I just noticed a discussion about this on one of the sale threads, but this is something I've been meaning to ask about and since it's a discussion here seemed better than there as a place to ask.

I recently came into possession of a vintage Phil Wood bottom bracket. It was on an early 70's Urago I bought. The metal sleeve was coated in some grime, but it cleaned up pretty well. The red paint on Phil's signature was even visible midway through my cleaning, though I seem to have inadvertently eradicated that.

Anyway, it's my first hands on experience with a Phil BB and I was wondering how close mine is to normal (in whatever way you choose to interpret normal).

First off, the bearings on one side will turn independent of the sleeve, but the outer shell of the bearings on the other side seem to have bonded to the sleeve a bit so that if I hold the outer shell and the sleeve I can't turn them in opposite directions. Near as I can tell, that shouldn't actually matter because the spindle turns freely and I don't think the outer shell of the bearing cartridge is actually supposed to spin. This was using French threaded cups, so its even possible some Loctite was applied to keep it from doing so.

More relevantly, the spindle spins freely, as I said. In the other thread [MENTION=340794]Dfrost[/MENTION] mentioned the viscosity of the bottom bracket. That's the real locus of my curiosity. The Phil BB I have almost feels dry. I don't feel any roughness (really, none), but I also don't really feel any significant resistance. Granted, this is under no pre-load in my description, but I would expect to at least be feeling the viscosity of some grease and I don't. Is that normal?

And if that is normal, do these things really last forever? I mean, I understand cup and cone bottom brackets lasting forever with proper maintenance but my experience with cartridge bearings has generally been that they last a long time and then you replace them because you can't really do proper maintenance.

Ass long as you don't abuse them, they could last the lifetime of the bike. That said, I came into an abused Phil BB that had solidly rusted into an ancient Legnano frame. I literally had to hammer the cartridge out of the frame, and in doing so knocked the bearings out of their mount. Not a big deal, since they needed replacing anyway: The races had completely rusted through.

Phil does not encourage home replacement of the bearings, but they do offer to rebuild their bottom brackets for a nominal charge. I decided to do it myself, anyway, and it wasn't difficult.

So is the lack of viscous feel normal?

Absolutely, and I continue to be amazed by the operating lifespan of some dry-feeling Phil bottom brackets and especially hubs, some of which haven't been serviced in over 30 years.


Go ahead and keep using them. Bearings eventually become rough and need replacement, but it might be a lot longer than you think.


I've popped out shields and greased old cartridges but it isn't uncommon for some bit of grit to be felt floating around as the bearing turns afterward, so the bearing surfaces may actually suffer versus leaving things alone.
Where the cartridge has been removed from an assembly, it is much easier to assure complete cleaning than if it is still pressed in place somewhere. I advise that unless complete flush cleaning can be assured that the shields be left alone, as a very tiny bit of grit can ruin this type of bearing's downsized, heavily-loaded surfaces.

The bearings came with a lip contact seal, use eventually wears that away, so in time the drag is reduced, I found that out decades ago with my first Phil bottom bracket. I have one of the original design units that exhibits the same one side is locked, other side rotates, I set the last torque to install on the rotating side. That just seemed reasonable.

Not sure it matters much, but can't hurt.

True about the lip seal. A well-used but uncontaminated cartridge bearing will be nice and smooth.

The outer races of the bearings on a Phil BB are "pinched" between the center sleeve and the mounting rings. When the rings are properly torqued in place, the bearings won't move. Doesn't really matter whether they stay attached to the sleeve or not when you remove them. Just make sure a loose one is seated in the sleeve before tightening the mounting ring.

The sleeve's width is precision milled to match the spindle's bearing saddle width plus the bearing width, so the bearings aren't side loaded when installed. That's bad for radial-contact cartridge bearings. Make sure the bearing seats are nice and clean on both the rings and the sleeve. Theoretically, overtightening the rings could bulge the center sleeve enough to cause some bearing side load, so don't overdo it.

When I install a Phil, when I have the rings close to tight enough, I give the ends of the spindle a good tap or two with a hammer to align everything. Then I tighten to spec. If it's still a little rough, another tap or two with a hammer should smooth things out. If not, I take it out and troubleshoot.

I think the relative "feel" of bearing assemblies is interesting! (That's the nerdiest thing I've said in a long time.)

You mention viscosity, but I'm really more interested in drag/friction/resistance (I'm not sure of the proper engineering term here, maybe "mechanical impedance"?) Like a few grams, I don't think drag matters to all but the most elite cyclists. But we don't let inconsequence stop a good debate here, do we?

I've noticed a difference in drag in bottom brackets. Cartridge units, especially new ones, seem stiff to me. I assume this is due to 1) new grease viscosity 2) possible friction/tightness in new parts/seals and 3) bearing preload. Some older cartridge bbs I've encountered seem more "broken in" and spin easier.

Overall, cup and cone bbs seem to have less resistance to me. They don't usually spin quite as smoothly though, depending on their quality/condition. I also suspect that many cup and cone bottom brackets aren't installed with proper preload.

If your bb feels dry you should probably lubricate it. As to Phill bbs, my guess is that two of their main benefits are longevity and maintainability.

If you want to nerd out more on friction: bb drag, cartridge bearings.

Now that's a good observation. I have this ideal in my mind that with a bike in the workstand the front wheel should show a strong pendulum effect, swinging back and forth over the heaviest spot in the wheel for quite a while. I've got some wheels that do that. As often as not I find that it's because I didn't adjust the cones properly. New hubs almost never do it when properly adjusted. Sometimes after the races are worn in it happens.

When it comes to brand-new hubs or wheels withg cup/cone bearings, I find it odd that they are NEVER adjusted properly at the factory.
These bearings are always adjusted at the factory with snug bearing pre-load, when what is needed is a finite amount of freeplay that should only fully go away when the full tension of the quick-release skewer is applied. Using a much lighter setting on the quick release adjusting nut should leave the wheel with some freeplay discernable at the rim.

I suspect that the average rider would be shocked to know how much resistance and roughness can be felt at the axle when the bearings are left in the factory-adjusted condition. Using a pair of loose dropouts or thick washers to simulate clamping the wheel into a frame will show the rider how tight and rough that the hand-turned axle's rotation becomes when the quick-release lever is secured, and makes it pretty clear how damaging that this is to the axle bearings, most especially the cones. The balls are typically chromium steel and the cup races have so much more surface area than the cone races, so the cones lose their case hardening first, just like a secondary asphalt roadway breaks up from being heavily used by heavy big-rig trucks.
One should never expect such over-tightened bearings to break in without damage. The best hubs tolerate this abuse for what is really an incredibly long time considering the extreme-pressure loading of the cone races, but it does shorten their lifespan greatly since they are being overloaded by huge factors (versus use with proper adjustment).

Bottom brackets and even solid/nutted (non-quick-release) hub axles should have some preload that is felt as turning resistance well beyond that imposed by even the heaviest of greases. There needs to be some pre-load since the bearing balls and races will compress locally under even the smallest force just like a spring does, and we don't want freeplay to occur under ordinary loading.

Headsets are another matter because the steer tube shortens in response to the expansion of the quill (according to Poisson's Ratio), which tightens the bearing preload adjustment quite measurably, sometimes leading to immediate damage during a first post-adjustment ride!
As well, the fork crown rocks due to steer tube flex in response to loading and road shock, which has a "pinching" effect on the bearings toward the front of the upper and lower headset and thus compresses the bearings toward the front of the headset with great leverage as the front of the steer tube shortens while the rear of the steer tube lengthens.
So the headset must not be overtight, at all, making headset adjustment the most sensitive of all the bike's bearings, and final adjustment for free movement should thus always be carefully checked with a quill tightened in place in the steer tube.

I've heard it suggested that hub bearings are over-tightening from the factory to prevent rattle-induced damage during pre-use handling. I don't know if there's anything to that. Maybe it's just easier to set the machines to do it that way.

Or a pair of 10mm box wrenches work quite well as “dropout simulators” clamped against the locknuts by the QR.

I don't think there's any damage possible from 'pre-use' handling. Its easier for machines to assemble parts 'tight' than it is 'loose'. Customers will complain about loose bearings all day long but they very rarely complain about tight bearings. Tight bearings will have a tendency to loosen up while loose bearings will....loosen up.

Phil stuff is nice but a sealed bearing is a sealed bearing (more or less). If it feels dry its dry. Phil stuff may or may not last longer. Owners of higher end anything tend to take care of things better than owners of lower end things.

Adjusting sealed bearing preload is exceptionally difficult. Most people, including mechanics don't know how to check and most don't know how to adjust it. Checking is an in/out movement not a back n forth movement.

2 round objects make contact at 1 point only. If they didn't they wouldn't be 'round'.

When clamed down cones don't move. All pressure from the bearings is literally concentrated at the bottom of the cone. Races on the other hand are constantly moving, with each rotation of the wheel the entire cup bearing path is touching the bearing.

Cone wear is heavily concentrated at one point. Cup wear is spread out.

Way back in the Campagnolo headset installation instructions it is stated to complete the adjustment After the stem is installed.
We had a set of dropouts at the shop and clamped them in the hub/ skewer assembly to evaluate and perform the final adjustment with the skewer closed.
The reason to perform the final tightening of the Phil lock rings at the side that slightly spins was to reduce the false torque that the friction of the lock ring against the stationary bearing might transmit.