I received a front and rear wheel built with the brand-that-shall-not-be-named's rim from a customer.
Let's start with the rear wheel.
Yeah, discretion won't cut it here.
But I do appreciate the effort.
The rear wheel had the chain drop inside the lowest gear,
so they're asking to replace the spokes.
If you ask the wholesaler for that brand, wouldn't that be the way to go (playing dumb).
That said, with this wheel
built with a Tni (Tni hubs) Evolite hub, and
with a shop name sticker shamefully plastered
on the rear hub body,
it seems like this brand's rim was built at a shop.
To put the conclusion up front: the customer's wheels were so
sloppily built that I was tempted to ask if they'd built it themselves.
I won't write the prefectures where that shop and the customer are located, but it's not on Honshu.
And by strange coincidence, there's also a shop with the same name
in Osaka Prefecture—or more precisely, Osaka City—though it's not a chain.
It's like how you see "Beijing" as a common name for Chinese restaurants,
but the first part of the shop name—equivalent to our "Bicycle Workshop"—
was different in their case, something like "Chinese Restaurant" or "Chinese Tavern."
Another odd coincidence: their technical skills are equally terrible.
From the chain drop, the inpoke spokes on the freewheel side have scratches and deformation.
Since it's a 24-spoke wheel, there are 6 inpokes on the freewheel side, and
the clearly bent spoke near the center of the image above
was from the initial impact of the fallen chain.
↑This is the 2nd and 3rd impacts,
↑When the chain jammed, they must've pedaled hard,
because around the 4th impact there's also damage to the hub flange.
The 5th and 6th spokes show scuff marks
but don't appear to need replacement.
The freewheel side was reasonably tight,
but the non-freewheel side was wobbly for an equal-diameter, equal-spoke-count build.
The rim isn't an offset rim.
There's some radial runout from spoke deformation,
but there's no phase with a big jump in lateral deflection.
At the provisional center, the rim was shifted abnormally toward the non-freewheel side.
If you mounted this on a frame, you could visually see it was off
from the gap with the seat stays and such.
I'm using the dimension taken on the right side for the left side here,
and both show nearly the same phase.
I tried varying the phase when applying the dimension on the left side,
but the result was about the same.
There's no lateral deflection that would fill in this center offset—
no matter where you measure, it comes out roughly this way.
This is a pretty complicated situation. If I centered this using only
the loosening tendency on the non-freewheel side,
the non-freewheel side tension would fall below the minimum viable for a wheel.
The freewheel side still has some tightening room,
but tightening the freewheel side alone isn't enough
to fill in this center offset.
In other words, I need to mix both freewheel-side tightening
and non-freewheel-side loosening in the right proportions.
To investigate, I fully loosened the nipples on the 5th and 6th inpokes
that appeared to have no chain damage, and
carelessly dropped a nipple inside the rim at one location.
I retrieved the nipple through the valve hole once,
then used the same method as for Zonda and Racing 3
rims—calling it back with a magnet to the rim hole.
When I received it, there was a Stan's rim tape on it,
and dried sealant had built up in the rim holes' recesses.
I wanted to work without removing this,
so I used the complicated method to recover the nipple.
But the rim tape gets replaced with a new one regardless of maker
(we use Stan's too), and
I later realized the customer had also asked for Imajji sealant to be added
and the tire to be reinstalled,
so that was a wasted step.
↑The timeline skips ahead, but this is the rim tape I replaced during work.
Also, if there was any hub issue, they wanted bearings and such replaced too,
but there was nothing wrong.
↑This is from when I dropped the nipple in the rim,
The tubeless valve was just barely short enough to be tightened with the nut,
so they'd extended the length with an intermediate-type extension valve
that doesn't have threads cut into the body itself.
The problem with this is: if you get a flat on the road and
can't remove the intermediate valve, you also can't remove the tubeless valve.
I removed it to recover the nipple.
Since the customer wanted the rim tape replaced anyway,
it was going to be removed regardless.
The timeline jumps again—this is actually a front wheel image—
but the intermediate extension valve is tightened fairly securely,
so it won't come loose by hand. Please carry appropriate tools.
I fully loosened the nipples on the 5th and 6th inpokes.
I secured them with tape so they wouldn't drop inside the rim.
No deformation.
So I'll reuse these two.
My judgment standard in cases like this is:
"Would I replace them if they were mine?"
There are so many crappy shops that go "Let's collect payment for full replacement, just in case",
but officially say "For safety we're replacing everything..."
I think that's the main source of the distrust people have toward shops.
If the customer had requested full replacement, I would've done it in this case too.
If they're leaving the decision to me,
then in this situation the answer is "no replacement."
↑The 4 deformed spokes.
Top to bottom in the image: 1st through 4th impacts.
There was something like a seal tape wrapped around the spoke threads
that you could peel off like unwrapping a bandage,
but it's almost meaningless.
Once I loosened the nipple slightly with a tool,
afterward it turned with fingers, so as a thread-locking agent
it basically didn't work.
Maybe it prevents sand from jamming in the nipple's threads,
but that's about it.
I marked the 6 inpokes that I either replaced or initially loosened with tape.
First, I'll try to dial out the runout
by touching only the nipples of the taped spokes.
I've taped the rim sides at locations with noticeable radial runout,
and when I spin the wheel, there's a pulsing radial deflection
obvious enough to see by eye.
Right under the taped inpoke, the rim deflects outward radially
because "there's still not enough tension",
but there's even more radial runout at completely different phases.
I roughly dialed in the lateral runout by adjusting only the 6 taped inpokes.
The rim is offset more than the original provisional center
because there's insufficient tension.
In fact, at this point the inpoke and outpoke tensions on the freewheel side are clearly different.
From here, when I bring the inpoke tension up to match the outpoke
and dial in just the lateral runout,
I get roughly the same amount of offset as the original provisional center.
This corresponds to the state one second before dropping the chain
inside the sprockets
(though the lateral runout should be less than that).
If the original wheel had been decent,
from here it would just be fine adjustment,
but the center is severely shifted to the non-freewheel side
and there's noticeable radial runout, so
this isn't spoke replacement or deflection correction—
this is a full wheel build.
To put it my way, before I fixed it,
these front and rear wheels never existed as actual wheels even once.
They're just fake wheels made by cobbling parts together.
At this precision, they wouldn't even pass the JIS standard inspection
for single-speed (Mama-chari) wheels.
If wheel building is this low-resolution of a process,
I have to think component installation on the frame and
disc brake bleeding resolution is equally sloppy.
I increased freewheel-side tension : decreased non-freewheel-side tension
at a ratio of about 7:3 (definitely not 5:5, that would loosen the non-freewheel side too much),
dialed in the radial and lateral runout, and when I thought the center looked about right
I checked with the centering gauge in the image above.
Considering it's used tubeless,
if you intentionally shift the rim toward the non-freewheel side,
that amount might make some sense,
but the original state was clearly abnormal.
Centering from there with just freewheel-side tightening
would've been extremely difficult, so
I centered it by loosening the non-freewheel side.
The freewheel side is rock-solid tight while
the non-freewheel side is a bit squishy...
which is normal for an equal-diameter, equal-spoke-count wheel.
The original assembly precision was well below normal
and I can't believe they got paid for it
(yet they had the nerve to plaster a shop sticker on the hub—that's hilarious),
but I've corrected it anyway.
Just replacing 4 spokes on a properly built wheel would've taken
about 1/3 the time I spent on this rear wheel.
The tubeless tire is a Schwalbe,
but as a general trend with aged tubeless tires,
the tire lift from pressurization gets worse over time.
The rear wheel wouldn't seat with our shop compressor,
so I used a CO₂ cartridge instead.
The valve surface is lightly frosted from the cold.
It thawed right away.
Fixed.
↑The 4th impact area + flange damage
↑The 5th and 6th spokes I didn't replace
You can see the Tni marking on the hub body in this image,
and I angled the previous shot to hide the hub slightly
to avoid the shop name sticker being visible in the frame.
Next, the front wheel.
The customer probably figured this was just an "add-on" to the rear wheel,
but this one was also out of bounds.
There's a center offset.
Rims from Mavic and others sometimes have the rear wheel rim
subtly offset to the left (non-freewheel side),
which makes you think maybe there's some philosophy behind it?
But when you see a rim-brake front wheel that's clearly uncentered,
you realize: oh, they just can't build wheels with proper precision.
For radial runout, I found the one phase around the entire rim
where the gauge barely touches
and took the photo, but
at almost every other phase
the gauge clears by this much.
This is a pretty big radial runout,
but the rear wheel's radial runout was even worse.
I dialed in the radial and lateral runout and centered it.
There was dried sealant on the inside of the tire's outer edge, but
I found that rubbing hard removes it, so
I used a sponge at the sink with running water
and scraped off as much as I could.
The Imajji sealant is visible at the lower right of the image.
Doing my best to be discreet—discretion aside—
This Schwalbe tubeless tire has
identical labels at opposite sides of the tread,
and both sides have labels—dual-label design.
Though it's dual-label, the tread pattern has rotational directionality
and there's a specific rotation direction. As long as you don't get those wrong and
align the valve hole phase "near the center of one of the labels,"
either side should work,
but there's a small "TLE" marking
on only the left side when the tire rotation direction is correct.
In a sense, this is the real basis for directional orientation.
When the tire is mounted in the correct rotation direction
and you align the valve phase "near the center of one of the two labels"
with the valve at the 12 o'clock position when viewed from the left,
two states are possible:
"TLE marking at 3 o'clock in the correct orientation" or
"TLE marking at 9 o'clock in reverse orientation."
This time I went with the former, but as long as the front and rear wheels match,
either works fine.
If anything, I could also use the TLE marking directly under the valve hole
on the left side as the single definitive orientation basis.
Let's start with the rear wheel.
Yeah, discretion won't cut it here.
But I do appreciate the effort.
The rear wheel had the chain drop inside the lowest gear,
so they're asking to replace the spokes.
If you ask the wholesaler for that brand, wouldn't that be the way to go (playing dumb).
That said, with this wheel
built with a Tni (Tni hubs) Evolite hub, and
with a shop name sticker shamefully plastered
on the rear hub body,
it seems like this brand's rim was built at a shop.
To put the conclusion up front: the customer's wheels were so
sloppily built that I was tempted to ask if they'd built it themselves.
I won't write the prefectures where that shop and the customer are located, but it's not on Honshu.
And by strange coincidence, there's also a shop with the same name
in Osaka Prefecture—or more precisely, Osaka City—though it's not a chain.
It's like how you see "Beijing" as a common name for Chinese restaurants,
but the first part of the shop name—equivalent to our "Bicycle Workshop"—
was different in their case, something like "Chinese Restaurant" or "Chinese Tavern."
Another odd coincidence: their technical skills are equally terrible.
From the chain drop, the inpoke spokes on the freewheel side have scratches and deformation.
Since it's a 24-spoke wheel, there are 6 inpokes on the freewheel side, and
the clearly bent spoke near the center of the image above
was from the initial impact of the fallen chain.
↑This is the 2nd and 3rd impacts,
↑When the chain jammed, they must've pedaled hard,
because around the 4th impact there's also damage to the hub flange.
The 5th and 6th spokes show scuff marks
but don't appear to need replacement.
The freewheel side was reasonably tight,
but the non-freewheel side was wobbly for an equal-diameter, equal-spoke-count build.
The rim isn't an offset rim.
There's some radial runout from spoke deformation,
but there's no phase with a big jump in lateral deflection.
At the provisional center, the rim was shifted abnormally toward the non-freewheel side.
If you mounted this on a frame, you could visually see it was off
from the gap with the seat stays and such.
I'm using the dimension taken on the right side for the left side here,
and both show nearly the same phase.
I tried varying the phase when applying the dimension on the left side,
but the result was about the same.
There's no lateral deflection that would fill in this center offset—
no matter where you measure, it comes out roughly this way.
This is a pretty complicated situation. If I centered this using only
the loosening tendency on the non-freewheel side,
the non-freewheel side tension would fall below the minimum viable for a wheel.
The freewheel side still has some tightening room,
but tightening the freewheel side alone isn't enough
to fill in this center offset.
In other words, I need to mix both freewheel-side tightening
and non-freewheel-side loosening in the right proportions.
To investigate, I fully loosened the nipples on the 5th and 6th inpokes
that appeared to have no chain damage, and
carelessly dropped a nipple inside the rim at one location.
I retrieved the nipple through the valve hole once,
then used the same method as for Zonda and Racing 3
rims—calling it back with a magnet to the rim hole.
When I received it, there was a Stan's rim tape on it,
and dried sealant had built up in the rim holes' recesses.
I wanted to work without removing this,
so I used the complicated method to recover the nipple.
But the rim tape gets replaced with a new one regardless of maker
(we use Stan's too), and
I later realized the customer had also asked for Imajji sealant to be added
and the tire to be reinstalled,
so that was a wasted step.
↑The timeline skips ahead, but this is the rim tape I replaced during work.
Also, if there was any hub issue, they wanted bearings and such replaced too,
but there was nothing wrong.
↑This is from when I dropped the nipple in the rim,
The tubeless valve was just barely short enough to be tightened with the nut,
so they'd extended the length with an intermediate-type extension valve
that doesn't have threads cut into the body itself.
The problem with this is: if you get a flat on the road and
can't remove the intermediate valve, you also can't remove the tubeless valve.
I removed it to recover the nipple.
Since the customer wanted the rim tape replaced anyway,
it was going to be removed regardless.
The timeline jumps again—this is actually a front wheel image—
but the intermediate extension valve is tightened fairly securely,
so it won't come loose by hand. Please carry appropriate tools.
I fully loosened the nipples on the 5th and 6th inpokes.
I secured them with tape so they wouldn't drop inside the rim.
No deformation.
So I'll reuse these two.
My judgment standard in cases like this is:
"Would I replace them if they were mine?"
There are so many crappy shops that go "Let's collect payment for full replacement, just in case",
but officially say "For safety we're replacing everything..."
I think that's the main source of the distrust people have toward shops.
If the customer had requested full replacement, I would've done it in this case too.
If they're leaving the decision to me,
then in this situation the answer is "no replacement."
↑The 4 deformed spokes.
Top to bottom in the image: 1st through 4th impacts.
There was something like a seal tape wrapped around the spoke threads
that you could peel off like unwrapping a bandage,
but it's almost meaningless.
Once I loosened the nipple slightly with a tool,
afterward it turned with fingers, so as a thread-locking agent
it basically didn't work.
Maybe it prevents sand from jamming in the nipple's threads,
but that's about it.
I marked the 6 inpokes that I either replaced or initially loosened with tape.
First, I'll try to dial out the runout
by touching only the nipples of the taped spokes.
I've taped the rim sides at locations with noticeable radial runout,
and when I spin the wheel, there's a pulsing radial deflection
obvious enough to see by eye.
Right under the taped inpoke, the rim deflects outward radially
because "there's still not enough tension",
but there's even more radial runout at completely different phases.
I roughly dialed in the lateral runout by adjusting only the 6 taped inpokes.
The rim is offset more than the original provisional center
because there's insufficient tension.
In fact, at this point the inpoke and outpoke tensions on the freewheel side are clearly different.
From here, when I bring the inpoke tension up to match the outpoke
and dial in just the lateral runout,
I get roughly the same amount of offset as the original provisional center.
This corresponds to the state one second before dropping the chain
inside the sprockets
(though the lateral runout should be less than that).
If the original wheel had been decent,
from here it would just be fine adjustment,
but the center is severely shifted to the non-freewheel side
and there's noticeable radial runout, so
this isn't spoke replacement or deflection correction—
this is a full wheel build.
To put it my way, before I fixed it,
these front and rear wheels never existed as actual wheels even once.
They're just fake wheels made by cobbling parts together.
At this precision, they wouldn't even pass the JIS standard inspection
for single-speed (Mama-chari) wheels.
If wheel building is this low-resolution of a process,
I have to think component installation on the frame and
disc brake bleeding resolution is equally sloppy.
I increased freewheel-side tension : decreased non-freewheel-side tension
at a ratio of about 7:3 (definitely not 5:5, that would loosen the non-freewheel side too much),
dialed in the radial and lateral runout, and when I thought the center looked about right
I checked with the centering gauge in the image above.
Considering it's used tubeless,
if you intentionally shift the rim toward the non-freewheel side,
that amount might make some sense,
but the original state was clearly abnormal.
Centering from there with just freewheel-side tightening
would've been extremely difficult, so
I centered it by loosening the non-freewheel side.
The freewheel side is rock-solid tight while
the non-freewheel side is a bit squishy...
which is normal for an equal-diameter, equal-spoke-count wheel.
The original assembly precision was well below normal
and I can't believe they got paid for it
(yet they had the nerve to plaster a shop sticker on the hub—that's hilarious),
but I've corrected it anyway.
Just replacing 4 spokes on a properly built wheel would've taken
about 1/3 the time I spent on this rear wheel.
The tubeless tire is a Schwalbe,
but as a general trend with aged tubeless tires,
the tire lift from pressurization gets worse over time.
The rear wheel wouldn't seat with our shop compressor,
so I used a CO₂ cartridge instead.
The valve surface is lightly frosted from the cold.
It thawed right away.
Fixed.
↑The 4th impact area + flange damage
↑The 5th and 6th spokes I didn't replace
You can see the Tni marking on the hub body in this image,
and I angled the previous shot to hide the hub slightly
to avoid the shop name sticker being visible in the frame.
Next, the front wheel.
The customer probably figured this was just an "add-on" to the rear wheel,
but this one was also out of bounds.
There's a center offset.
Rims from Mavic and others sometimes have the rear wheel rim
subtly offset to the left (non-freewheel side),
which makes you think maybe there's some philosophy behind it?
But when you see a rim-brake front wheel that's clearly uncentered,
you realize: oh, they just can't build wheels with proper precision.
For radial runout, I found the one phase around the entire rim
where the gauge barely touches
and took the photo, but
at almost every other phase
the gauge clears by this much.
This is a pretty big radial runout,
but the rear wheel's radial runout was even worse.
I dialed in the radial and lateral runout and centered it.
There was dried sealant on the inside of the tire's outer edge, but
I found that rubbing hard removes it, so
I used a sponge at the sink with running water
and scraped off as much as I could.
The Imajji sealant is visible at the lower right of the image.
This Schwalbe tubeless tire has
identical labels at opposite sides of the tread,
and both sides have labels—dual-label design.
Though it's dual-label, the tread pattern has rotational directionality
and there's a specific rotation direction. As long as you don't get those wrong and
align the valve hole phase "near the center of one of the labels,"
either side should work,
but there's a small "TLE" marking
on only the left side when the tire rotation direction is correct.
In a sense, this is the real basis for directional orientation.
When the tire is mounted in the correct rotation direction
and you align the valve phase "near the center of one of the two labels"
with the valve at the 12 o'clock position when viewed from the left,
two states are possible:
"TLE marking at 3 o'clock in the correct orientation" or
"TLE marking at 9 o'clock in reverse orientation."
This time I went with the former, but as long as the front and rear wheels match,
either works fine.
If anything, I could also use the TLE marking directly under the valve hole
on the left side as the single definitive orientation basis.
Products Mentioned in This Article
* Amazon affiliate links — prices may vary
Original Japanese post: のむラボ日記 #6872