The cogs in the animation would move in the same way if they were rollers and not attached to a carrier, dependent on friction instead of gear teeth (but note that cartridge bearings have ball cages that act like the planetary carrier in the diagram). Cup/cone/ball bearings are angular contact, which skews the rotational axis of the balls but movement is still basically the same. A deep groove ball bearing as used in a cartridge bearing bottom bracket is not exclusively radial, it also has an element of axial load capacity just like a CCB bottom bracket, albeit less than an angular contact bearing. The deep groove differs from CCB in that the "cup" and outer race are separate parts (unless it's a press fit where the bearing seats directly in the frame).

It's more of a "not self loosening" effect but that's the idea, supposedly invented by the Wright brothers who were bike mechanics before they became aeronautical engineers.

Precession is the description of why things don't just line up, but it isn't an explanation of the forces involved: The spindle, balls and cup form an impossible gear train because the cup is not allowed to move at the rate the bearings are trying to turn it. If the spindle, bearings and cup were cogs, the spindle would be locked up. A sun and planetary gear system wouldn't be useful if the center input gear could turn without the outer ring turning in kind, and BB cups don't turn.

The reality is that the balls are smooth and they are not rolling on either the spindle or the race, they slipping to make it possible for the spindle to turn without the cup turning in kind. Those friction forces of the balls trying to roll but being forced to slip like a car tire on ice is what pushes the cup looser or tighter.

It is also a reminder why preload is foolish - the balls are already backslipping, and preload just increases that friction more.


I don't see why you think your illustrations help clarify anything, but I don't think you get what is actually happening in a bearing system.

(bold) Respectfully, whoa-whoa-whoa. Bearing balls, when functioning properly, are rolling contact, that's actually how they are categorized as bearings. If they are not rolling but staying stationary and sliding against one or both races, that is known as bearing ball skidding, and it will flat spot the bearing ball in nothing flat, and then function is impeded from thereon. That is part of the recommendations for a slight preload on angular contact bearings, it reduces skidding. This is from online, a bearing company:



If you have a bearing without *rolling* elements, but sliding contact, that is a bushing.

Regarding precession, it's due to a radial force, that rotates direction with respect to the threads. The BB threads are set up for precession to tighten the BB cup or cartridge. If you have a fixed gear bike and they were to ride in reverse with a good amount of pedal force, the BB would back out. If you have a mid-drive bike that transmits motor power mid-spindle, and the rider is not pedaling, or just has weight on the pedals with no turning motion, the BB will not back out, because the radial force on the BB is not rotating orientation, it's all one direction from chain tension or rider weight. But if, on that same BB, the ball bearings actually seize up or stop rotating so that they are sliding on the BB, then the force on the BB cups is opposite intended direction and it will back out if the frictional force is sufficient.

Ball. Bearings. Roll. That's why they are round.

I'll have to look at your time of post, it may have been very late local to you. I've made mental errors in the wee hours.

The real mystery is why Italian spec bottom brackets have a metric cup diameter (36mm) but an English thread pitch (24 tpi).

For both the cups and the shell? I know of engineering applications where one part had slightly (IIRC, 5%) different thread pitch in order to be self-locking.

The bearings aren't static, but there is simply no way that anything round can strictly roll when the spindle race diameter is different than the cup race diameter. Something has to slip, as you can clearly see in the gif I posted.

And really, how can you read about "excessive skidding" without wondering why there would be any skidding? That skidding is what has to happen for the reasons I stated.

Bearing preload makes sense when the bearing is going to support thousands of pounds on the axle, heat up greatly or have large load variations. None of which apply to bikes. You have all the information necessary, but keep coming to the wrong conclusions. And you didn't even understand the bearings you are preloading.

I posted at 10am. Apparently you are equally good with time conversions.

It isn't self locking. It is, as stated, English thread with a metric width cup.

The bearing balls rotate. At the outer periphery, i.e., the outer race or cup, they roll with a surface movement at a given rate. Now transfer that same ball rotation rate to the bearing inner periphery, i.e., the inner race or spindle; You will have the same surface rate, no slipping, however because the inner race is smaller diameter, it will have a *higher rotational speed*, revolutions, than the speed differential at the outer race. If, for example, the inner race contact diameter is 1/2 the outer race contact diameter, the inner race rotational speed with be 2X the speed differential compared to between the balls and the outer race.

The Nuvinci/Enviolo variable speed hub uses this principle by varying the contact diameter to vary the speed ratio, with no ball slippage at any of the contact points. It is, in effect, a variable race ball bearing.

There is no rotational rate that is going to perfectly match both the precession of the balls, the rolling rate for that precession on the outer race and the rolling rate on the smaller inner race. So something has to slip to a degree. Sorry.

Please see my second paragraph to last post, just added.

Nuvinci has four out of four moving surfaces. A BB has two out of three. The Nuvinci does what the BB should not - turn the cups. The BB wants to turn the cups, but they can't, so the bearings slip fractionally instead.

Nuvinci: Fair enough.

Ball bearing:

If you watch the balls for awhile it will be clear that their outer circumferences are not able to roll on the static outer race or the moving inner race at the same rate. You can see that the red dot on the ball is not moving as quickly as the red dot on the inner race. If all three had teeth, the structure would not turn at all.

Yeah that's not a very good animation, I've seen the same elsewhere on the web, I think it's a copy of a copy of a copy.

I haven't ruled out that you're right, by the way, but that just seems contrary to rolling bearing theory. But I need to prove it if I'm right or wrong.

I think that you believe, that the rolling contact speed between the spinning balls and inner race, must then be the rolling contact speed between the balls and outer race, because it's the same ball, and the outer race is fixed, except that can't be because the outer race is larger diameter. But in addition to spinning, that ball is also translating, and that, I think, compensates for the contact speed differential. I haven't quite got it mathed out yet. What I got is (for this specific animation):

For each 1 revolution of the inner race, each ball rotates 6 times. That means a 6:1 ratio in diameters. To get the outer race rolling diameter, it's 6/6 + 2/6 = 8/6 or 4/3 times the inner race diameter.

That's it so far. I'm straining my brain to figure out how to take into account the translation of each ball, and whether added or subtracted to the rolling contact due to the balls spinning. If it were a linear bearing, rail moving balls over a fixed rail, it's easy, it's the "2 for 1" moving rail translation versus the ball translation. But with diameters, it's going to take more thought.

EDIT: I think that 6:1 ratio above based on rotations is way off. The race diameter ratio is a lot closer to 3:2, with balls at 0.5 diameter.

I agree that it doesn’t matter that the inner and outer race contact circles are different diameters

I kind of think it’s more to do with the point of highest contact force between the bearings and outer race moving around part of the circle during the power phase.

And probably the fact that they’re not perfect spheres contributes, and also have to have a tiny bit of room to move

I've seen it suggested that after WWII Italy received American machine tools to help with economic recovery, and they were not capable of cutting metric threads.

If they were rolling on both races at the same rate the bearing would need to have infinite diameter i.e. be a straight line. Because they roll at different rates on each race the balls also circle the races, at a rate determined by the difference in the two race diameters.

That's hilarious, I actually had a picture in my brain of an old South Bend, no metric, but hadn't put together the connection. That's not out of the realm of possibility. And if not from the USA, British machine tools. But USA machine tool production was going gangbusters.

I can't count the number of USA machine tool builders that no longer exist. Most named after the city where they started.

You also said that this is a cartridge bearing bottom bracket, so it's quite modern. I'm surprised that you can't get an inexpensive tool to tighten it correctly. What kind is it?

Yes, both the cups and the shell. Not to mention, the Whitworth thread angle and profile used with Italian thread spec.

I hadn't realised they use Whitworth thread form, so more likely British machine tools than American, and possibly pre-WWII?

And if the bearing seize, they go from the nice planetary Kontact gave us to a locked system that promptly unscrews cup. So keep your bearings lubed. (Pedal bearings can do the same and unscrew the pedal.)

It's made by a somewhat unknown company called first components (Taiwan based I believe). There are not many companies that make eccentric BB's for threaded shells, and this is the only one I know for a T47 shell.

I said in a previous post that its eccentricity was irrelevant. What I meant is that the point of my original question was simply to understand the general principle behind the reasoning of reverse threading for bottom brackets since it seems counter intuitive.

However the fact that it's eccentric could have contributed to my specific issue of it coming loose.
Maybe I just didn't tighten it properly in the first place, but it's also possible that the clocking of the eccentric cup has an effect on it. If you were to draw an imaginary vertical line that splits the BB in half, the spindle is currently positioned in front of that. I suspect that when I'm mashing and apply downward forces on the pedals the position of the spindle might want to unthread the cup. I put some Teflon tape on the thread and tighten the cup as much as possible, but if it loosens again I might want to try a slightly different gear/chain length combo so that I can clock that spindle behind that line.
The tool I would need to tighten it (and torque it), other than the manufacturer provided spanner wrench, is a 64mm socket. Not very common and very expensive if I can even find it.
Anyway a picture says a 1000 words

New comments added to previous diagram:


This pictured example would represent the non-drive (left) side of the bottom bracket; The pedaling force forward is to the left, counter-clockwise, and thus same is the closest contact point between the cup (blue) and the shell (red) due to rotating radial load at the bottom bracket, indicated by the green arrow (assuming any radial slack between the two threads). But then notice how the precession causes the cup to slowly rotate opposite, clockwise, to the right, thus the left side of the BB requires right hand thread to keep tight. The drive (right) side is opposite, precession to the left, so requires left hand thread.