headset brinelling / fewer ball bearings



jasong

New Member
Nov 24, 2003
268
0
0
I'm curious why manufacturers put in a retainer in headsets that have ball bearings, if guides like Barnett's suggest that they should immediately be removed on new bikes. Is it just to speed up initial assembly? After servicing an older bike, and trying to get the retainer back in, it just wasn't going to go and not create a lot of friction.. So, I looked into the guide for their thoughts and was happy to see their recommendation of getting rid of it and filling the cup with more balls.

Also, most people, if removing (or if it didn't exist) the retainer, leave out 2-3 balls in the headset, to avoid brinelling?

Which made me wonder about a few wheels I've received through mail order. The front wheels have, at least twice, missed a ball on both sides (there was ample space to insert an extra and not change spacing). Was this done purposely for reasons similar to those for avoiding brinelling in the headset?

Headset brinelling seems much more likely, because there is much less rotation, and the balls, if given minimal space between each other, could be seeing the same position in the race for a long period, and with sustaining substantial pressure in this fixed position, and would not have an obvious place to move to if sitting next to another ball. However, wheels by nature are rotating, even through impact, making that much less likely.

Thoughts? Others leave a ball out on the front hubs? Barnett's guide said nothing of the sort in their section on hubs. What could be negative effects of missing these balls?

Would also like to add that I pulled the seal on a Cane Creek cartridge bearing (headset) and it's completely full. That's not an apples-apples comparison of course since those can be tossed and the balls aren't touching anything more permanently installed.

Thanks,
 
I think manufacturers started using the retainers just to make servicing and assembling headsets easier. That way when one pulls the fork one doesn't have to try to account for all the little bearings bouncing around on the floor etc... Yes, technically the more bearing surface the better, which would suggest that pitching the retainers and putting loose bearings in the headset would be the way to go. But the bottom line is that if it's properly maintained and adjusted, a headset with retainers can last a really, really long time!

Really there are so many good cartridge bearing headsets becoming available now that eventually it'll be a mute point.
 
In most situations the balls would be best without the retainers. The biggest example is bottom brackets. When they were open type(thank god they are not common now) They had to be rebuilt every couple hundred miles or else the bearing retainers would toast the assembly. Thats why some went to plastic retainers. There is a debate in the theoretical category that having a retainer contact a rotating ball is less friction than the balls contacting eachother. I don't know how much that works out in practice, but I am more concerned with the load bearing ability of the higher number of balls.
 
:rolleyes:
Conniebiker said:
The biggest example is bottom brackets. When they were open type(thank god they are not common now) They had to be rebuilt every couple hundred miles or else the bearing retainers would toast the assembly. Thats why some went to plastic retainers.
Oh, BS :rolleyes:
 
Conniebiker, have you also seen [front] hubs shipped and missing a bearing on both sides? I was curious if the intent there was for similar reasons (which don't seem to make as much sense as with a headset application, where movement isn't as frequent or shocks might be transmitted while the bearings are in a specific location). Thanks for your response.

Conniebiker said:
In most situations the balls would be best without the retainers. The biggest example is bottom brackets. When they were open balls.
 
jasong said:
Conniebiker, have you also seen [front] hubs shipped and missing a bearing on both sides? I was curious if the intent there was for similar reasons (which don't seem to make as much sense as with a headset application, where movement isn't as frequent or shocks might be transmitted while the bearings are in a specific location). Thanks for your response.
I've yet to see hubs with missing balls . Maybe you just got screwed.
 
Conniebiker said:
Do you know Sydney from bikeforums.net?
You two seem so alike.
Sorry, but your claim that bearing retainers would "toast the assembly" if the BB wasn't rebuilt every 200 miles seems a bit outrageous to me also. Wasn't this the most common type of BB for decades before the cartridge-type?

My Shimano 600 BB from 1990 lasted 10 years and 10K miles without even so much as a single repack of the bearings.
 
dhk said:
Sorry, but your claim that bearing retainers would "toast the assembly" if the BB wasn't rebuilt every 200 miles seems a bit outrageous to me also. Wasn't this the most common type of BB for decades before the cartridge-type?

My Shimano 600 BB from 1990 lasted 10 years and 10K miles without even so much as a single repack of the bearings.
Alright so it should be 200 miles without significant decrease in performance under competitive usage. To me that is normal life. All conditions all year, 200-400 mi a month, more during XC race season. If I did not perform maint monthly if not bi-weekly I had shot bearing retainers. I have yet to see one in heavy usage that lasted. Plain ball will last considerably longer. The limiting factor is grit. If you ride clean roads constantly then it would be near indefinite. There is also quality factors. I have had the original set of large bottom bracket bearings in my 69 Schwinn with no problems but the ones that are commonly available in shops(which is not really that common either) are what I base my claim on.
Just because it was a common system at the time was not to say it was the best. They used square taper cranks back then(and my Raceface ones are good), but they are prone to many problems. New designs all but eliminated a lot of them. Cartridge bottom bracket's claim to fame is thier virtually grit proof seal, eliminating the main enemy.
On the missing balls: I have gotten a few that way. It is usually just one per hub, since the hub will still spin that way. I always repack my hubs out of the box anyhow so I just check in the process.
 
Conniebiker said:
Alright so it should be 200 miles without significant decrease in performance under competitive usage. To me that is normal life. All conditions all year, 200-400 mi a month, more during XC race season. If I did not perform maint monthly if not bi-weekly I had shot bearing retainers. I have yet to see one in heavy usage that lasted. Plain ball will last considerably longer. The limiting factor is grit. If you ride clean roads constantly then it would be near indefinite. There is also quality factors. I have had the original set of large bottom bracket bearings in my 69 Schwinn with no problems but the ones that are commonly available in shops(which is not really that common either) are what I base my claim on.
Just because it was a common system at the time was not to say it was the best. They used square taper cranks back then(and my Raceface ones are good), but they are prone to many problems. New designs all but eliminated a lot of them. Cartridge bottom bracket's claim to fame is thier virtually grit proof seal, eliminating the main enemy.
On the missing balls: I have gotten a few that way. It is usually just one per hub, since the hub will still spin that way. I always repack my hubs out of the box anyhow so I just check in the process.
Where do you dredge up all this inane and lame BS? :rolleyes: :rolleyes:
 
i think retainers are around due to a littel bit of tradition and a lot due to ease of assembly..
ancient ashtabula style (one-piece) cranks are just about impossibel to put together without retainers.
newer three-piece square taper bottome brackets are doable without retainers but they make the job quite easier
same with headsets
removing the retainers and adding a coupel extra ballz is an old trick to get a littel more life out of a slightly brinneled headset that 'hangs' in it's favorite position (strait ahead, usualy)
more balls and more surface area help with wear too
as far as hubs go, remember they are often designed to have a littel bit of room in them... packing them full up with bearings is not a good idea.. there should be a littel space

as far as old style bbs vs. new sealed cartridge bearing bbs:
sure the new ones will last longer between services.. they are sealed, afterall
but if you ask me, i'd rather have the ability to repack my old well designed three piece bottom bracket at home every once in a while
as opposed to tossing out the sealed one and buyin a new one or mucking about with bearing pullers and presses and junk like that
 
Yeah, it´s not really an issue, since the hubs always seem to come adjusted too tight and checking out what´s inside takes another minute. It just has happened with such frequency it seemed that perhaps an intention, unknown to me, was there.

Conniebiker said:
On the missing balls: I have gotten a few that way. It is usually just one per hub, since the hub will still spin that way. I always repack my hubs out of the box anyhow so I just check in the process.
 
Conniebiker said:
Alright so it should be 200 miles without significant decrease in performance under competitive usage. To me that is normal life. All conditions all year, 200-400 mi a month, more during XC race season. If I did not perform maint monthly if not bi-weekly I had shot bearing retainers. I have yet to see one in heavy usage that lasted. Plain ball will last considerably longer. The limiting factor is grit. If you ride clean roads constantly then it would be near indefinite. There is also quality factors. I have had the original set of large bottom bracket bearings in my 69 Schwinn with no problems but the ones that are commonly available in shops(which is not really that common either) are what I base my claim on.
Just because it was a common system at the time was not to say it was the best. They used square taper cranks back then(and my Raceface ones are good), but they are prone to many problems. New designs all but eliminated a lot of them. Cartridge bottom bracket's claim to fame is thier virtually grit proof seal, eliminating the main enemy.
On the missing balls: I have gotten a few that way. It is usually just one per hub, since the hub will still spin that way. I always repack my hubs out of the box anyhow so I just check in the process.
OK, now the 200 miles makes more sense if you were referring to BMX racing conditions. And, agree with your statements that contamination is the main issue to BB bearing life, and that on clean roads, life will be almost infinite. Of course, no one rides on perfectly clean roads all the time, which is why seals are a good idea.

Curious, but what brand of hubs are you buying where you find missing balls? Do you check the factory specs for the number of balls, and the amount of grease before proceeding to add stuff? If I'd spent good money on a new hubs, and they came without grease, or missing ball bearings, I'd return them under warranty at once.
 
The only times I have gotten the missing balls was a long time ago on Deore disc, discounted wheels. (but that wasn't why). I bought them and didn't really do anything with them for a long while, then I realized there was a problem after I lost the paperwork. It is really not that big a deal to me since I have stock of all the more common bearing sizes.
I much prefer Sunrace Juju sealed cartridge hubs. They go years without discoloring the grease since there are 2 layers of seals in addition to the cartridge seals themselves. They did however prompt me to repacking out of the box. The first race I did, (disc brakes mind you) I melted all the grease right out. Apparently they forgot to specify to thier distributor about the application. I always repack with Valvoline 614 automotive now. Never had a problem and I check them every 4kmi or so.
 
It's possible the hubs with the "missing bearings" could just be designed that way. Maybe they were built from the factory wrong. Best way to find out would be to call the factory to find out. I'm sure they would send new bearings for it or send a new hub.
As far as the headset is concerned, since the bearings do not rotate consistently and the retainer would tend to hinder bearing movement the chances of brinneling are higher.Removing the retainer would allow the bearings to individually move easier (hence removing a bearing(s) to give more space) Also remember that a properly adjusted headset is critical to preventing this. I don't think hub brinneling is a factor due to the bearings constantly rotating and changing location on the bearing race (properly adjusted of course).
 
DHK, I think these were a Sora hub and a velocity hub. Both were inexpensive (<$100) wheelsets. Where have you seen a spec for amount of grease? Even reading through Park/Barnett's they don't give a spec for amoutn of grease, and ball count is implied as enough to fit (of the right type, which seems more or less universal excluding Campy). Ball bearing count seems to be rather standard in front hubs of this and higher caliber ball/cone design (10 3/16"). Have you seen exceptions?

dhk said:
Curious, but what brand of hubs are you buying where you find missing balls? Do you check the factory specs for the number of balls, and the amount of grease before proceeding to add stuff? If I'd spent good money on a new hubs, and they came without grease, or missing ball bearings, I'd return them under warranty at once.
 
jasong said:
Where have you seen a spec for amount of grease? Even reading through Park/Barnett's they don't give a spec for amoutn of grease,
Bwahahahahahahahahahahahaha!! Do you want that in ounces, grams or scruples and to 4 significant places?
 
I'm putting back together a pair of 70's vintage racers with Campy NR gear that have a non sealed square tapered axle BB. Both BB's were in excellent working order. The seal between axle and cup is fairly tight, and the cups are threaded around the axle hole so the rotation of the axle tended to force out any debris. The typically thorough design that Campy was known for back then. As I recall, one repacked their BB every 1-2k miles with that style, and that was probably too much.

Big advantage of sealed BB's is that you don't have to repack the bearings, and they are less prone to dirt.

I am curious about the notion that a cage reduces the number of bearings. If the cage is properly designed, it shouldn't reduce the number of bearings, as it holds them in near perfect alignment. I have repacked both caged and non caged bearings, and can safely state that the non caged headset bearings are a pain to deal with. Little devils go bouncing everywhere...


dhk said:
OK, now the 200 miles makes more sense if you were referring to BMX racing conditions. And, agree with your statements that contamination is the main issue to BB bearing life, and that on clean roads, life will be almost infinite. Of course, no one rides on perfectly clean roads all the time, which is why seals are a good idea.

Curious, but what brand of hubs are you buying where you find missing balls? Do you check the factory specs for the number of balls, and the amount of grease before proceeding to add stuff? If I'd spent good money on a new hubs, and they came without grease, or missing ball bearings, I'd return them under warranty at once.
 
jasong said:
DHK, I think these were a Sora hub and a velocity hub. Both were inexpensive (<$100) wheelsets. Where have you seen a spec for amount of grease? Even reading through Park/Barnett's they don't give a spec for amoutn of grease, and ball count is implied as enough to fit (of the right type, which seems more or less universal excluding Campy). Ball bearing count seems to be rather standard in front hubs of this and higher caliber ball/cone design (10 3/16"). Have you seen exceptions?
Have never seen specs for the amount of grease to use in bike hubs, only general guidelines to use a layer in the cups so the balls are about 1/2 covered when they are pressed into the grease. My guess is that many people overpack the bearings with grease, resulting in more drag in the hub and seepage past the seals.

For sealed cartridge bearings, have seen drawings which specify the weight of grease to use in ounces. Believe the general rule is about 1/3 (33%) grease fill for most cartridge bearings used in industrial equipment.

Have no idea what's standard for the number of balls in a cone hub; was just curious as to how you'd know balls were missing if you didn't have the specs. Regular cartridge bearings aren't completely full of balls due to the fact they can only be assembled with a certain number which need to be kept evenly spaced via retainers.
 
jasong said:
DHK, I think these were a Sora hub and a velocity hub. Both were inexpensive (<$100) wheelsets. Where have you seen a spec for amount of grease? Even reading through Park/Barnett's they don't give a spec for amoutn of grease, and ball count is implied as enough to fit (of the right type, which seems more or less universal excluding Campy). Ball bearing count seems to be rather standard in front hubs of this and higher caliber ball/cone design (10 3/16"). Have you seen exceptions?
Have never seen specs for the amount of grease to use in bike hubs, only general guidelines to use a layer in the cups so the balls are about 1/2 covered when they are pressed into the grease. My guess is that many people overpack the bearings with grease, resulting in more drag in the hub and seepage past the seals. Would like to know the exact amount of lube Shimano or Campagnolo recommends for hubs, if it's published anywhere.

For sealed cartridge bearings, have seen drawings which specify the weight of grease to use in ounces. Believe the general rule is about 1/3 (33%) grease fill for most cartridge bearings used in industrial equipment.

Have no idea what's standard for the number of balls in a cone-and-cup wheel hub; was just curious as to how you'd know balls were missing if you didn't have the specs. Regular cartridge bearings aren't completely full of balls due to the fact they can only be assembled with a certain number due to the clearances between the inner and outer races.