I've taken in a set of Mavic Cosmic SLR 40 front and rear wheels from a customer.
The customer said it was okay to mention, so I'll note that this was purchased from a shop called Fitte.
The spokes are twisted in various places, and ceramic bearings were installed at the time of purchase.
The bearing seal on the left side of the rear hub body was crushed—apparently from using the wrong size tool—and the customer made a partial attempt to straighten it themselves.
They're asking me to correct these issues during inspection.
The front hub bearing rotation showed no abnormalities.
Looking at the seal, it says KOYO, which is Koyo Seiko, a bearing manufacturer. The company has since been merged into JTEKT.
It's kind of like the relationship between Prince Motor Company and Nissan, if you follow.
Since the end shapes on both sides are identical, I've taken some images from a pulled-back position too.
Measuring the dimension on the nominally right side according to the MAVIC marking on the hub body
and applying it to the nominally left side, there was this much gap between the centering gauge and the hub end.
Rim brake front wheels don't develop center offset over time, so whether it's an overseas online purchase or not, any Japanese brick-and-mortar shop that sells wheels with this much offset without fixing it is, frankly, a crap shop.
You might think they only replaced the bearings and didn't actually inspect it, but there are faint grip marks on the nipples, and the flattened orientation of several spokes is subtly twisted, so it looks like they deliberately caused this.
You might say the nipple grip marks and spoke twist happened during wheel building at Mavic, but I have some evidence that's not the case, which I'll explain when I get to the rear wheel.
I mean, if spokes are subtly twisted in the out-of-box condition, you should straighten them before selling, right?
I centered it up.
There was barely any lateral runout, so I can do that "just-the-runout" pretend truing thing.
Someday I'd like to get to a point where I can center the wheel without twisting spokes!
Now for the rear wheel.
This is bad. The hub rotation is quite rough and gritty.
Based on the spline pattern of the freebody, the wheel appears to be unused or barely used.
Several nipples show faint anodize loss—in other words, grip marks—but
on the spokes corresponding to those nipples, and with a correlation that can't be ruled out,
there's characteristic paint chipping on the "edge" face of the flattened section near the start of the butted portion on the rim side.
↑similar damage in other areas
And the spokes showing this pattern appear in 7 consecutive locations on one side of the wheel, so it couldn't have happened during the manufacturer's wheel building.
Unusually for a Mavic rear wheel, the rim was shifted toward the right side (freewheel side).
↑by about this much
Based on the pattern of twisted spokes with grip marks and the nipple condition, it appears the rim was shifted to the left, and someone loosened the left side nipples to do truing work.
There was no lateral runout "alone."
I removed the rear hub's left cone end.
I removed the snap ring visible in the image and the wave washer below it
and pulled the hub shaft out to the right.
This wheel appears to be unused or nearly unused, but the Instant Drive 360 face ratchet is almost completely devoid of its dedicated grease.
Since it's clear they replaced the bearings, if they thought there wasn't enough grease, they should have applied fresh grease even to new bearings.
The bearing seal says ONI, so the ceramic bearing appears to be from ONI Bearings (鬼ベアリング).
↑This is the left-side hub body seal,
which was crushed even worse before the customer tried to straighten it.
The right-side hub body bearing showed no seal crush.
However, when I put my finger in both bearings and spun them, I found that the cause of the roughness was the right-side hub body bearing—the one whose seal wasn't crushed.
I carefully removed both bearings without deforming their seals.
The left seal's crushing was present when I received the wheels.
↑The bearing seal on the right in this image is the crushed one from the hub body's left side, which the customer had partially straightened.
Since both bearings were the same size, I transplanted the clean seal to the left-side bearing and will replace the right-side bearing with a standard steel ball bearing.
This orange seal is different from the front hub bearing—it's under JTEKT's name.
I installed the seal from the original right side onto the left-side bearing.
I removed the ratchet ring and its retaining C-ring
(obviously removing the latter first).
I removed the bearing.
Good. Even the removed bearing on its own shows rough, gritty rotation.
You see, when bearings are press-fitted, the rotation gets a bit duller, and while there's definitely some resistance during press-fitting, when it's very minor, sometimes removing just the bearing and spinning it with your finger makes you wonder "wait, this spins so smooth—did we really need to replace it?" because it seems so much better.
So if something's really wrong, it's actually reassuring when the bearing itself is gritty enough on its own that it provides convincing evidence of the problem. That's why I wrote "good" earlier.
I can't claim to be someone who does good deeds—I swing sticks around to mess with dragonfly mating and such—but I do take pride in the fact that I don't do business where I replace parts that don't need replacing and force unnecessary costs on customers.
But this work wasn't done in front of the customer, so it's actually convenient if the problem parts are obviously bad—it's clearer that way.
↑On the left in the image is a standard 6802 bearing.
With an inner diameter of 15mm, outer diameter of 28mm, and width of 7mm,
using the metric notation system for non-standard bearings, it would be a 152287 bearing.
The hub shaft inner diameter is obviously 12mm for a 12mm through-axle outer diameter,
but if that hub shaft's outer diameter is 15mm, you can use a 6802 bearing, though a 1.5mm-thick pipe—the radial difference—feels a bit insubstantial. It would be nice if there were a bearing like a 152287 but with inner diameter changed to just 17mm so the hub shaft could be 17mm. This demand from the high-end spoke bike industry created the 172287 bearing shown on the right in the image above.
↑It has the same width and outer diameter
↑but with an inner diameter 2mm larger
I press-fitted it into the hub.
Once I'd assembled it to wheel condition, the rotation lost all its roughness.
In other words, the crappy shop's ONI bearing
got way smoother rotation
than a standard 172287
bearing pressed in correctly and carefully
at around 1000 yen per unit.
I'm returning the "screw-up ONI bearing"—the right bearing probably damaged by overloading the inner race, plus the left bearing with the crushed seal—to the customer, but as a bonus for comparison testing, I've included another 172287 bearing that I press-fitted separately from the one in the hub.
The freebody bearing showed no abnormalities inside or out.
I applied grease to the face ratchet—something other than Mavic's white variety.
I've told the customer what it is, and I've provided them with more than they'd reasonably use for personal use.
Hmm, the twisted spokes really do tend to have
characteristic grip marks. This happens to be the spoke with the largest paint chip.
I've never seen this kind of damage on the same butted spokes of other Mavic wheels, so as I wrote earlier, it couldn't have happened during the manufacturer's wheel building.
I centered it up.
According to the customer, when they told the shop about the cogging in the rotation, they were told "the freebody bearing is OEM, and there's plenty of grease, so that's why it coggs." Based on the earlier exchanges, it was clear the shop was trying to cover up their mistake with excuses, so the customer sent the wheels to us.
Freebody bearing is OEM →
If it's new, that wouldn't cause cogging roughness
Plenty of grease →
The amount of grease on the face ratchet has no bearing on how the hub shaft rotates when held by hand. Besides, the face ratchet grease was depleted.
Also, if plenty of grease causes cogging, then hub greasing would be making rotation worse, wouldn't it?
I have one request at the end of this post.
Please only applaud this article if you're someone who
"had the shop mess up,
but got talked into accepting their excuse that there was no shop fault,
all while internally praying the customer was dumb enough to believe it,
sweating profusely with darting eyes,
and despite giving explanations in an oddly authoritative tone,
received no redo work or refund."
Well, surely there aren't that many…
↓There are that many?!
The customer said it was okay to mention, so I'll note that this was purchased from a shop called Fitte.
The spokes are twisted in various places, and ceramic bearings were installed at the time of purchase.
The bearing seal on the left side of the rear hub body was crushed—apparently from using the wrong size tool—and the customer made a partial attempt to straighten it themselves.
They're asking me to correct these issues during inspection.
The front hub bearing rotation showed no abnormalities.
Looking at the seal, it says KOYO, which is Koyo Seiko, a bearing manufacturer. The company has since been merged into JTEKT.
It's kind of like the relationship between Prince Motor Company and Nissan, if you follow.
Since the end shapes on both sides are identical, I've taken some images from a pulled-back position too.
Measuring the dimension on the nominally right side according to the MAVIC marking on the hub body
and applying it to the nominally left side, there was this much gap between the centering gauge and the hub end.
Rim brake front wheels don't develop center offset over time, so whether it's an overseas online purchase or not, any Japanese brick-and-mortar shop that sells wheels with this much offset without fixing it is, frankly, a crap shop.
You might think they only replaced the bearings and didn't actually inspect it, but there are faint grip marks on the nipples, and the flattened orientation of several spokes is subtly twisted, so it looks like they deliberately caused this.
You might say the nipple grip marks and spoke twist happened during wheel building at Mavic, but I have some evidence that's not the case, which I'll explain when I get to the rear wheel.
I mean, if spokes are subtly twisted in the out-of-box condition, you should straighten them before selling, right?
I centered it up.
There was barely any lateral runout, so I can do that "just-the-runout" pretend truing thing.
Someday I'd like to get to a point where I can center the wheel without twisting spokes!
Now for the rear wheel.
This is bad. The hub rotation is quite rough and gritty.
Based on the spline pattern of the freebody, the wheel appears to be unused or barely used.
Several nipples show faint anodize loss—in other words, grip marks—but
on the spokes corresponding to those nipples, and with a correlation that can't be ruled out,
there's characteristic paint chipping on the "edge" face of the flattened section near the start of the butted portion on the rim side.
↑similar damage in other areas
And the spokes showing this pattern appear in 7 consecutive locations on one side of the wheel, so it couldn't have happened during the manufacturer's wheel building.
Unusually for a Mavic rear wheel, the rim was shifted toward the right side (freewheel side).
↑by about this much
Based on the pattern of twisted spokes with grip marks and the nipple condition, it appears the rim was shifted to the left, and someone loosened the left side nipples to do truing work.
There was no lateral runout "alone."
I removed the rear hub's left cone end.
I removed the snap ring visible in the image and the wave washer below it
and pulled the hub shaft out to the right.
This wheel appears to be unused or nearly unused, but the Instant Drive 360 face ratchet is almost completely devoid of its dedicated grease.
Since it's clear they replaced the bearings, if they thought there wasn't enough grease, they should have applied fresh grease even to new bearings.
The bearing seal says ONI, so the ceramic bearing appears to be from ONI Bearings (鬼ベアリング).
↑This is the left-side hub body seal,
which was crushed even worse before the customer tried to straighten it.
The right-side hub body bearing showed no seal crush.
However, when I put my finger in both bearings and spun them, I found that the cause of the roughness was the right-side hub body bearing—the one whose seal wasn't crushed.
I carefully removed both bearings without deforming their seals.
The left seal's crushing was present when I received the wheels.
↑The bearing seal on the right in this image is the crushed one from the hub body's left side, which the customer had partially straightened.
Since both bearings were the same size, I transplanted the clean seal to the left-side bearing and will replace the right-side bearing with a standard steel ball bearing.
This orange seal is different from the front hub bearing—it's under JTEKT's name.
I installed the seal from the original right side onto the left-side bearing.
I removed the ratchet ring and its retaining C-ring
(obviously removing the latter first).
I removed the bearing.
Good. Even the removed bearing on its own shows rough, gritty rotation.
You see, when bearings are press-fitted, the rotation gets a bit duller, and while there's definitely some resistance during press-fitting, when it's very minor, sometimes removing just the bearing and spinning it with your finger makes you wonder "wait, this spins so smooth—did we really need to replace it?" because it seems so much better.
So if something's really wrong, it's actually reassuring when the bearing itself is gritty enough on its own that it provides convincing evidence of the problem. That's why I wrote "good" earlier.
I can't claim to be someone who does good deeds—I swing sticks around to mess with dragonfly mating and such—but I do take pride in the fact that I don't do business where I replace parts that don't need replacing and force unnecessary costs on customers.
But this work wasn't done in front of the customer, so it's actually convenient if the problem parts are obviously bad—it's clearer that way.
↑On the left in the image is a standard 6802 bearing.
With an inner diameter of 15mm, outer diameter of 28mm, and width of 7mm,
using the metric notation system for non-standard bearings, it would be a 152287 bearing.
The hub shaft inner diameter is obviously 12mm for a 12mm through-axle outer diameter,
but if that hub shaft's outer diameter is 15mm, you can use a 6802 bearing, though a 1.5mm-thick pipe—the radial difference—feels a bit insubstantial. It would be nice if there were a bearing like a 152287 but with inner diameter changed to just 17mm so the hub shaft could be 17mm. This demand from the high-end spoke bike industry created the 172287 bearing shown on the right in the image above.
↑It has the same width and outer diameter
↑but with an inner diameter 2mm larger
I press-fitted it into the hub.
Once I'd assembled it to wheel condition, the rotation lost all its roughness.
In other words, the crappy shop's ONI bearing
got way smoother rotation
than a standard 172287
bearing pressed in correctly and carefully
at around 1000 yen per unit.
I'm returning the "screw-up ONI bearing"—the right bearing probably damaged by overloading the inner race, plus the left bearing with the crushed seal—to the customer, but as a bonus for comparison testing, I've included another 172287 bearing that I press-fitted separately from the one in the hub.
The freebody bearing showed no abnormalities inside or out.
I applied grease to the face ratchet—something other than Mavic's white variety.
I've told the customer what it is, and I've provided them with more than they'd reasonably use for personal use.
Hmm, the twisted spokes really do tend to have
characteristic grip marks. This happens to be the spoke with the largest paint chip.
I've never seen this kind of damage on the same butted spokes of other Mavic wheels, so as I wrote earlier, it couldn't have happened during the manufacturer's wheel building.
I centered it up.
According to the customer, when they told the shop about the cogging in the rotation, they were told "the freebody bearing is OEM, and there's plenty of grease, so that's why it coggs." Based on the earlier exchanges, it was clear the shop was trying to cover up their mistake with excuses, so the customer sent the wheels to us.
Freebody bearing is OEM →
If it's new, that wouldn't cause cogging roughness
Plenty of grease →
The amount of grease on the face ratchet has no bearing on how the hub shaft rotates when held by hand. Besides, the face ratchet grease was depleted.
Also, if plenty of grease causes cogging, then hub greasing would be making rotation worse, wouldn't it?
I have one request at the end of this post.
Please only applaud this article if you're someone who
"had the shop mess up,
but got talked into accepting their excuse that there was no shop fault,
all while internally praying the customer was dumb enough to believe it,
sweating profusely with darting eyes,
and despite giving explanations in an oddly authoritative tone,
received no redo work or refund."
Well, surely there aren't that many…
↓There are that many?!