Bicycle Mechanics - Replacing steel bearings with Ceramic - general question

Hi all,

I'm new here and not an experienced mechanic. I have 7 year old Trek 7300 adult bike that has seen some miles and I am beginning to replace some components.

In my reading (Sheldon Brown etc etc) I have come across the amazing factoid (?) that Ceramic bearings in the Bottom bracket can reduce total rolling friction from 4% to 0.5 percent. I have also read that (generally) loose balls are better than balls in a retainer, AND that light oil can (or should?) be used with ceramic Bearings. Finally I have also read that there are ceramic / ceramic races and balls but also there is a hybrid of ceramic balls with steel races.

OK, here's my beginner, probably too dumb to be worthy of a polite response, question -

What happens if somebody were to replace the steel bearings with high grade equivalent ceramic bearings?

I'm a tourist, the bike can use some further replacements, so I'm willing to take a chance, but then I don't want to do something just outright stupid....

What's the experience or best conjecture on doing such a thing? Certainly this has been tried? Perhaps some shops are doing it routinely already? Conversely if a mechanic were to do this would it be considered unorthodox or an outright mistake?

Also regarding the other bearings on a bike, would this also be true for hub bearings?

Regarding rolling friction I thought perhaps I could somehow measure before and after by doing something like fastening a known weight to the crank at 3:00, letting the crank "drop" then counting the number of turns the rear wheel makes maybe 5 times and averaging - before and after. If there is a significant change in friction this should give some measure of it?

Will appreciate any comments on this whole idea of substituting ceramic for steel into existing and lightly worn races.


Sincerely,

Michael

(My very first post ever here....)

In my reading (Sheldon Brown etc etc) I have come across the amazing factoid (?) that Ceramic bearings in the Bottom bracket can reduce total rolling friction from 4% to 0.5 percent.

Huh? that's contrary to what I've read. Any links? Is this the same Sheldon Brown that wrote this about ceramic bearings?:

Most bearing balls are super-hard steel. Lately, some suppliers have been promoting very expensive bearings that use ceramic balls.
These are a silly gimmick, aimed at gullible cyclists with too much money. They have no real benefit in practical use on bicycles.
"A fool and his money are soon parted."

What happens if somebody were to replace the steel bearings with high grade equivalent ceramic bearings?
First, welcome to the forum and, no, it's not a dumb question.

The brief answer to your question is that ceramic bearings are a tremendously expensive answer to a question nearly nobody asked. They do reduce friction in components with nearly no friction to begin with and do it at great cost. The Pros use them because their sponsors are paying for them but the real advantage even to them is almost zero.

As a tourist, keep your hubs and bottom bracket clean, well lubed and properly adjusted and use good quality steel bearings. You will never ever see any advantage to ceramic bearings over good quality steel ones but your wallet will thank you.

And, yes. loose balls instead of retainer held balls are slightly better since a few more balls will fit in the race and share the load over more bearings. Here, the cost penalty is about zero.

You have to decide whether or not ceramic bearings is worth the extra outlay. Ceramic bearings for other applications are often OEM packed with grease, so I don't know where that 'use light oil' comes from unless you are racing short distances or have a BB oil port and don't mind adding/checking oil levels frequently.

If the grease you are currently using is too tacky, then use another brand of grease or buy an NLGI 1 or 1.5 grease. This will make a bigger difference than ceramic bearings.

Waste of time even thinking about it.

If you wanted that miniscule advantage you'd be best to avoid the hybrid ceramic bearings which offer no advantage. With ceramic it's all or nothing. If you're not racing and your bike isn't worth at least $6k+, ceramic is a waste of money.

If you're not racing and your bike isn't worth at least $6k+, ceramic is a waste of money.
My take is that even if both of these things are true, you are wasting your money.

My take is that even if both of these things are true, you are wasting your money.

It's not.

The psychological advantage sometimes is enough to be useful, whether it makes you 0.00001% faster or 0.0000000000000001% faster. WR events have their own special preparation but they're guaranteed to be using ceramic.

Wow, thanks! 3 responses already. Amazing site/community here....

Here are a few further notes on where things came from:

1. The 4% steel versus as low as 0.5% for ceramic came from a Cycling news article Sept 14, 2006 - titled "WIll this new bottom Bracket make you faster?" H

2. No, Sheldon Brown certainly did not recommend and I certainly didn't mean to imply he did. The note about using loose balls came from him in a paragraph that pointed out that retainers are useful to manufacturers for assembly and to save cost be dropping a bearing (or two?).

3. There are various sources for individual ceramic bearings at prices that seem very close to steel to me - I might be missing something (like minimum order size/ shipping) but I'm thinking rather general anyway.


My impression of the overall situation is that between technology transfer, reduced sales overall, and new web based sources (bless them) that the price for (I believe first quality - but this is making some assumptions) Ceramic individual bearings have dropped significantly for the average Joe. I think I will do more research but a quick google of sources for loose ceramic bearings has given me a number of sources for a bag of 10 ceramic 0.250 Dia bearings for from 19 to 25USD.

Here I'll snip in their description:

Pack of 10 Loose Ceramic Balls 1/4 inch 6.35mm Si3N4, Grade-5 quality balls, this is a popular size that could be used in many Bicycle, or any application that uses this size 1/4 Inch. Balls are made of Ceramic Si3N4.


Item: Loose Balls
Size: 1/4 Inch = 6.35mm
Quality: Grade 5
Material: Ceramic
Quantity: 10 Loose Balls.

According to the Wheels Mnfg site :


Grade 5 Si3N4 bearings are twice as hard as steel bearings
70% stiffer, 75% smoother, 60% lighter


Does that change anything already said?

I'm a bit of a "armchair mechanic" and I enjoy keeping track of the shifting edge of these things.

So does the above price and information change things?

Also, I guess, I'd like to hear about "worse case" / possible problems.....

Thanks so much, I'm learning!

Michail

It doesn't.

You really aren't going to gain any benefit from ceramic for a performance hybrid. You are literally burning money. You'd be better off spending that extra money on high quality high TPI tyres (this WILL make you faster) inflated properly and/or a professional bike fit instead.

I can't stress how irrelevant bearing resistance is on a properly maintained high quality loose ball hubs (you'd be an idiot to put those on crappy hubs).

If you're tourist, or even a sport rider, your best bet is to get a sealed BB, be it shimano, phil, or one of the others. Then you get less friction than an open BB and low to no maintenance.

If you're tourist, or even a sport rider, your best bet is to get a sealed BB, be it shimano, phil, or one of the others. Then you get less friction than an open BB and low to no maintenance.

This is opposite of reality.

Loose ball bbs that came with last generations bikes have significantly less drag than modern shimano carts/phil carts with phil bearings. Even a $2 loose ball bb can be adjusted to spin more than a shimano cart. External bb systems are even worse.

The actual amount of drag from the BB is so miniscule (relative to total drag) that saving any of that drag is insignificant. So let's say a BB saps 4-seconds from you in a 100-mile ride. Reducing that loss to 0.5-seconds isn't really that huge of a savings. For example, it's not going to be the difference between 25th and 1st in a race.

. There are various sources for individual ceramic bearings at prices that seem very close to steel to me.....

Ceramic individual bearings have dropped significantly for the average Joe. I think I will do more research but a quick google of sources for loose ceramic bearings has given me a number of sources for a bag of 10 ceramic 0.250 Dia bearings for from 19 to 25USD.....
Does that change anything already said?

No, it doesn't change anything. A package of 10 1/4" ceramic balls for $20 is no where near close to good quality steel bearings. First, you will need two packages for a typical rear hub so the bearing balls alone will be $40 plus shipping.

For comparison, Grade 25 (very high quality) 1/4" chrome-steel bearing balls are less than $5 in packages of 100. Here is a source: http://www.loosescrews.com/index.cgi?d=single&c=Bearings&sc=Loose%20Balls&tc=Grade%2025&item_id=BB-CH5&id=964320116153

So the penalty is $35 for NO useful benefit.

operator's comment about the psychological benefit is correct and the comment about current record events always using ceramic bearings is also correct but someone else is paying for them, not the rider.

The actual amount of drag from the BB is so miniscule (relative to total drag) that saving any of that drag is insignificant. So let's say a BB saps 4-seconds from you in a 100-mile ride. Reducing that loss to 0.5-seconds isn't really that huge of a savings. For example, it's not going to be the difference between 25th and 1st in a race.

This is not correct. It's the amount of watts the BB takes. You can be certain this is significant with a well adjusted loose ball bb vs an external system. If you've seen both of these, campy record square taper vs, record "ultra torque" - you'll agree with me without argument.

This is not correct. It's the amount of watts the BB takes. You can be certain this is significant with a well adjusted loose ball bb vs an external system. If you've seen both of these, campy record square taper vs, record "ultra torque" - you'll agree with me without argument.

link/pic?

link/pic?

Only a video can show you this.

This is opposite of reality.

Loose ball bbs that came with last generations bikes have significantly less drag than modern shimano carts/phil carts with phil bearings. Even a $2 loose ball bb can be adjusted to spin more than a shimano cart. External bb systems are even worse.

Hey you know op, your right...in your world of perfectly square and aligned bottom bracket bracket faces and use of frictionless grease.

For the rest of us in reality however where we might not want to oil bathe our bottom brackets... your wrong.

1. The 4% steel versus as low as 0.5% for ceramic came from a Cycling news article Sept 14, 2006 - titled "WIll this new bottom Bracket make you faster?" H

You have to understand where this "knowledge" comes from. Journalists typically know next to nothing about what they're reporting on. They re-write what some else has told them. In 2006 ceramic bearings are just entering the market. The only source of data is the people selling the bearings. The "facts" of the article are probably a direct swipe from a press release. Swallow with several grams of salt.




Bicycling magazine, for instance, places vendor data ahead of common sense.

Bear in mind also, cheap ceramic bearings are often machined(is this even the word for ceramics) to lower tolerances than comparably priced steel bearings. You pay more for an inferior product.

This is not correct. It's the amount of watts the BB takes. You can be certain this is significant with a well adjusted loose ball bb vs an external system. If you've seen both of these, campy record square taper vs, record "ultra torque" - you'll agree with me without argument.Are you talking about the study Jobst Brandt did a couple of years ago comparing a Campy Record versus an FSA Megaexo?

I would also think that the rubber seals on the bearings would represent more drag than the difference between steel vs. ceramic bearings. Or even the difference in shear-friction between running grease versus oil.

I haven't seen anyone yet point out that even if you use bearings that are absolutely perfectly spherical, you are still spinning them in between cups and an axle that are unlikely to be ground and polished to anything close to a tolerance that could take advantage of such a bearing.

Which is why good ceramic bearings come in a cartridge that is designed around the bearing.

I personally think ceramic bearings make sense in derailleur pulleys-the fastest spinning bearings on a bike, then hubs. My DT 240s feel amazing with Enduro ceramic bearings in them, but as far as BBs go, they spin too slowly to make the kind of difference worth paying for....

My DT 240s feel amazing with Enduro ceramic bearings in them, but as far as BBs go, they spin too slowly to make the kind of difference worth paying for....
My Campy Chorus and Shimano Dura Ace hubs with the stock steel balls also feel "amazing". They will "pendulum" for a dozen cycles as the very slight wheel imbalance settles to the bottom after I hang the bikes on their storage racks. How much energy could these possibly absorb at riding speeds? If ceramic bearings cut the losses to zero, the difference on the road would be immeasurable.

I haven't seen anyone yet point out that even if you use bearings that are absolutely perfectly spherical, you are still spinning them in between cups and an axle that are unlikely to be ground and polished to anything close to a tolerance that could take advantage of such a bearing.

Off topic a little, but since the parts that ride on the bearings are not of high tolerance, would it matter if i installed Grade 25 or Grade 200 ball bearings?

Off topic a little, but since the parts that ride on the bearings are not of high tolerance, would it matter if i installed Grade 25 or Grade 200 ball bearings?
The races are really pretty good on most hubs and Grade 25 balls (the much higher grade ones) sell for so small a premium over Grade 200 that they are well worth using.

I personally think ceramic bearings make sense in derailleur pulleys-the fastest spinning bearings on a bike, then hubs. My DT 240s feel amazing with Enduro ceramic bearings in them, but as far as BBs go, they spin too slowly to make the kind of difference worth paying for....

The bolded statements makes very little sense. Bottom bracket bearings see the highest loads of any bearing on a bike whereas derailler pulley bearings see basically zero load (hence why most deraillers have simple bushings instead of bearings). Better bearings make a difference under heavy load. Almost any bearing can be adjusted to feel decent at zero load. I can agree with hubs possibly benefiting from ceramics but again only because they see such (relatively) high loads.

The bolded statements makes very little sense. Bottom bracket bearings see the highest loads of any bearing on a bike whereas derailler pulley bearings see basically zero load (hence why most deraillers have simple bushings instead of bearings). Better bearings make a difference under heavy load. Almost any bearing can be adjusted to feel decent at zero load. I can agree with hubs possibly benefiting from ceramics but again only because they see such (relatively) high loads.

Nice derailleurs have bearings. My 7900 rear derailleur came with ceramic bearings. And my old Ultegra had ceramic bushings. Nearly every bike I work on has a lot of drivetrain drag because no one ever lubes the rear derailleur pulleys. Far worse than hubs or BBs that actually have grease in them.

Say what you will, but most humans don't create "heavy loads" on BB bearings. The state of lubrication and, on cup and axle BBs, adjustment, has more to do with how much drag is created than the tolerance of whatever bearing happens to be in the races. The old-school BB on my SunTour Superbe Pro equipped Fuji (ca. 1986) spins with less drag than the ceramic bearing'd BB on my Madone. Can I measure it or even feel it when riding? No.

BTW, Grade 25 bearings were spec'd only for ground and polished races. The cheaper cast metal races can't make use of a better bearing because there's too much runout.

Thanks much for all the input!

As I try to piece together the bigger picture I have to think that with such a thing as this - with a near total lack of unbiased practical lab results - that this forum can turn in to a "Greek Philosopher" debate club. And indeed there are many factors involved with rolling friction. I looked to see what is available for books on Bearings and saw only a few - and they started around $200.

I rather think that lower costs and more competitive distribution has sneaked up on some of us and the additional cost of ceramic vs steel is now so low that this is ripe for some rethinking / practical field testing. Also I don't think anybody mentioned that ceramic bearings can/should last longer by several times, and will not corrode or "weld" over time like steel to steel can do.

If I were a mechanic working on other people's bikes I WOULD NOT BE putting ceramic bearings in on the bet that they might be better - I think that the risk is to great on the negative side - but for my purposes of testing, and perhaps just having something to test/think/talk about plus or negative - I think I shall still do it when the chance comes up.

There was one particular comment that rather bothers me or rather threatens to become an interesting "gedanken gang" or thought experiment for me for awhile. This comment was something like "if you don't have a $6,000 bicycle it won't matter". Hmmmm.

Might be off-topic but as I thought about that I came up with the counter thought that:

for about $500 a machine could be selected and put together by a good mechanic that would be very near mechanical efficiency wise (rolling resistance) to the best machines you that can be purchased anywhere for any amount of money - measured from the pedals to the rear hub independent of spokes/rim/tires.

Just my humble non-mechanic guess on things based on how the world seems to work.


Thanks!

Michael

Off topic a little, but since the parts that ride on the bearings are not of high tolerance, would it matter if i installed Grade 25 or Grade 200 ball bearings?

Yeah, it does matter. Given suitable materials, the biggest difference is how well matched the bearings are in size. Bearing sets with a wide range of sizes concentrate the load on single balls, leading (relatively) rapidly to failure. Bearing balls are graded primarily by sphericity (e.g. 25 μ vs. 200 μ), but the lot size tolerance is as great between grades. Grade 200 won't last as long, but not due primarily to the greater sphericity tolerance.

... with a near total lack of unbiased practical lab results - that this forum can turn in to a "Greek Philosopher" debate club.
Know any Greek philosophers that hold Engineering degrees?



I rather think that lower costs and more competitive distribution has sneaked up on some of us and the additional cost of ceramic vs steel is now so low...
Don't be tempted into thinking that economy of scale applies here. These bearings have been used for years in much larger applications. A "novelty fee" should still apply. Any significantly lower after-market prices are the result of lower-cost materials - not just in the bearings, but in other components of the assembly. In other words, the numbers that have been bandied about for "ceramic bearings" probably don't apply to these ceramic bearings.


Also I don't think anybody mentioned that ceramic bearings can/should last longer by several times, and will not corrode or "weld" over time like steel to steel can do.
Ceramic bearings do have different failure modes. That doesn't imply that practical failure rates are different one way or the other. (Unless you leave your BB soaking in salt water for a few months...) For that you need .. um .. practical lab results.


but for my purposes of testing, and perhaps just having something to test/think/talk about ... I think I shall still do it when the chance comes up.
Now this is an issue that a scientist would pick up on much more quickly than a Greek philosopher. Your "test" has absolutely no significance to the issue. Nor does any other uncontrolled test with a sample size of one. But if you want to do it for fun, by all means...


for about $500 a machine could be selected and put together by a good mechanic that would be very near mechanical efficiency wise (rolling resistance) to the best machines...
True, because he'd be using steel bearings. ;)

I'm not going to stop and write a lengthy explanation because 1) I've done it before in this forum (search!) and 2) I don't really have the time. That said, it seems several people here are mixing up seal drag (what you feel spinning a bearing with nearly no load) and bearing drag due to the load (what actually causes measurable losses), which you cannot easily determine just by spinning a crank and seeing how long it goes for. Those unclear on this should look into it more before returning to the "Greek Philosopher" debate club.

As I try to piece together the bigger picture I have to think that with such a thing as this - with a near total lack of unbiased practical lab results
Not published here, but with some digging you may find something (try the local library, journal databases, etc).


- that this forum can turn in to a "Greek Philosopher" debate club. And indeed there are many factors involved with rolling friction.
Yep and yep.

I looked to see what is available for books on Bearings and saw only a few - and they started around $200.
Local library


I rather think that lower costs and more competitive distribution has sneaked up on some of us and the additional cost of ceramic vs steel is now so low that this is ripe for some rethinking / practical field testing.
I disagree, but go for it if you want to.

Also I don't think anybody mentioned that ceramic bearings can/should last longer by several times, and will not corrode or "weld" over time like steel to steel can do.
Weld? Really? I'm not sure what you wanted to say there, but it wasn't weld. Also, even the most minimal preventative maintenance will prevent hubs and BB shells from corroding internally.


If I were a mechanic working on other people's bikes I WOULD NOT BE putting ceramic bearings in on the bet that they might be better - I think that the risk is to great on the negative side - but for my purposes of testing, and perhaps just having something to test/think/talk about plus or negative - I think I shall still do it when the chance comes up.


There was one particular comment that rather bothers me or rather threatens to become an interesting "gedanken gang" or thought experiment for me for awhile. This comment was something like "if you don't have a $6,000 bicycle it won't matter". Hmmmm.
That one is easy. Unless you already have a $6000 bike, there are surely more cost effective ways to go faster.

Might be off-topic but as I thought about that I came up with the counter thought that:
for about $500 a machine could be selected and put together by a good mechanic that would be very near mechanical efficiency wise (rolling resistance) to the best machines you that can be purchased anywhere for any amount of money - measured from the pedals to the rear hub independent of spokes/rim/tires.
Sort of like the 80/20 rule to an extreme, although you are correct. The $$ in cycling comes from ease of shifting and durability. A good example is chains. Most bike chains are all pretty close right out of the box as far as losses go. However, add some road grit, or take shifting quality and drive train noise into account, then you get some differences.

will not ... "weld" .. like steel to steel can do.Weld? Really? I'm not sure what you wanted to say there, but it wasn't weld.
He's (incorrectly) implying that welding (or more properly micro-welding) is similar to corrosion. It isn't. It's a failure mode whereby the load causes the ball and race metals to stick (weld) together. As the ball moves on it literally rips (tiny) chunks from one or the other. The resulting chunks and holes subsequently become problems in their own right...

If some modification saved 4 seconds per 100 miles it would be a pretty good mod, since 4 seconds would be 120 feet on a level highway.
Charlie
PS I have no idea if ceramic bearing can save 120 feet per 100 miles- that seems too good to be true, but...

Timeout, please?!

I did put "weld" in quotes - which to me means "lighten up on pickymeter stuff" - and SOMEWHERE (and I COULD find it) some "authority" referred to what happens with to steel to steel contact over longer periods of time as a "weld" - the eternity approaching us here, known as a Wisconsin Winter, would be one example :).

I'd better "confess" that I'm actually a software guy writing a application for the bicycle domain. Its rather a "builder" site like this one for Assault Rifles: http://www.ar15builder.com/ this site I'm working on should be 100X more powerful but also more peaceful :)

Recently I have put together some rather basic "script files" and was thinking to move along to something meatier than the clearance / interference of seatpost od in a frame id (my very first script file for the bicycle domain).

Several weeks ago I tried to contact John Barnett as I was impressed with his manuals and more or less got "shot down" before I could talk with him. UBI seems like they could be interested but they are 2300 long miles away from my home in NE Wisconsin.

"Script files", as I am using the term are focused little pseudo programs that non-programmers can write. That is a major objective of this project I've been working on for several years. I think of them as a "swarm" - and they can actually be as simple as a "review" or as complicated as necessary to build a table of cadence versus speed (as I just finished). These scripts can allow opinions to be just that - which could be of value with proper expertise backing it up.

...so when some kid is dragging and dropping a fork onto his new dream mountain bike frame "swarms" of 1,00s or even 1,000s of script files could watch over his every "decision". (It could also be automatically delmiting his choices but that wouldn't be as much fun.)

Hmmmm.

I haven't seen anyone yet point out that even if you use bearings that are absolutely perfectly spherical, you are still spinning them in between cups and an axle that are unlikely to be ground and polished to anything close to a tolerance that could take advantage of such a bearing.

Which is why good ceramic bearings come in a cartridge that is designed around the bearing.

I personally think ceramic bearings make sense in derailleur pulleys-the fastest spinning bearings on a bike, then hubs. My DT 240s feel amazing with Enduro ceramic bearings in them, but as far as BBs go, they spin too slowly to make the kind of difference worth paying for....
There is no part on a bike that can make use of the quality of ceramic bearings. They are designed for invironments where they are in high heat and very high speeds.
For our use a very good quality bronze bushing would be more than adequate. 700 series wheels at 60 mph are turning at 670 rpm, not exactly blazing fast.

I did put "weld" in quotes - which to me means "lighten up on pickymeter stuff" - and SOMEWHERE (and I COULD find it) some "authority" referred to what happens with to steel to steel contact over longer periods of time as a "weld" - the eternity approaching us here, known as a Wisconsin Winter, would be one example :).

I'd better "confess" that I'm actually a software guy writing a application for the bicycle domain...

In other words, you don't really know what you're talking about. You're just extrapolating from what you've read.

Well, there's place a for that. We all do it insofar as we can. But you've come among experts - people that really do know what they're talking about. A little deference is in order, IMNSHO.

As for your app, if you weight opinions equally then you're compiling a grand mess. There *are* correct answers to technical questions. (Yes, I know Google is supposedly "the death of the scientific method", but that's a pile of crap.)

He's (incorrectly) implying that welding (or more properly micro-welding) is similar to corrosion. It isn't. It's a failure mode whereby the load causes the ball and race metals to stick (weld) together. As the ball moves on it literally rips (tiny) chunks from one or the other. The resulting chunks and holes subsequently become problems in their own right...

Also known as "galling."

Nice derailleurs have bearings. My 7900 rear derailleur came with ceramic bearings. And my old Ultegra had ceramic bushings. Nearly every bike I work on has a lot of drivetrain drag because no one ever lubes the rear derailleur pulleys. Far worse than hubs or BBs that actually have grease in them.

I've work on a decent number of bikes and have yet to come across a derailler pulley that offered significant drag regardless of how mucked up it was. There just isn't enough load put on those pulleys for it to matter. However, I've felt the extra friction created by a bottom bracket on it's way out while pedaling hard even though spinning it by hand felt just fine.


Say what you will, but most humans don't create "heavy loads" on BB bearings. The state of lubrication and, on cup and axle BBs, adjustment, has more to do with how much drag is created than the tolerance of whatever bearing happens to be in the races. The old-school BB on my SunTour Superbe Pro equipped Fuji (ca. 1986) spins with less drag than the ceramic bearing'd BB on my Madone. Can I measure it or even feel it when riding? No.

As has been pointed out, you are feeling the seal drag on the newer bottom brackets, something that was not present on the cup and cone style bearings. You mention how adjustment can affect the drag on a bearing. Well, if you have crap parts you can't get them adjusted very well. Uneven races and/or balls will make a perfect adjustment impossible. As for heavy loads, if bottom brackets are so overbuilt, why aren't there manufacturers out there touting lighter duty bottom brackets for weight savings? Smaller balls meaning less weight, right? Oh right, they tried that with ISIS and it failed (more often than other systems).


BTW, Grade 25 bearings were spec'd only for ground and polished races. The cheaper cast metal races can't make use of a better bearing because there's too much runout.

All bearing races are ground and polished to some extent. If you wanted to feel some friction, try slapping some bearings into an as cast race. It wouldn't be pretty. Some races obviously are ground and polished to tighter tolerances than others, in addition to being made of higher quality material/better heat treatment.

^^This guy seems to know what he's talking about. Pay attention.

In other words, you don't really know what you're talking about. You're just extrapolating from what you've read.


Yes, my experience on many of these things will be thinner than the chrome plating on a Flying Pigeon spoke. I am a life long (61yrs) cyclist and crossed the alps among other stupid bike tricks when I was much younger.

And YES, the expertise here is astounding - and I have been in (almost) 100% in deference mode. I just don't do 100% deference, I think I could shoot a some holes into some of the logic given (such as "ceramic bearings were designed for....yadda yadda... therefore.. yadda not useful yadda for low rpm yadda). Am I missing something? That ability to handle higher RPM can be converted to reliability / or "traded" for life by running almost or perhaps bare dry - that's just good armchair logic. Sloshing those bearing through miles of grease has to cost a few ergs of energy, no way that it doesn't.

Some of the replies have hit me square between the eyes - the one about the difference between "low level" params such as minimum friction and long term reliability. The mentioning of failure mode. "Weld" versus Galling - Great stuff! I can makes notes on that kind of thing.

We worked with some open source guys that have authored an astounding web framework (imho) and to work with them we agreed to make our front-end open source. That's when I had the idea of making the new demo sites / testing domain bicycles because I have loved the darned things. That was 8 months ago.

I have to find a good source of expertise, this is not going to work here unless they give me 5 yrs to get the first version up. This is a herd of cats kind of situation. Its here but not in this format exactly. The initial software concerns are mostly toward compatibility issues between components for purchasing / building or repairing a bicycle. There is complete project persistence between sessions. I really think it is going to be very useful but the real value may be awhile off yet until we can find the right folks to work with. Spending all day reading in repair/maintenance books and trying to pick meaningful examples. That's how this ceramic issue came up just as my bike is needing some repairs from a broken hanger.




Michael

Wow, what a pissing contest!



You know it is like this. No one really needs more than a Honda Accord but yet we don't all drive them. Most Hummer owners go grocery shopping in them amd drive on pavement all the time. The beauty of it all is that we can choose what makes us feel good. That being said, if you want ceramic bearings and can aford it than go for it. If it makes you enjoy the bike more you will ride the bike more and that is what it is all about. There is always a conflicting opinion with evedence to back it up so listen to yourself and not what everyone else thinks.



Enjoy the ride. It is all you get.



Greg

^^This guy seems to know what he's talking about. Pay attention.

No. But he sounds like it. And on the interwebs, it's often enough.

I work on bikes 10 hours a day, 5-6 days a week and have for nearly a quarter century. I say what I say based on actual repeated observations. When I have a bike in the stand with lots of drag in the drivetrain, and squirting PG2000 in the derailleur pulleys frees things up considerably, I think it's safe to say the pulleys were the issue. This tells us we might need to take them apart and check their condition-not ignore them! Might the BB need attention as well? Anyone can feel drag in a BB on its way out! We're talking about friction in properly adjusted new or slightly used parts.

No. But he sounds like it. And on the interwebs, it's often enough.

Yes, definitely trust the guy pushing ceramic bearing derailler pulleys :rolleyes:

No. But he sounds like it. And on the interwebs, it's often enough.

I work on bikes 10 hours a day, 5-6 days a week and have for nearly a quarter century. I say what I say based on actual repeated observations. When I have a bike in the stand with lots of drag in the drivetrain, and squirting PG2000 in the derailleur pulleys frees things up considerably, I think it's safe to say the pulleys were the issue. This tells us we might need to take them apart and check their condition-not ignore them! Might the BB need attention as well? Anyone can feel drag in a BB on its way out! We're talking about friction in properly adjusted new or slightly used parts.

If you think you can directly correlate how a bearing feels when spinning on a stand to how it will perform on the road then you need to revisit my comment above about seal drag vs load. Yes, a bearing which feels like sandpaper on the stand will be bad on the road, but that's about as much as you can surmise. I'm sure you're a very good mechanic, especially after 25 years at it, but it doesn't mean you know any of the science behind it. Along with that, it makes you much more likely to believe the myths that have been stuck in the cycling industry for forever.

As far as I can tell, everything joejack said was correct, or at least much closer to correct than 90% of the other stuff in this forum.

Contrary to popular expectation, the wheel people I know have found their ceramic bearings (including the ones with ceramic races) to fail about as fast under poor weather commuting use as their steel bearings, but the steel ones are lots cheaper.

Wow, what a pissing contest!



You know it is like this. No one really needs more than a Honda Accord but yet we don't all drive them. Most Hummer owners go grocery shopping in them amd drive on pavement all the time. The beauty of it all is that we can choose what makes us feel good. That being said, if you want ceramic bearings and can aford it than go for it. If it makes you enjoy the bike more you will ride the bike more and that is what it is all about. There is always a conflicting opinion with evedence to back it up so listen to yourself and not what everyone else thinks.



Enjoy the ride. It is all you get.



Greg

I repeat myself when under stress, I repeat myself when under stress, I repeat myself when under stress, I repeat myself when under stress, I repeat myself when under stress,

Hey! Anyone remember King Crimson?

Greg

and I have been in (almost) 100% in deference mode. I just don't do 100% deference,
As a registered Elitist, even the "almost" gains you some respect from me. :p


I think I could shoot a some holes into some of the logic given (such as "ceramic bearings were designed for....yadda yadda... therefore.. yadda not useful yadda for low rpm yadda). Am I missing something? That ability to handle higher RPM can be converted to reliability / or "traded" for life by running almost or perhaps bare dry - that's just good armchair logic.
Not good logic. I know the fallacy has a name; I can't think of it at the moment. Call it an "unwarranted conclusion". Since it hasn't been stated why ceramics are better in yada yada applications, no conclusion about performance in a different application can be made.

What ceramic gains you in those high-stress applications is primarily resistance to heat. Even with hardening or carburizing, steel - especially when heat cycled - will eventually degrade. Ceramic is intrinsically hard and won't. (There's a flip side to that, too: ceramic is more susceptible to shock loads.) In low rpm - i.e. cool - applications, ceramic's primary advantage is irrelevant. There will be a resistance to corrosion, though with metal races that won't matter much. And while there may be an advantage in rolling resistance it is small - likely too small to justify the cost.

Now I'm no lubrication engineer, but it seems to me that a bearing's own worst enemy is itself - the little bits the break off it are hard enough to do a lot more damage should they get in the way. Lube - especially grease - suspends them and keeps most out of the way. Thus ceramic won't offer any advantage when run dry. But I don't know this for sure.

But to be fair, cost seems to me the only downside. So if that's not an issue (and all else, like Grade, being equal), go for it.


OT:

We worked with some open source guys that have authored an astounding web framework (imho) and to work with them we agreed to make our front-end open source. That's when I had the idea of making the new demo sites / testing domain bicycles because I have loved the darned things. That was 8 months ago.

I have to find a good source of expertise, this is not going to work here unless they give me 5 yrs to get the first version up. This is a herd of cats kind of situation. Its here but not in this format exactly. The initial software concerns are mostly toward compatibility issues between components for purchasing / building or repairing a bicycle. There is complete project persistence between sessions. I really think it is going to be very useful but the real value may be awhile off yet until we can find the right folks to work with. Spending all day reading in repair/maintenance books and trying to pick meaningful examples. That's how this ceramic issue came up just as my bike is needing some repairs from a broken hanger.
Open source! :thumb: Can you name the framework? Is it Spring? (IMO they have a superb web tier design. I just wisht I knew it better.) How are you doing persistence? Are you planning on much scalability?

I suppose we should take this to PM, but I bet you'd get some good feedback from a preview link posted here.




Wow, what a pissing contest!
Bull cookies! This is one of the better discussions I've participated in here. It's great to deal with someone that really wants to know why!*

Better that than someone for whom Truth is subjective. [hint hint]



* Or in this case why not...

Hey! Anyone remember King Crimson?

OH yeah. One of my favorite groups, though the one song most people know - 21st Century Schizoid Man - is my least favorite.

But to be fair, cost seems to me the only downside. So if that's not an issue (and all else, like Grade, being equal), go for it.
That's probably the most succinct summary statement in this whole discussion. :thumb:

I agree there is really no functional downside to ceramic bearings but the functional upside is insignificant. The major objections are both cost and the absurd claims some marketers have made for them. If you have the money to spend, go for it but don't expect miracles.

Open source! :thumb: Can you name the framework? Is it Spring? (IMO they have a superb web tier design. I just wisht I knew it better.) How are you doing persistence? Are you planning on much scalability?

The language we are using is Ruby (open source and from a Japanese author), the framework I'm in love with is "Ramaze", and the author is one amazing Ramazian that uses the nickname Manveru is an Austrian in Tokyo....

What I would say about Ramaze (and Rack on which it is based) is that it has allowed me to use the framework "as an extension to Ruby" without all the piles of "magic incantations" needed to work with a package such as Rails - this is the key from my perspective as a looong time app guy - with little to no personal interest in web "design" itself (as a programmer).

Here's a link to their site "learn" page: http://ramaze.net/learn

There's a IRQ link from this site. I monitor this IRQ generally 24-7 to pick-up tidbits of info.

=================

Persistence has come largely from totally isolating the "model" into an "Engine" and this engine communicates by way of six "Channels" - the channels are flash memory on my system and will be even faster on the eventual server. This rather automatically gives us persistence as the channels are keyed by the unique userID - so if a user thread just "leaves" / goes down or whatever should happen - when a user should come back the channels will be there and the front-end will respond to him one week or a month later like he never left.

BUT that is for current / open project - a big part of the concept for us is that the Project is given a name and a bike type is selected - and when the project is "saved" it will be there by name for a given client. We build a complete "virtual bike" at the moment this basic info is known - right now 164 departs of data (some logical rather than actual). The whole things is driven from a single "indented" or "outline" type of file (YAML for Ruby). However the departments are made "smart" by class extensions. The extensions include all sorts of information but at the bottom of it all they have a list of "picklists" for each department. And - last but not least - the "picklists" are actually, in some cases matrices. And I am at the stage of trying to build sets of these things - these picklists but also testing the whole thing downstream. I have a half dozen scripts now firing and it takes only 1 sec for over 1000 to fire... about scripts below - this is sort of a word we've taken on to describe the simple source files that the experts will generate - but not directly. It is more like the "assembly code" of our system that gets compiled from a even friendlier front-end to allow the experts not to get their hands dirty in programmer grease :) hehehe

What then "glues" it all together - and I personally am very pleased with how this has gone - is a "DSL" / D)omain S)pecific L)anguage which allows the Bicycle expert to write "expert scripts" ONE TIME forever - think about this - and the script then gets "fired" against those departments/parts that it involves >>automatically whenever<< that combination should come up. Writing "expert scripts" is like dropping a live bullet into the magazine of a virtual machine gun that fires away at Compatibility question automatically whenever those particular components are matched.

We did a drag & Drop front end that I am just now thinking can be paired with the Park Tool like "bicycle map" - the D&D generates a signal - and that is all we need to make a part of a diagram visible / or invisible removed =. You can see that front-end without any real "brains" here: http://buildmyworld.info/build/ (up and running the last 8 months while I have been doing my best to put the "brains" together in a utility version which should be going up next week or soon - it has a "stuck in the 90's ugly HTML" front-end but Web Designer people can
turn that into something nice - they don't need me - I'm alergic to CSS and that stuff (too much real app stuff to do here).

Ruby is just super duper for writing DSLs - on this I could go on for days - the "method missing" handling is the door to that kingdom. My DSLs break down into 4 or so parts and each one is really easy to view - if you are interested I have 6 examples or so - one builds a cadence table, one checks for seattube ID versus seatpost OD - dumb stuff maybe but think of a "swarm" of 100's (actually eventually 1000's) of these?! each leaving just a bit of determined information for successive "waves" to use. It is very powerful stuff imho.

Its been a couple of years of long hours of work and I'm anxious to now start showing the concept off to people that can see where this can go.... it is no substitute for "real expertise" but it will can eliminate the repeated and repeated stuff and it will be world-wide and I'm hoping it can take on a sort of "wikipedia" kind of mechanism for Bicycle Experts - like yuse guys here.

UBI has shown "some interest" but they just opened a new campus. BBI sensed that this thing could be competition - and I guess that is true. Park Tool is rather in my neighborhood and they are international, so I have ordered their BB2 manual to study.

Oh shoot, I was trying to clean up my post and ended up submitting it before I was done editing it somehow. My wife was calling for me to make my token contribution to dinner... and I knew I was pushing my "Just let me finish this" time limit. I'll be glad to answer any interest in this more properly but will drop for now. Within a week the utility version (ugly but smart) should be up and that will demonstrate Navigation and Selection. Its not an entirely trivial project....