Technical FAQ: Saving freehubs, aero is everywhere

Dear Lennard,
I’ve destroyed every aluminum freehub I’ve ever ridden (see attached photo). I gouge the cassette into the hub so far I sometimes can’t get it removed to replace worn gears without filing off metal chips.

Is this a regular issue for other riders? How do you prevent this kind of wear?

I know that steel freehub bodies don’t have this issue, since the harder metal doesn’t deform against the cassette, but most road wheels (especially nice ones) don’t use steel freehubs.

Any maintenance and purchasing advice would be very appreciated.
— Ken

Dear Ken,
Well, the tearing-up of the splines and the difficulty of removing the cogs are as sure as death and taxes if you’re a powerful rider using an aluminum freehub body that is Shimano/SRAM compatible and you’re using separate cogs. The problem is, besides the soft material, that the splines are so low.

It’s hard to imagine one of those cogs being able to plow completely through the wide, indexing spline on the freehub, but I have seen complete spline failure on steel freehub bodies that have, in the interest of weight savings, removed the centers of the splines, leaving thin, separated strips of steel. The circumstance was similar: separate steel cogs pushing into the splines, but in this case they kept going until they tore right through all of the splines and just spun freely on the freehub body.

The cheapest way to eliminate the problem is to not pedal hard; this doesn’t happen to 100-pound riders. But you probably don’t want to do that, so here are two more expensive options that will allow you to pedal as hard as you want.

1. Use a SRAM Powerdome cassette. The only engagement of the 10 largest cogs with the cassette is the set of splines on the largest cog, and those are aluminum and won’t damage your freehub body in the least. Check out the photo from the back side on the link. The first cog also has splines, but you’re generally too torque-limited to get that one to dig into the freehub body anyway, and even if it did dig in, you could still remove it easily.

2. Use a Campagnolo freehub body and cassette. The splines are much deeper on Campy freehub bodies, so the pressure is distributed more, and the cogs don’t tend dig into them. If you’re using an 11-speed drivetrain, a Campy wheel is interchangeable with a SRAM or Shimano wheel, generally without derailleur readjustment, because the spacing between cogs is the same with 11-speed cassettes from any of the three brands.

If you’re going to use a Shimano cassette, get one that has the most cogs riveted to aluminum carriers. This would be Dura-Ace, and the largest five cogs will be integrated onto two aluminum carriers; this guarantees that at least the innermost five cogs won’t dig into the freehub body. But judging by your 11-speed freehub body, it looks like only the smallest six cogs have dug into it. Low-end Shimano cassettes have (or at least had) three long, thin bolts holding the entire cog stack together, which should eliminate this problem, but I don’t believe this exists in 11-speed form.

There was a brief time in the 10-speed era when Shimano tried to address this issue by making Dura-Ace cassette cogs have longer spline teeth and Dura-Ace aluminum freehub bodies have deeper splines. But the market rejected the incompatibility with prior generations and with Ultegra and 105. So Shimano’s subsequent retreat on changing the spline configuration took us to your current predicament.
― Lennard

Feedback on last week’s column:

Dear Lennard,
I saw your response to the question of whether the gloves used by the U.S. women’s track team might have slowed them down. According to an MIT wind tunnel study, gloves slow you down more than a non-aero front wheel in a TT. I suspect the same is true on the track. The article was published in a rival magazine but the data is solid.
— Ted

Dear Lennard,
Here is my comment on your note on the women’s track team gloves.

[related title=”More Technical FAQ” align=”right” tag=”Technical-FAQ”]

Air drag can sometime be reduced by increasing the surface roughness of the object. This is why the golf ball has dimples. In a range of speed which depends on the object size, generating turbulence in the boundary layer, the air near the object, moves the separation point: the point where the boundary layer separate from the object downstream. The roughness increases the skin friction but decreases the pressure drag caused by the separation. At high velocities the boundary layer is already turbulent; adding roughness increases the total drag.

The U.S. track bike chain was moved to the left to reduce drag, this is an indication that a lot of time was spend in the wind tunnel, so maybe the loose gloves were there to reduce the drag.

It is evident that the U.S. running support team has done a lot of work in the wind tunnel too. At Rio the US running sprint track team tops and shorts have little pink bumps just where one would like to trip the boundary layer. The fabric roughness also has been strategically adjusted.

The speeds of the foot sprints are on the order of 100m in 10 seconds, or 10m in 1 second, or 3600 X 10m in an hour, 36 km/hour, 22.5 miles/hour. We cyclists know that air drag at that speed is not negligible. A time difference of 1/100 of a second over 10 sec is a difference of 0.1 percent over the total time. Reducing air drag by a few percent is thus very worthwhile, unless you are Usain Bolt.

Even the U.S. long jumper has dimples on his socks!
— Pierre

Dear Pierre,
Thanks for that explanation. The boundary layer argument is always one which that is in a dance with increasing the size of the object. I’m pretty sure that if you strip the fuzz off of a tennis ball, the smaller inner ball will not fly as far as the larger, fuzzy version of itself. But when you increase the size of the object in order to create the roughness to disrupt the boundary layer, there is obviously a point where size trumps surface roughness and drag goes up rather than down.

I think it is particularly interesting given your example of the US women’s 4X100 relay team, because my wife and I frequently wondered aloud to each other while watching the speed events on the track — the ones with starting blocks, whether running or hurdling — about what big hair some of the competitors had and how it seemed sure to be slowing them down. I can imagine a boundary-layer effect in that instance, too, but it’s very hard for me to imagine that some of the huge hair on some very fast Jamaican and American sprinters wasn’t, on balance, slowing them down, and not just from aerodynamic drag, but sometimes also from having to accelerate all of that additional mass.

So the question for me in this instance of the pursuit team’s gloves is: Which wins out, size or surface roughness?
― Lennard

Dear Lennard,
The best explanation of why left-side-drive offers some advantage that I’ve seen came from LOOK’s marketing material for their R96 frame: they suggest that it’s simply because the left side of the bike has a lower average speed than the right side.

My calculations, based on a 42mm chainline, suggest the speed difference is 0.21%, meaning the drag difference is most likely under 0.5% (that’s half a percent of the drivetrain’s drag, not the whole bike), which isn’t a lot given the drawbacks.

The main drawback is that for now at least you have a very limited number of rear wheels which are safe to use with a left-side-drive cog. If one of those wheels is the best for drag and frictional losses then great, but it’s more likely that the best is in the larger part of the market; you could well take a bigger marginal loss on those factors than your marginal gain from moving the drive side to the slower side of the bike. There is a similar deficit of choice for left-side-drive cranks. It’s possible also that BBs are asymmetrically designed at that level, I don’t know.

Now that Felt and LOOK have both built left-side-drive track bikes it will be more worthwhile for wheel and crank manufacturers to build products that can perform that function safely, and in a few years the drawbacks I listed above may disappear entirely.
— Phil

Dear Lennard,
I broke my collarbone a few years ago at the end of a great ride when a dog ambushed us from a ditch. I thought I had hit the dog. But the more I have thought about it, I think I panicked and locked up the front brake. I hardly use the front brake now and have told myself to never lock it up for any reason. A few months ago a small dog came at me and I did not panic, no front brake. I hit the dumb dog, and he managed to wiggle out from under my wheel before I rode over him and even though it pulled my wheel so that I went off the road, I did not fall and all was well.

I think many broken collarbones and other injuries could be avoided with front anti-lock brakes. Even the pros panic and lock up their front wheel from time to time.
— Forrest

Dear Forrest,
I think you’re right. Right after Interbike last year, I wrote about the Ultra Cycle Brake Safe, which makes any cable-actuated brake into an anti-lock brake. I don’t know whether the product ever came to be, though.
― Lennard