunday and I was one full minute slower than my last time up a few months ago. I was carrying a couple extra things on the bike, but my weight is more or less the same. Did I miss my morning coffee? Was it the extra hours spend on the bike during the week? The garmin popped up with the segment an I could see the seconds I was falling back to my PR with every pedal stroke. Yet I didn't feel that weak: I didn't feel like I should be a minute slower!
C'est la vie! But I was in the garage last night and making some adjustments when I hand spun the rear wheel, and there was some brake pad drag (disk brakes). I estimate the wheel came to a stop within 10 revolutions rather than maybe 30. Did that make a wattage differece? Under load would it be more? Then I spun the front one and wow, that came to a stop within 3 revolutions! There was a ton of drag there! But is it really a significan difference under load on a slow climb? Will that account for my full 60 second difference on a 5min climb?? (I hope and wish!)
This happens all the time with MTB disk brakes. Most often not as bad as my bike is now but the brake pads can easily be hitting the rotor and causing a minor drag. But even that small drag is probaby way more significant than the difference between the average ceramic vs steel bearings. I really wonder how much all this makes a difference though, under load?
I've been using a ceramic hob for years but its not improved my cycling at all.
Latest related review on the Tacx ceramic bearing jockey wheels is here: http://road.cc/content/review/199145-tacx-t4035-jockey-wheels
Plenty of opinions on bearings (backed up by years of research) from Keith Bontrager in this fascinating Velocast interview. Well worth a listen:
http://velocastcc.squarespace.com/tech5/2016/4/23/episode-34-keith-bontr...
Interesting article.
I would definitely use if I was a professional racer.
For someone who just enjoys riding, keeping the price realistic and reliability: I'll stick to steel bearings and grease.
I think I'll go back to an internal bb next time, for my lightweight commuter bike.
My thinking is, smaller spindle, less stiff (in theory), bigger bearings, much longer lasting (in reality).
The external bearing design is great but only for the first 1000 miles or so.
@kev-s
you make a very good point about cheap bearings. Its something I saw in mountain biking full suspension frames some years back, they used to come fitted with quality bearings on the pivots, and then they started turning up with unbranded generic bearings that would cr*p out after 2-3 months of riding, or actually feel graunchy from new!
Obviously a massive cost saving for the frame manufacturer if saving $40 on a set of bearings and selling 30,000 frames a year!
We used to refit the frames with quality Japanese or German bearings sourced by an industrial bearing supplier in Park Royal, and these would last several seasons without issue. Well worth the £5-6 each bearing cost.
Mebbe worth adding that there's been a big market in counterfeit bearings for the last handful or years or more. Tbh I dunno how you'd descriminate, but google is your friend.
As for ceramic bearings or buggerall lube - the no of watts you'd save is down in the noise..
When it comes to wheel bearings your best bet is to replace the cheap chinese bearings that come in the majority of sub £500 wheels these days
Rather than pay out loads for ceramics try a nice set of japanese skf or nsk bearings from your local bearing supplier, these will outlast any chinese bearing, will cost around £10 more per bearing than the chinese equvilant but no way near as much as a ceramic one would be
These bearings are used in industrial machinery and in manufacturing process lines, conveyor bets, rollers etc.. so are hard wearing and smooth rolling, designed to last
SKF = swedish (Svenska Kullager Fabriken)
Ezo = japanese
SKF is Sweedish. I know because I work for them. And yes, when my endro bearings that come on my bottom bracket or wheels give out, I switch them to SKF bearings.
As for the denting Paul Lew is talking about, that is called brinelling. It also happens to standard bearings as the ballset is typically the hardest object in your bearing, although there is still some compression experienced in steel ballsets. There will be slightly less in ceramic which may lend itself to being more prone to brinelling.
SKF is indeed a Swedish company but they do manufacture bearings all over the world. I have bought SKF bearings made in Japan, Italy, France.
I've replaced all my Campag Ultra-Torque orginal bearings with Boca Bearings' 'Yellow Seal' ceramics. Not so much for reduced friction but for far greater durability than afforded by the rubbish that come fitted. Have never seen the alleged pitting issues.
On what do you base your opinion? I have several bikes with Campagnolo Ultra-Torque and my experience is that after twenty or thirty thousand km of hard cycling tthere is no play, they turn smoothly as are just as light as when new.
Campagnolo original bearings not good? What have you been smoking?
we used to fit ceramic speed bearings to all the S-Works custom builds in our workshop because they came supplied with the frame
the poor BB design (OSBB) with the press fitted nylon cups which tended to 'walk' under load actually caused the bearings to fail long before any wattage savings could be experienced
we'd replace them with regular BB30 steel bearings and those nylon cups bonded with DP420 epoxy, and riders said there was no difference in feel at crank?
I know in mountain biking we would not use ceramic bearings due to constant impacts / shock loads, can't see it being any different in road cycling due to the state of our highways
Bought some Ceramic bearings for my BB from one of the main competitors to Ceramic Speed (similar major cost). Initially impressed with them dry (straight out the box in my hand) but by the time the supplied grease was added and the seals were on there was no difference to the old quality steel bearings.
I certainly didn't see my watt output increase. I have two bikes with identical drive trains apart from the ceramic bearings and swapping between them provides no measurable difference on an indoor trainer with power measurement (Kickr).
I would say save your money and spend some of it on quality steel bearings. Get ones with the highest ABEC rating - it doesn't guarantee the material quality only the roundness, but at least it is something
I think you're a bit confused there, because nothing you do to your bike whether aero, ceramic, carbon, lower rolling resistance tyres or any other upgrade, will affect your power. Your power meter or your trainer is only measuring the effort you yourself are making.
So if you're riding at say 200w, on your commuting bike with a rusty chain and half flat tyres and your powermeter says 200w, and you're doing say 15mph, and then you put your carbon super dooper wheels on that same bike with ceramic bearings and race tyres , and ride at the same 200w, you should notice some SPEED improvement at the same given power output -maybe you can see 20mph on your speedo. So the fast stuff you added allowed you to reduce drag/friction/weight or whatever and therefor go a bit quicker, it didn't affect in any way your fitness/strength/power output in watts
Thats the result of this formula : SPEED= sum(expensiveness+looks)x weight savings= total awesomeness.
It won't affect the power output by his legs, but it will affect the power reading given by a trainer as it measures power at the rear hub, therefore includes drivetrain losses.
Obviously, Vectors won't care what bearings you have in your BB.
It won't affect the power his legs put out, but it WILL affect the power at the rear hub, which is where the KICKR measures power.
If he was talking about power at the crank, you'd be right, but hes not.
Either way, its not exactly scentific as he is only measuring difference time to time. Obviously this doesnt factor other variables. To measure it, you'd need to measure the difference between the power at the crank and the power at the hub, to see what gains (if any) are made.
The problem is, the KICKR is only accurate to within +-5%. Ceraamic bearings dont give anywhere near that benifit, so you would need super accurate equipment to measure it. I suspect the benifit is so small, its not actaually measurable in a reporducable way. Thats probably why no one has published the results.
I think you're a bit confused there, because nothing you do to your bike whether aero, ceramic, carbon, lower rolling resistance tyres or any other upgrade, will affect your power. Your power meter or your trainer is only measuring the effort you yourself are making.
So if you're riding at say 200w, on your commuting bike with a rusty chain and half flat tyres and your powermeter says 200w, and you're doing say 15mph, and then you put your carbon super dooper wheels on that same bike with ceramic bearings and race tyres , and ride at the same 200w, you should notice some SPEED improvement at the same given power output -maybe you can see 20mph on your speedo. So the fast stuff you added allowed you to reduce drag/friction/weight or whatever and therefor go a bit quicker, it didn't affect in any way your fitness/strength/power output in watts
Thats the result of this formula : SPEED= sum(expensiveness+looks)x weight savings= total awesomeness.
Pages