Velo News

  1. Home » Bikes and Tech » Technical FAQ » Technical Q&A with Lennard Zinn: Old-school rims and bearing drag

Technical Q&A with Lennard Zinn: Old-school rims and bearing drag

By Lennard Zinn

Old-school rims at Paris-Roubaix
Dear Lennard,
I noticed that all the teams for Paris-Roubaix used old-school tubular rims for the race. How come they don’t use the fancy composite tall rims as they do during the rest of the season? Did it have to do with their stopping power in the dry, dusty conditions or the punishment of the cobbles?

Dear Peter,
Both. Carbon rims, especially deep-section ones, are so much stiffer that they are likely to crack rather than flex and bend like an aluminum rim would. In a crash, jagged carbon edges are worse than bent aluminum. And of course, replacement cost is much higher. And reduced stopping power and increased brake grabbiness are characteristics you do not want to have to deal with in Paris-Roubaix, whether it is wet or dry.

Drag could be bearings or poorly faced BB shell
Dear Lennard,
I really enjoy your tech coverage and rely on your manuals for mechanical guidance in my home shop. Quick question regarding your piece on bottom bracket bearing drag. I have a Truvativ Gigapipe BB on a Rouleur carbon crankset and it is absolute junk. After repeatedly adjusting it per Truvativ’s specs, it is so tightly compressed that the crank barely spins when I have removed the chain and apply light pressure. While riding, it feels much more difficult to pedal than a 3-year-old Octalink BB I have on another bike. So, what’s wrong with my Gigapipe? Is it just a worthless product or is mine perhaps defective? I rode it for about 300 miles to try to break it in, but then I got tired of feeling like I was pedaling in cement and just kept riding my old bike, which works great. I couldn’t deduce from the data in your story what the net effect of drag at the BB is in terms of overall power output. What do the variations in the data mean when interpolated out over the course of an hour ride at LT?

Dear Andrew,
I don’t know what’s wrong with that BB. Perhaps the bearings themselves are suspect; maybe they are not up to spec. Or, your BB shell could be so poorly faced (i.e., the opposite faces are not parallel to each other and perpendicular to the bore of the shell) that there is not enough flexibility in the system to deal with it (you should have your shop stick another integrated crank in there and see how it spins!). One other possibility is that your BB shell is overly long (should be 68mm wide for English thread), and the system is putting a side load on both bearings, while that system should only be side-loading the left one. Or, you could just be over-reacting to friction of the bearing seals, which will slow the spinning of the cranks by hand but is providing very little total drag, and that drag is not load-dependent.

I think that the GXP system is actually a superior one relative to other integrated-spindle/external-bearing (IS/EB) systems when mounted on an imperfectly faced frame. But even GXP has limited ability to deal with a poorly faced frame. Most IS/EB cranks rely on the consumer to set the bearing side load to eliminate lateral movement. The consumer tightens a bolt on the end of the non-integrated arm (generally the left arm; Race Face is the exception to this) to take out the slop and then tightens tangential pinch bolts to secure the arm to the spindle. This means that both bearings are side loaded.

The Truvativ GXP design pinches the left bearing’s inner race (which has a smaller ID than the right bearing) between the left arm and a shoulder on the spindle, thereby isolating the drive-side bearing from axial loads. (Unlike with pinch-bolt systems, you tighten the GXP left crank on to high torque as you would your Octalink.) The GXP right bearing has more radial loading already due to the chain being on that side; the left bearing takes some axial loading, but since its radial loading is less, the bearing wear is more even on the two sides, and the side load is less. Because of this isolation of lateral loads on a single bearing on one side of the shell, GXP can deal with minor problems with facing better than an IS/EB crankset that has the same bearing ID on both sides, a constant-diameter spindle, and side loads both bearings. If there are any facing problems with one of these systems the freedom of movement will be poor. The bearing wear rate will also be very high. Frame manufacturers had gotten used to not worrying much about BB-shell facing during the era of sealed cartridge bottom brackets, since neither cup, or at least only one cup, had a lip that tightened against the face of the shell. Performance riders really worried about facing of their BB shells in the days of loose-bearing bottom brackets, because both cups were forced into alignment with the face of the shell due to a lip or a lockring on the cup. But now it is something that frame manufacturers need to be careful about again.

As for the increased fatigue or decreased speed due to bearing friction measured in our test, I think it’s trivial. As you saw from our graphs, the GXP BB actually had the highest drag of any of them at 0.41 watts (same as a couple of the Shimano IS/EB cranks) with the seals and at 0.27 watts with the seals removed. But presumably at LT, you are putting out an amount of power that dwarfs these numbers. Say you do 410 watts at LT. The GXP BB drag we measured is 1/1000th of this, or 0.1 percent. If you think you can measure the speed difference after a drop of 1/1000th of your power, you have a pretty accurate stopwatch. Of course, as I said repeatedly in the article, we were measuring this in a very low-load situation, like spinning at 90RPM super easy. So there may be, and likely is, a multiplier in this that increases the power loss.

Feedback on previous columns
Dear Lennard,
Regarding the April 10 letter “Carbon and cork,” where the guy wanted to change out his upcoming carbon rims with his aluminum rims freely, since he most likely had to readjust the pads anyway, it takes only a few seconds extra to change out the pad. You said this and mentioned the ease (Shimano, SRAM) and difficulty (Campy) of this. I think, forget changing the pads. Have a set of spare pads already in spare holders. Just change out the holder. Even easier.

Dear Michael,
Great idea. Pricier, but if you’ve already popped for all-carbon wheels, that expense will be irrelevant, and the time savings over many changes will be enormous.

Dear Lennard,
This is in response to the article on carbon seat posts seized in aluminum seat tubes. I have an additional tip that proved very successful. I first tried heating as described, but no luck. Then a buddy suggested creating an even bigger temperature differential by crushing little pieces of dry ice and pouring them into the seat tube. Leave it for 10 minutes and let it get real, real cold on the inside, then heat on the outside. Voila — the post slipped right off!

VeloNews technical writer Lennard Zinn is a frame builder (,a former U.S. national team rider and author of numerous books on bikesand bike maintenance including the pair of successful maintenance guides”Zinnand the Art of Mountain Bike Maintenance” and “Zinnand the Art of Road Bike Maintenance” as well as “Zinn’sCycling Primer: Maintenance Tips and Skill Building for Cyclists.”
Zinn’s column is devoted to addressing readers’ technicalquestions about bikes, their care and feeding and how we as riders canuse them as comfortably and efficiently as possible. Readers can send brieftechnical questions directly to Zinn (’s column appears each Tuesday here on