Another day of wheels (and so on).
A customer brought me the rear wheel of a wheel called ACR22 made by Velarete, which is some kind of frame builder.
My colleague was shirking their duties, but
the customer said it was fine to not keep the name private, so...
Some branding going on here, but
the rim is a Tni AL22 or Kinlin XR200,
and the hub is Novatech 482SL, which is the precursor to Tni's Evolite hub.
Originally it was built 24H all-black four-cross lacing from the start,
but apparently the anti-freewheel side spokes broke multiple times
in a very short period after purchase,
so they tried swapping the anti-freewheel side spokes to Pillar aero ones,
and that's the current state.
Even with that, the wheel quickly developed runout, so
they asked Velarete about it, but
both the spoke failures and the runout got a "beats me" response, so
through a friend's connection, it ended up here.
Now, about these frequent spoke failures—
if they say they don't know, let me teach them (condescending tone).
When something happens at an abnormally frequent rate,
it means there's something abnormal causing it.
And abandoning the chance to solve that means
they should accept criticism backed by reasoning—you idiot.
First, the freewheel side uses Sapim black race spokes in a four-cross lacing, but
The freewheel side isn't tensioned properly at all.
This is the root cause of the spoke failures.
Spoke failures can't be completely avoided, but
with low spoke tension, the play between the hub flange hole and the spoke head increases,
which raises the risk of spoke head failures.
Also, the lower the tension,
the greater the variation in spoke tension on the same flange side.
Spoke tension—the tighter the better, up to a point, but this is way too low.
The failures actually occurred all on the anti-freewheel side, but
with the freewheel side at this tension level, if it was built same-gauge same-cross on both sides,
the anti-freewheel side spoke tension must have been even more scattered,
especially at the low end.
People who build nothing but velodrome wheels tend to
build floppy wheels on other rims too, the way they do with Araya 16B Gold
(the only currently-produced velodrome rim) when building those,
but whether this low tension reflects some philosophy of Velarete's, I don't know.
And I don't know if they build velodrome wheels often or not, either.
As for the anti-freewheel side,
Pillar aero spokes with the slit hole feature are used, but
The spoke deformation is way larger than what the fore-aft width of the butted section would suggest.
Many of you have probably already figured it out by this point, but
this spoke combination—it's reverse gauge different-cross lacing.
How stupid do you have to be to pull something like this off?
Well, if the actual wheelbuilding were decent despite the stupidity,
it would've been better, but...
By the way, when the customer brought this rear wheel in,
I had them feel up the newly-built rear wheel from my nomlabo wheel #5 that was at the shop—
24H semi-competition four-cross lacing without any twist ties—
and had them squeeze the spokes.
Hmm, weird—the anti-freewheel side spokes on this one are
way thinner than the rear wheel you brought, yet
they deform way less.
I spent about 30 minutes explaining to the customer why that happens,
using two whiteboard pages
(I filled one side, erased it, and filled it again)
covering the physics of spoke tension and lacing geometry.
Same rim, same spoke count, similar hub
(the 660 hub is actually less favorable than the 482SL in terms of left-right spoke deformation difference), yet
the rear wheel I built has
(even excluding the twist ties)
less spoke deformation and less left-right asymmetry—and I explained that this isn't magic,
but the concrete mechanical reasons why.
Earlier I wrote "if the actual wheelbuilding were decent"; here's what I meant:
I found the phase with maximum lateral runout on the anti-freewheel side.
It's the only phase in the wheel's rotation where the gauge contacts the rim.
I marked it with tape.
With the gauge in the same position, I found the phase with maximum runout toward the freewheel side,
where the rim and gauge are farthest apart.
↑From the front-rear angle, it looks like this.
This is the sum of the maximum amounts of lateral runout on both sides.
I marked this position with tape too.
The tape positions relate to each other this way,
but I deliberately placed the center gauge away from that phase.
I avoided the situation where you'd deliberately search out the spot with the largest center deviation
or end up with the same result by accident,
but the rim was still pretty far off-center toward the freewheel side.
When I placed the center gauge at two other locations directly below the tape marks,
the rim was still off-center toward the freewheel side.
Low-tension rear wheels do move toward the freewheel side faster with age than high-tension ones,
but it's hard to imagine this wheel, which is already so wildly out of true it's almost touching the brake pad,
has been ridden many kilometers to get this way.
And a properly-centered wheel that goes wildly off this quickly doesn't happen in the short term, so
this center offset must have been present from the moment the wheel was built.
Whether the person who built this wheel was the builder themselves or an apprentice, I don't know
(I don't even know if they have an apprentice), but their "eye is pretty bad."
By "bad eye" I don't just mean poor vision—
I mean their tolerance threshold for what counts as "straight" is pretty low.
This wheel, as it appears to the builder's eye,
was built as straight as possible,
and since they're a pro, they should be able to charge money without embarrassment for it.
But if their work ends up like this, either their eye is genuinely bad
or they cut corners somewhere.
In my view, I haven't seen the Velarete frames themselves, but
I suspect their frames are made with the same level of "bad eye"—
the same low standards for precision.
If that's not the case, then they're only cutting corners on wheels.
Maybe their frames are fine and they're just bad at wheelbuilding!
Someone might say that. But if their eye were good, they'd realize
that a wheel they built doesn't hold a candle to other wheels in the world
in terms of spoke tension and build precision, even if they charged money for it,
and yet here they are slapping stickers on it and branding it for sale.
That tells me this isn't laziness but a genuine case of "bad eye."
And as an aside, making a reverse-gauge different-cross lacing when rebuilding?
That's "stupid"—not speculation, just plain fact.
You might say, "You can't make frames, so don't talk down to people!"
But even people who can't build wheels can tell if a wheel is out of true,
and just the same, I can judge a frame's alignment even if I don't match
the standards of a proper builder. (There are individual differences in criteria.)
And while I can't make frames, I'm fortunate to know a reliable frame builder
and have them make frames for me.
Now, Velarete's frames—I've never examined one in detail, so
I genuinely don't know if they're good or bad (the output of their eye quality),
but judging from this level of wheelwork, they don't seem like much.
That's just my opinion.
For whatever reason, the nipples were 14mm semi-long brass nipples.
Up to this point, I haven't done any actual wheelbuilding work.
I've only put the wheel on the stand, marked it with tape,
and placed the center gauge—I haven't turned a single nipple.
The first task was to peel back the rim tape, but
the spoke length on the freewheel side was unacceptably short,
and on the anti-freewheel side it was long, though still within tolerance.
↑Anti-freewheel side
↑Freewheel side
↑Anti-freewheel side
↑Freewheel side
Wheelsmith Spoke Prep has been applied.
We use it at the shop too, but I was shocked
the wheel had such bad runout despite having Spoke Prep applied.
This Spoke Prep comes in two colors—
orange (skin tone) and blue (light blue)—with no functional difference,
so you can switch colors between left and right sides to avoid mixing them up, which is convenient.
But orange and blue do have
a clear performance difference (←trade secret)
so I only use blue.
Rim-brake front wheels and track-fixed rear wheels don't have offset,
and track-fixed single-threaded rear wheels have about 0.5mm offset in theory,
so I often build them with same-length spokes on both sides,
and wheels with more offset than that—free-hub rear wheels and
disc-brake front wheels—I usually do reverse-gauge different-cross lacing,
so spoke mix-ups are impossible,
which means I don't need to use different Spoke Prep colors for left and right.
After the rebuild, the left and right spokes are clearly different,
and mix-ups are impossible, so
if they'd noticed the performance difference in Spoke Prep,
I'd expect them to use blue on both sides.
But they're using orange on one side, so they haven't noticed the difference.
I don't know if they did a full disassembly or just the anti-freewheel side,
but the Pillar aero spoke on the anti-freewheel side is the one they swapped in later,
and it has orange applied.
Also, if you've noticed the performance difference, it would be best
to use blue on the low-tension side, which is the anti-freewheel side.
I know another wheelbuilder who changes Spoke Prep colors between left and right
on wheels with offset, and even they always use blue on both sides
for rim-brake front wheels and such, which is why they say blue wears out faster.
I disassembled the wheel and collected the spokes.
↑Pillar aero
↑Sapim race
Comparing lengths.
When building a freewheel hub rear wheel with a non-offset rim and same lacing on both sides,
the left-right length difference is usually about 2mm.
The 482SL hub is a high-low flange, so this length difference comes out to about 2.5mm in theory
regardless of rim inner diameter, but
if you don't have a spoke cutter and can only choose lengths in 1mm increments—
2mm or 3mm difference—then 2mm is fine.
There was a 6mm difference in length between left and right.
Both spokes have a 2mm outer diameter on the round part before the threads start, so
using that as a reference, even visually you can see the 6mm difference.
With a difference this large, if one spoke length was correct,
the other would definitely be too long or too short.
In reality, both are wrong. As I wrote earlier,
the freewheel side race was too short,
and the anti-freewheel side Pillar aero was slightly long.
The Pillar ones could be fixed with a little cutting, but
they probably don't own a spoke cutter.
If they don't, they should have sourced the right spoke lengths from the start,
but that's laziness too.
Or maybe their "eye is so bad" they think this is acceptable.
Man, this guy probably does the same thing with frames—
using mismatched components that are the wrong length. ←I'm being repetitive
Built it.
Before I saw the spoke lengths on the original wheel, I was planning
to reuse the race spokes on the freewheel side and swap to CX-RAY on the anti-freewheel side
for something semi-competition level, but
I went with cut Pillar aero on the freewheel side,
and CX-RAY on the anti-freewheel side.
I'll do the twist-tie lacing later, but
it might be good to show the customer the wheel first without twist ties,
before deciding.
Since we can't source Pillar aero spokes at the shop,
if the freewheel side spokes break again and if there's a situation
where they can't go back to Velarete (wink wink),
I'll swap everything to competition spokes.
For spoke weight ratio, I'm using 85% as a rough estimate for 2.0-1.8-2.0mm round butted spokes,
but DT Competition has longer 2.0mm sections on both ends, so
actual measurement gives about 86.5%.
Sapim race comes out to about 85%, but
before I weighed them, I calculated at 85% and got 74.4g.
Then when I actually weighed the spokes, they came to 74.0g, so
the actual spoke weight ratio for this batch of race is 84.48%.
Sapim CX, which like Pillar aero uses the slit-hole design,
shows a theoretical weight ratio of 98.9% from the spec weight,
but actual measurement gives 100.3%.
This batch of Pillar spokes has a thinner flat section than CX,
so I estimated around 95% spoke weight ratio, but
actual measurement came to 98.28%.
That's close enough to 100% like a plain 14g spoke that there's no problem.
Except that this reverse-gauge lacing is still problematic.
It's the equivalent of building a round-spoke rear wheel with competition on the freewheel side
and champion on the anti-freewheel side on purpose.
My rebuilt wheel, photographed from the right side
clockwise from the valve hole:
↑Anti-freewheel side
↑Freewheel side
↑Anti-freewheel side
↑Freewheel side
The freewheel side Pillar spokes are cut,
so the end is silver. The anti-freewheel side CX-RAY
didn't need cutting, so the end is black.
To Velarete:
In case you come across this post someday,
I'm writing this now.
Feel free to talk back to me, but
explain the rational reason why the rebuilt spokes are in reverse-gauge different-cross lacing
and why the spoke lengths differ by 6mm first.
If you can't do that, stay quiet forever.
Just do business with people who won't complain about your work,
and get by however you can.
A customer brought me the rear wheel of a wheel called ACR22 made by Velarete, which is some kind of frame builder.
My colleague was shirking their duties, but
the customer said it was fine to not keep the name private, so...
Some branding going on here, but
the rim is a Tni AL22 or Kinlin XR200,
and the hub is Novatech 482SL, which is the precursor to Tni's Evolite hub.
Originally it was built 24H all-black four-cross lacing from the start,
but apparently the anti-freewheel side spokes broke multiple times
in a very short period after purchase,
so they tried swapping the anti-freewheel side spokes to Pillar aero ones,
and that's the current state.
Even with that, the wheel quickly developed runout, so
they asked Velarete about it, but
both the spoke failures and the runout got a "beats me" response, so
through a friend's connection, it ended up here.
Now, about these frequent spoke failures—
if they say they don't know, let me teach them (condescending tone).
When something happens at an abnormally frequent rate,
it means there's something abnormal causing it.
And abandoning the chance to solve that means
they should accept criticism backed by reasoning—you idiot.
First, the freewheel side uses Sapim black race spokes in a four-cross lacing, but
The freewheel side isn't tensioned properly at all.
This is the root cause of the spoke failures.
Spoke failures can't be completely avoided, but
with low spoke tension, the play between the hub flange hole and the spoke head increases,
which raises the risk of spoke head failures.
Also, the lower the tension,
the greater the variation in spoke tension on the same flange side.
Spoke tension—the tighter the better, up to a point, but this is way too low.
The failures actually occurred all on the anti-freewheel side, but
with the freewheel side at this tension level, if it was built same-gauge same-cross on both sides,
the anti-freewheel side spoke tension must have been even more scattered,
especially at the low end.
People who build nothing but velodrome wheels tend to
build floppy wheels on other rims too, the way they do with Araya 16B Gold
(the only currently-produced velodrome rim) when building those,
but whether this low tension reflects some philosophy of Velarete's, I don't know.
And I don't know if they build velodrome wheels often or not, either.
As for the anti-freewheel side,
Pillar aero spokes with the slit hole feature are used, but
The spoke deformation is way larger than what the fore-aft width of the butted section would suggest.
Many of you have probably already figured it out by this point, but
this spoke combination—it's reverse gauge different-cross lacing.
How stupid do you have to be to pull something like this off?
Well, if the actual wheelbuilding were decent despite the stupidity,
it would've been better, but...
By the way, when the customer brought this rear wheel in,
I had them feel up the newly-built rear wheel from my nomlabo wheel #5 that was at the shop—
24H semi-competition four-cross lacing without any twist ties—
and had them squeeze the spokes.
Hmm, weird—the anti-freewheel side spokes on this one are
way thinner than the rear wheel you brought, yet
they deform way less.
I spent about 30 minutes explaining to the customer why that happens,
using two whiteboard pages
(I filled one side, erased it, and filled it again)
covering the physics of spoke tension and lacing geometry.
Same rim, same spoke count, similar hub
(the 660 hub is actually less favorable than the 482SL in terms of left-right spoke deformation difference), yet
the rear wheel I built has
(even excluding the twist ties)
less spoke deformation and less left-right asymmetry—and I explained that this isn't magic,
but the concrete mechanical reasons why.
Earlier I wrote "if the actual wheelbuilding were decent"; here's what I meant:
I found the phase with maximum lateral runout on the anti-freewheel side.
It's the only phase in the wheel's rotation where the gauge contacts the rim.
I marked it with tape.
With the gauge in the same position, I found the phase with maximum runout toward the freewheel side,
where the rim and gauge are farthest apart.
↑From the front-rear angle, it looks like this.
This is the sum of the maximum amounts of lateral runout on both sides.
I marked this position with tape too.
The tape positions relate to each other this way,
but I deliberately placed the center gauge away from that phase.
I avoided the situation where you'd deliberately search out the spot with the largest center deviation
or end up with the same result by accident,
but the rim was still pretty far off-center toward the freewheel side.
When I placed the center gauge at two other locations directly below the tape marks,
the rim was still off-center toward the freewheel side.
Low-tension rear wheels do move toward the freewheel side faster with age than high-tension ones,
but it's hard to imagine this wheel, which is already so wildly out of true it's almost touching the brake pad,
has been ridden many kilometers to get this way.
And a properly-centered wheel that goes wildly off this quickly doesn't happen in the short term, so
this center offset must have been present from the moment the wheel was built.
Whether the person who built this wheel was the builder themselves or an apprentice, I don't know
(I don't even know if they have an apprentice), but their "eye is pretty bad."
By "bad eye" I don't just mean poor vision—
I mean their tolerance threshold for what counts as "straight" is pretty low.
This wheel, as it appears to the builder's eye,
was built as straight as possible,
and since they're a pro, they should be able to charge money without embarrassment for it.
But if their work ends up like this, either their eye is genuinely bad
or they cut corners somewhere.
In my view, I haven't seen the Velarete frames themselves, but
I suspect their frames are made with the same level of "bad eye"—
the same low standards for precision.
If that's not the case, then they're only cutting corners on wheels.
Maybe their frames are fine and they're just bad at wheelbuilding!
Someone might say that. But if their eye were good, they'd realize
that a wheel they built doesn't hold a candle to other wheels in the world
in terms of spoke tension and build precision, even if they charged money for it,
and yet here they are slapping stickers on it and branding it for sale.
That tells me this isn't laziness but a genuine case of "bad eye."
And as an aside, making a reverse-gauge different-cross lacing when rebuilding?
That's "stupid"—not speculation, just plain fact.
You might say, "You can't make frames, so don't talk down to people!"
But even people who can't build wheels can tell if a wheel is out of true,
and just the same, I can judge a frame's alignment even if I don't match
the standards of a proper builder. (There are individual differences in criteria.)
and have them make frames for me.
Now, Velarete's frames—I've never examined one in detail, so
I genuinely don't know if they're good or bad (the output of their eye quality),
but judging from this level of wheelwork, they don't seem like much.
That's just my opinion.
For whatever reason, the nipples were 14mm semi-long brass nipples.
Up to this point, I haven't done any actual wheelbuilding work.
I've only put the wheel on the stand, marked it with tape,
and placed the center gauge—I haven't turned a single nipple.
The first task was to peel back the rim tape, but
the spoke length on the freewheel side was unacceptably short,
and on the anti-freewheel side it was long, though still within tolerance.
↑Anti-freewheel side
↑Freewheel side
↑Anti-freewheel side
↑Freewheel side
Wheelsmith Spoke Prep has been applied.
We use it at the shop too, but I was shocked
the wheel had such bad runout despite having Spoke Prep applied.
This Spoke Prep comes in two colors—
orange (skin tone) and blue (light blue)—with no functional difference,
so you can switch colors between left and right sides to avoid mixing them up, which is convenient.
But orange and blue do have
a clear performance difference (←trade secret)
so I only use blue.
Rim-brake front wheels and track-fixed rear wheels don't have offset,
and track-fixed single-threaded rear wheels have about 0.5mm offset in theory,
so I often build them with same-length spokes on both sides,
and wheels with more offset than that—free-hub rear wheels and
disc-brake front wheels—I usually do reverse-gauge different-cross lacing,
so spoke mix-ups are impossible,
which means I don't need to use different Spoke Prep colors for left and right.
After the rebuild, the left and right spokes are clearly different,
and mix-ups are impossible, so
if they'd noticed the performance difference in Spoke Prep,
I'd expect them to use blue on both sides.
But they're using orange on one side, so they haven't noticed the difference.
I don't know if they did a full disassembly or just the anti-freewheel side,
but the Pillar aero spoke on the anti-freewheel side is the one they swapped in later,
and it has orange applied.
Also, if you've noticed the performance difference, it would be best
to use blue on the low-tension side, which is the anti-freewheel side.
I know another wheelbuilder who changes Spoke Prep colors between left and right
on wheels with offset, and even they always use blue on both sides
for rim-brake front wheels and such, which is why they say blue wears out faster.
I disassembled the wheel and collected the spokes.
↑Pillar aero
↑Sapim race
Comparing lengths.
When building a freewheel hub rear wheel with a non-offset rim and same lacing on both sides,
the left-right length difference is usually about 2mm.
The 482SL hub is a high-low flange, so this length difference comes out to about 2.5mm in theory
regardless of rim inner diameter, but
if you don't have a spoke cutter and can only choose lengths in 1mm increments—
2mm or 3mm difference—then 2mm is fine.
There was a 6mm difference in length between left and right.
Both spokes have a 2mm outer diameter on the round part before the threads start, so
using that as a reference, even visually you can see the 6mm difference.
With a difference this large, if one spoke length was correct,
the other would definitely be too long or too short.
In reality, both are wrong. As I wrote earlier,
the freewheel side race was too short,
and the anti-freewheel side Pillar aero was slightly long.
The Pillar ones could be fixed with a little cutting, but
they probably don't own a spoke cutter.
If they don't, they should have sourced the right spoke lengths from the start,
but that's laziness too.
Or maybe their "eye is so bad" they think this is acceptable.
Man, this guy probably does the same thing with frames—
using mismatched components that are the wrong length. ←I'm being repetitive
Built it.
Before I saw the spoke lengths on the original wheel, I was planning
to reuse the race spokes on the freewheel side and swap to CX-RAY on the anti-freewheel side
for something semi-competition level, but
I went with cut Pillar aero on the freewheel side,
and CX-RAY on the anti-freewheel side.
I'll do the twist-tie lacing later, but
it might be good to show the customer the wheel first without twist ties,
before deciding.
Since we can't source Pillar aero spokes at the shop,
if the freewheel side spokes break again and if there's a situation
where they can't go back to Velarete (wink wink),
I'll swap everything to competition spokes.
For spoke weight ratio, I'm using 85% as a rough estimate for 2.0-1.8-2.0mm round butted spokes,
but DT Competition has longer 2.0mm sections on both ends, so
actual measurement gives about 86.5%.
Sapim race comes out to about 85%, but
before I weighed them, I calculated at 85% and got 74.4g.
Then when I actually weighed the spokes, they came to 74.0g, so
the actual spoke weight ratio for this batch of race is 84.48%.
Sapim CX, which like Pillar aero uses the slit-hole design,
shows a theoretical weight ratio of 98.9% from the spec weight,
but actual measurement gives 100.3%.
This batch of Pillar spokes has a thinner flat section than CX,
so I estimated around 95% spoke weight ratio, but
actual measurement came to 98.28%.
That's close enough to 100% like a plain 14g spoke that there's no problem.
Except that this reverse-gauge lacing is still problematic.
It's the equivalent of building a round-spoke rear wheel with competition on the freewheel side
and champion on the anti-freewheel side on purpose.
My rebuilt wheel, photographed from the right side
clockwise from the valve hole:
↑Anti-freewheel side
↑Freewheel side
↑Anti-freewheel side
↑Freewheel side
The freewheel side Pillar spokes are cut,
so the end is silver. The anti-freewheel side CX-RAY
didn't need cutting, so the end is black.
To Velarete:
In case you come across this post someday,
I'm writing this now.
Feel free to talk back to me, but
explain the rational reason why the rebuilt spokes are in reverse-gauge different-cross lacing
and why the spoke lengths differ by 6mm first.
If you can't do that, stay quiet forever.
Just do business with people who won't complain about your work,
and get by however you can.