I received the front and rear wheels of a Ksyrium Pro (Mavic wheelset) from a customer.
They'd acquired it at an auction and wanted me to inspect it before use.
Let me start with the rear wheel.
Yeah, right off the bat it's a problem.
The right side hollow end bolt wasn't tightened—it came loose by hand.
Dirt that had been packed into the hub axle's threads was pushed out.
But the real problem isn't that it came loose by hand.
This is something I see a lot of people using without realizing it, but the part of the end piece that the frame rests against has snapped from twisting.
Setting that aside for now, the inside of the freebody was running dry—the grease had deteriorated and the rotation felt rough and crunchy.
I cleaned the parts.
The seal attached to the freebody on the hub has a yellow plastic material—looks like Delrin—at the point of constant contact. It's interesting that it's not just covering the end face of the freebody but also extends slightly inward because
if the end bolt remains loose and you keep riding, the freebody runs eccentric and rubs continuously against the hub body. That's within the manufacturer's design expectations. The polished-looking silver areas in the image above aren't from cutting—they're where the freebody has rubbed away the black anodize coating.
With a hard anodized hub body, the anodize layer is thick, so the area directly under the freebody turns entirely silver. That's why this wear is hard to spot visually.
↑This right end piece has the frame-contact section twisted off.
In the image above, on the left side of the circle on the end face, you can see the marks where it snapped.
This is actually a pretty common failure—so common, in fact, that I think it's less about user error and more about it just being a fragile part. If our shop carried Mavic products, I'd absolutely stock this part as a standard item.
↑From the small parts catalog. The end piece extends to include the frame contact point.
Breakage cases are common, and cases of people riding them without noticing are equally common. The previous owner of this wheel continued using it after the damage and flattened the threads, but I suspect there was no "malicious intent" in the legal sense when they quietly sold it at auction without mentioning it.
The nominal diameter for this section is 10mm. My measurement of the left end was 9.8mm, but on the right end where the threads had been flattened, the hub axle section measured 8.5mm. A difference of over 1mm in diameter between left and right at the base—that's not a trivial problem at all. Imagine tightening a quick release on the front wheel with only one side of the hub end floating 1.3mm out from under the quick release arm. This would cause the rim to tilt left in the upper half of the wheel for reasons separate from hub centering drift, and since the axle is tilted, it could even affect shifting on 10-speed or higher systems.
The nipples showed signs of previous truing work, but
the wheel center was off.
Even with the broken right end, it doesn't affect the reference surface for the centering gauge, so I did both truing and centering.
I was going to give the customer the part number for the right end so they could replace it themselves, but the original setup was such a mess—the right end was loose yet the left side ball adjustment was over-tightened. When I properly tightened the right end, the hub became so stiff it wouldn't turn. So if the customer replaces the right end, the wheel center shouldn't shift, but the ball adjustment might change very slightly.
There was a risk that the wheel would be used either staying slightly stiff or with slight play, so
the customer found the right end and sent it to me.
As it turned out, the wheel center didn't change and the ball adjustment didn't change either.
Now for the front wheel.
The hub bearing rotation was grinding, and when I opened it up, I found it had been switched to bearings with contactless seals.
↑This is a Cosmic hub I serviced separately, and the bearings are still original. The seal has pleats between the inner race and the seal—it's a contact-type seal in constant contact.
I've written about this many times before, but most Mavic hubs have no seal function on the end parts, so the bearing seal is the final line of defense. Much of the dirt in the image above is what was packed into the hub axle's threads and pushed out when we loosened them. But
even with the ball adjustment nut removed, you can see there's nothing like an O-ring or labyrinth seal structure—no dust cap function whatsoever.
Back to this wheel—the seal has no pleats and there's a gap between the inner race and seal. Both bearings were grinding, and the removed bearing alone showed such cloudy rotation you could see it needed replacing. Man, I was relieved.
Because here's the thing: with a bearing still pressed in, if the rotation is only slightly cloudy, once you pop it out as a loose bearing, the cloudiness often disappears, and it looks like "was this replacement really necessary?" That's why I'm glad this one was clearly bad even loose.
I replaced it with a contact-type bearing.
The seal groove looks shiny because I applied grease to the hub axle, inserted it, and then pulled it back out.
The front wheel also had centering drift. Plus, though I'm not sure if it was original, there was noticeable radial runout.
I said "not sure," but there were no signs of tampering on the nipples near where the runout was, so it was probably original.
I did both radial and lateral truing and centering.
The wheel in the image above is spinning. Before truing, the wear marks in the brake zone were perfectly circular, but after truing they're moving up and down within the brake zone. This is visible until fresh wear marks form. I was going to photograph the radial runout using the distance of the prominent wear marks from the rim edge, but I found an even clearer method.
This Ksyrium rim has a design where the brake zone cutting extends inward at the phase directly below the rim spoke holes. The innermost part of that clean zone doesn't get touched by brake pads, staying pristine. The width of that clean section is noticeably different between the phases with and without radial runout—you can really see it. Man, that was bad (past tense).
They'd acquired it at an auction and wanted me to inspect it before use.
Let me start with the rear wheel.
Yeah, right off the bat it's a problem.
The right side hollow end bolt wasn't tightened—it came loose by hand.
Dirt that had been packed into the hub axle's threads was pushed out.
But the real problem isn't that it came loose by hand.
This is something I see a lot of people using without realizing it, but the part of the end piece that the frame rests against has snapped from twisting.
Setting that aside for now, the inside of the freebody was running dry—the grease had deteriorated and the rotation felt rough and crunchy.
I cleaned the parts.
The seal attached to the freebody on the hub has a yellow plastic material—looks like Delrin—at the point of constant contact. It's interesting that it's not just covering the end face of the freebody but also extends slightly inward because
if the end bolt remains loose and you keep riding, the freebody runs eccentric and rubs continuously against the hub body. That's within the manufacturer's design expectations. The polished-looking silver areas in the image above aren't from cutting—they're where the freebody has rubbed away the black anodize coating.
With a hard anodized hub body, the anodize layer is thick, so the area directly under the freebody turns entirely silver. That's why this wear is hard to spot visually.
↑This right end piece has the frame-contact section twisted off.
In the image above, on the left side of the circle on the end face, you can see the marks where it snapped.
This is actually a pretty common failure—so common, in fact, that I think it's less about user error and more about it just being a fragile part. If our shop carried Mavic products, I'd absolutely stock this part as a standard item.
↑From the small parts catalog. The end piece extends to include the frame contact point.
Breakage cases are common, and cases of people riding them without noticing are equally common. The previous owner of this wheel continued using it after the damage and flattened the threads, but I suspect there was no "malicious intent" in the legal sense when they quietly sold it at auction without mentioning it.
The nominal diameter for this section is 10mm. My measurement of the left end was 9.8mm, but on the right end where the threads had been flattened, the hub axle section measured 8.5mm. A difference of over 1mm in diameter between left and right at the base—that's not a trivial problem at all. Imagine tightening a quick release on the front wheel with only one side of the hub end floating 1.3mm out from under the quick release arm. This would cause the rim to tilt left in the upper half of the wheel for reasons separate from hub centering drift, and since the axle is tilted, it could even affect shifting on 10-speed or higher systems.
The nipples showed signs of previous truing work, but
the wheel center was off.
Even with the broken right end, it doesn't affect the reference surface for the centering gauge, so I did both truing and centering.
I was going to give the customer the part number for the right end so they could replace it themselves, but the original setup was such a mess—the right end was loose yet the left side ball adjustment was over-tightened. When I properly tightened the right end, the hub became so stiff it wouldn't turn. So if the customer replaces the right end, the wheel center shouldn't shift, but the ball adjustment might change very slightly.
There was a risk that the wheel would be used either staying slightly stiff or with slight play, so
the customer found the right end and sent it to me.
As it turned out, the wheel center didn't change and the ball adjustment didn't change either.
Now for the front wheel.
The hub bearing rotation was grinding, and when I opened it up, I found it had been switched to bearings with contactless seals.
↑This is a Cosmic hub I serviced separately, and the bearings are still original. The seal has pleats between the inner race and the seal—it's a contact-type seal in constant contact.
I've written about this many times before, but most Mavic hubs have no seal function on the end parts, so the bearing seal is the final line of defense. Much of the dirt in the image above is what was packed into the hub axle's threads and pushed out when we loosened them. But
even with the ball adjustment nut removed, you can see there's nothing like an O-ring or labyrinth seal structure—no dust cap function whatsoever.
Back to this wheel—the seal has no pleats and there's a gap between the inner race and seal. Both bearings were grinding, and the removed bearing alone showed such cloudy rotation you could see it needed replacing. Man, I was relieved.
Because here's the thing: with a bearing still pressed in, if the rotation is only slightly cloudy, once you pop it out as a loose bearing, the cloudiness often disappears, and it looks like "was this replacement really necessary?" That's why I'm glad this one was clearly bad even loose.
I replaced it with a contact-type bearing.
The seal groove looks shiny because I applied grease to the hub axle, inserted it, and then pulled it back out.
The front wheel also had centering drift. Plus, though I'm not sure if it was original, there was noticeable radial runout.
I said "not sure," but there were no signs of tampering on the nipples near where the runout was, so it was probably original.
I did both radial and lateral truing and centering.
The wheel in the image above is spinning. Before truing, the wear marks in the brake zone were perfectly circular, but after truing they're moving up and down within the brake zone. This is visible until fresh wear marks form. I was going to photograph the radial runout using the distance of the prominent wear marks from the rim edge, but I found an even clearer method.
This Ksyrium rim has a design where the brake zone cutting extends inward at the phase directly below the rim spoke holes. The innermost part of that clean zone doesn't get touched by brake pads, staying pristine. The width of that clean section is noticeably different between the phases with and without radial runout—you can really see it. Man, that was bad (past tense).