Carbon steel or chromium steel? After what grade do marginal returns begin to drop off?

There is (or used to be, anyway) a rule of thumb about high-end bikes and components that you get 95% of the performance for 50% of the cost. Grant Petersen at Rivendell has applied it to his ready-made bikes vs. his customs. I've found it to be generally true.

You talking about buying loose balls ?
the cost of the higher [grade 25] steel loose balls is not significant.
bag of a thousand is pretty cheap.
industrial bearing supply less than thru Bike shops.

and the races last longer if you repeatedly maintain them, and toss used balls.

Yes. I already have been through my work. Could you qualify your second statement fietsbob?

I've heard and read reports of those who ditch the stock headset bearing retainers and bearings in favor of using higher-grade bearings á la Grant Petersen. If I plan to use loose ball bearing hub and headset systems I'm curious to know if there is a noticeable increase in performance and longevity using higher-grade bearings instead of the stock balls.

I guess it all depends on your definition of 'Noticeable'..
When I looked at the film "Stars and Watercarriers", & before the Time Trial days,
no wing shaped Time trial-special bikes then,
the team mechanics cleaned and re did the hubs and BB bearings,
on their racers bikes, to make sure there was best possible performance.
that day , from their team's men , so it must have mattered to them, at that time.

The balls only need to be replaced when the finish has dulled.

No. The performance will not be improved. Longevity is a matter of proper maintaince.

WoS- Your question is somewhat confusing. If you are talking about just the "balls" alone, they are generally so cheap, it's probably best to replace with the best grade that is readily available. Don't expect a real performance improvement however. Most of it comes from maintenance as mentioned.

The replacement bearings in better grade hubs and loose ball headsets are stainless, so you don't have to choose between carbon or chromium.

As far as the other components are concerned you can easily get equal performance for a lot less money. But there's always a market for the high end stuff, and I'm glad there is- it enables the manufacturers to provide a better mid-range part for the likes of you and me.

The "standard" for bicycle loose bearing balls is chrome steel Grade 25. Grade 200 (the lower the nuumber, the better the roundness and finish) is adequate but Grade 25 costs very little more and are enough better to justify the small price increment. Beyond Grade 25 (Grade 10 for example) there is no benefit.

I use grade 25 bb's because they are only pennies more than lower spec bb's. Better performance?...probably not noticeable. I'm more concerned about preventing damage to the more expensive races in the hubs, BB's, and headsets. So when I routinely disassemble, clean, and grease and reassemble these parts I just throw out the old bb's and replace them with new ones. Many would view that as unecessary and wasteful, but that's what I do.

Where did you get that idea?

There are indeed stainless steel bearing balls available (usually 440C) but they are a specialty item used only where their corrosion resistance is of value. I don't know of any bicycle component that uses them either OEM or as replacements.

Campy claims their bearing sets are individually matched and therefore charges much more for their name on them. But they aren't stainless steel and, if there is any performance benefit to the extra "matching", no one has really demonstrated it.

Yeah sorry- it may be naive of me to believe manufacturers advertising. When Shimano says their hubs use stainless balls, should I not believe them? No real way of knowing I guess since they are all round and shiny.

Can you reference any literature or ads where Shimano claims their hub or bottom bracket bearing balls are stainless steel? They do list the ball diameter and number for their hubs but i've never seen a reference to their material of construction. The only claim i've seen for stainless steel is the spokes on their prebuilt wheels.

Well, now that you mention it... neither did I. Looks like another "moment" on my part. My assumption came from here-https://www.universalcycles.com/shopp...4&category=745 I had no reason to not believe their claim. I know- it's advertising, and the internet, but would like to believe they were truthful. I've not seen any deceptive practices from these guys before. So, they may be lying or got this info from somewhere- like the Shimano rep? Not that all reps know what they're talking about. I found their claim believable because at least they didn't list the Tiagra hub as having SS balls.

So this discussion caused me to go to the Shimano site also, and I didn't find any mention of SS balls in their hubs either. The moral of this story- don't believe everything you read, and please disregard any thoughts I had on this subject.

Bike Tools etc. carries a (very) small selection of loose balls made from 440C stainless steel.

At least on one specification sheet, 440C ball bearings have a lower value on the Rockwell scale than normal chromium steel ball bearings, meaning the stainless steel ball bearings are softer. Whether that difference is significant for bicycle use, I don't know.


I see/hear this alleged Campagnolo claim many times, but I've never seen a Campagnolo rep or ad material state that. Nevertheless, the numeric grade given to ball bearing grade specification refers only to the sphericity and lot diameter variation; the overall ball diameter tolerance is also specified but is a much larger number.

For example, Grade 25 ball bearings are constraint to within 0.000025" sphericity and lot diameter variation. However, the overall ball diameter tolerance is 0.0001", and this is 2 orders of magnitude larger. So, at least on paper, it would seem prudent to "match" loose balls to the same lot to gain 2 orders of magnitude of uniformity. However, I don't know whether that has any meaningful impact to smoothness or durability. Sounds good though!

Buy AISI 52100 grade 25 balls.
Don't remove the ball cages in hubs, the cages separate the balls, preventing sliding contact between them as the hub turns.
In a headset, it's ok to lose the cages and add more balls since the balls are not rolling to any great degree and the additional balls reduce the unit load on any single contact.

Interestingly I could find no mention of SS balls in the 6700 freehub on the NorthAmerican site but could on the Australian site.
https://www.shimano.com/publish/conte...0.-type-..html

I suspect the actual wheel hub bearings are chrome steel, but the actual freehub bearings are SS.

Well, Shimano uses loose ball in almost all of their hubs so they don't seem worried about sliding contact between adjacent balls. Also, even if the cages keep the balls from rubbing on each other, they will rub on the cage itself so that seems like a wash to me. I've always reguarded cages (or retainers if you prefer) as there to speed and ease assembly, not as a performance improver.

As part of their BS shimano put SS bearings in their DA hubs. I have a set in a jar to compare the bearings I remove for my hubs for cleaning. If they look like the SS I discard them.

The only reason for cages in cup and cone hubs is to make it easier to assemble the hubs.

Cranks spin at 60 to 120 rpm. Wheels at 30mph are spinning at 335 rpm. For our purpose we would be fine with high quality bronze bushings.

I suppose so, except no one uses bronze bushings for these applications, probably for several reasons: 1) adjustability...race and cone bb's allow you to take up bearing slack. You could do this with adjustable conical bronze bearings and seats. That's how they do it with things like lathe headstocks, including many high precision jeweler's lathes. 2) wear. Bronze bushings are softer than steel bb's and race/cones. Bike wheels and cranks may spin at relatively low rpm's, but they do so under large loads and they are exposed to a lot of weather and road grit, which would screw up bronze bushings.

To the OP...I don't know if it's actually been stated here and I missed it
or just has not yet come up, but one reason for losing the bearing cages
is that you can usually fit in one or two more bearings in your assembly.

This makes the support points of the bearings and races both a little more
numerous and a little closer together, which, if you think about it, probably
improves their load performance marginally. Anyway, that's why I usually do
it, particularly in headsets.

+1 on the Grade #25 Chrome Steel balls being all anyone will ever need on
a bike, performance wise, and the advisability of replacing them if there
is any question with regard to their status. Most bearing assemblies
are consciously designed with the idea that the bearings will be replaced
a few times during service life. The races in the good ones are hardened
and will go for a long time if serviced and lubricated.

XXXMart bikes are and exception to this rule, and would probably benefit
from the aforementioned bronze bushings.

Not true.
Cages maintain consistent spacing between the balls preventing contact or skidding between adjacent balls. Keep in mind the contacting surfaces are sliding in opposition to each other with a combined surface speed of twice the ball rotation speed.
The entire point of a ball bearing system is to eliminate sliding friction. When the balls are forced together (as they are in a radially loaded full complement bearing) they are in sliding contact.
A 52100 ball is Rc60 minimum at the surface. Cages are made of much softer materials, most with self lubricating properties, plastics, bronze or at worst, in bearings with limited max speeds, a soft annealed steel. Contact of the balls to the cages results in cage wear only.

Full complement, cage less, bearings are only used in predominantly axially loaded, low speed applications.