Another day with wheels (and so on).
I received a rear wheel from a customer built with a 120mm-high tubular rim.
It's a trike rear wheel, but the trike (three-wheeler) I'm referring to here
is a recumbent with two front wheels and one rear wheel.
The customer has given me permission—or rather, specifically requested—so I'll write this,
but this was built by HC Works, a damn fool shop
that mainly deals with recumbents and trikes.
I'll explain what makes it damn foolish as I go,
but since there's a possibility of similar mistakes
on other wheels out there, this also serves as a warning.
A BCT brand 2:1 spoke pattern hub, 24H,
all-black Champion 14-gauge, which reads as 40-cross lacing.
With this rim being high-profile, the spokes are short,
but the structure on the inner side of the rim holes
is much the same as a 20-inch HE low-profile rim (spoke length is nearly identical).
With spokes this short, once tensioned to a certain point,
spoke deflection when squeezed becomes almost completely rigid,
but as you can see, this is abnormally loose on the freewheel side.
There's almost no runout or centering issues.
It was built with just a soft tug from the preliminary assembly stage.
As I'll mention later, when I half-disassembled it,
loosening all nipples two full turns allowed me to
loosen them by hand after that.
When I press the measurement probe of a Hozan tension meter...
The spokes deflect like this.
I checked all freewheel-side spokes.
H1ST here is 74. The next one was 72.
Most locations are under 80, with a few reaching around 90.
The highest reading was 109.5.
The lower the tension, the more variation in spoke tension.
The relationship between H1ST and second ST for plain spokes
isn't particularly secret knowledge, so I'll show the chart too,
but when H1ST is under 80, the second ST is under 150N,
which becomes "too low to measure."
At the highest H1ST of 109.5, it's around 500N,
but only the final-cross pair exceeded 100.
Just to be clear, this isn't the main point of this post.
The anti-freewheel side 0-cross lacing on this rear wheel isn't radial.
It has a phase offset—it's twisted.
It might be easier to see if tilted.
It's hard to believe this wasn't noticed before finishing the wheel,
so this must be "normal" for HC Works.
I expect to hear excuses like "this condition is unavoidable with this hub and rim,"
so I'll provide counterevidence below.
With small-wheel builds, there are cases where a 36H hub
must be laced to a 24H rim with equal numbers left and right
(such as when building with Alfine internal hub),
and in those cases phase offset does occur.
But this rear wheel can be built without phase offset.
Colima makes a wheel called the MCC,
which uses 0-cross carbon spokes on both front and rear, left and right.
The front wheel is radial, but
the rear wheel has both sides tilted in the porcupine direction.
Since it's a drive wheel, this must reflect consciousness of
the force direction when pedaling twists the freewheel body hard.
If this rear wheel absolutely had to be laced this way,
the anti-freewheel 0-cross should tilt porcupine-direction,
but it's actually the opposite.
Initially I thought the anti-freewheel spokes
had just been shifted one hole at a time,
but that's not the case.
When I place a silver spoke on the radial line
from the 0-cross flange hole...
The rim hole along that extension already has a freewheel-side spoke.
Up to this point you might think "maybe this wheel can only be laced this way,"
but that's not true at all.
The freewheel side uses 4-cross (2 crosses), and if I place a silver spoke
parallel to the hub axle from directly below the final cross,
the anti-freewheel phase aligns with a flange hole.
Doing the same from directly below the first cross,
the anti-freewheel phase falls exactly halfway between flange holes.
The spoke in the image above isn't red—it's the hub color reflecting off the silver spoke.
This means the hub has a phase equivalent to
a generic 32H hub with every other anti-freewheel flange hole blanked off.
My shop whiteboard has a blank side and a grid side,
and with the grid I can draw fairly accurate diagrams freehand.
If I make each grid square equal to 45° on the wheel diagram,
45° of a 32H rim equals four rim holes.
I extended that to 180°.
From there I drew a wheel diagram
in hub-rim-hub format using generic 32H hub lacing.
Let me try 0-cross lacing on both sides.
When spoke path doesn't create phase offset and
is perpendicular to the hub and rim lines in the diagram,
that 0-cross lacing becomes radial.
I erased the spokes.
This hub has every other anti-freewheel flange hole blocked off,
so I'll block them in the diagram too.
This makes the hub 16:8H at 24H
(though the diagram only shows 180° of it).
I changed the rim to 24H. 45° contains three rim holes.
I marked the rim hole distribution for the 2:1 pattern on the 24H rim.
Here, the fundamental mistake is not aligning the phases between the rim and left flange holes
while prioritizing anti-freewheel radial lacing.
I laced the freewheel side 0-cross.
All spokes deviate from the radial line, but
they alternate, deviating at equal angles in opposite directions,
so with tangent lacing, the combined direction of the final-cross pair
aligns with the radial line, preventing phase offset.
Connecting the remaining anti-freewheel hub and rim holes...
gives you this.
Or this.
Only the porcupine direction reverses.
This is the initial condition.
If only the freewheel tangent lacing is in place without the anti-freewheel spokes,
there's no phase offset,
but threading the anti-freewheel spokes
causes the freewheel tangent to twist slightly along with it.
In both the diagram and the actual initial condition,
the 2:1 was XI (X-eye) lacing,
but a wheel can be laced without phase offset with either XI or Zh (Zhe) lacing,
where the anti-freewheel radial spoke passes through the freewheel's final cross when viewed from the side.
Of the three parallel lines, the top and bottom hub lines are fixed,
but the middle rim line can be shifted left or right arbitrarily.
For a 24H rim with "anti-freewheel radial lacing" prioritized,
I drew spokes first.
At that point, I'd only drawn
the rim holes for those anti-freewheel spokes.
I added the freewheel spoke rim holes.
Two of them, dividing the anti-freewheel spoke holes into thirds.
I drew the freewheel spokes.
No phase offset, anti-freewheel spokes remain radial.
The above shows Zh lacing, but
it can also be XI lacing.
I prefer the obtuse angle at the final cross,
so I'd probably go with Zh.
I removed all spoke tension.
With tension gone, the freewheel spokes
take on the anti-freewheel twist more noticeably,
and you can see the final-cross tip is twisted.
If I remove the anti-freewheel spokes from here,
the freewheel tangent lacing shows no twist.
From here, I'll remove the freewheel nipples but
leave all the spokes threaded through the flange holes untouched,
converting to a phase-offset-free 2:1 pattern.
The required spoke length changes,
but since this is a mock-up, that's fine.
As I'll explain later, the spoke length itself was wrong anyway
(or how exactly was the anti-freewheel twisted spoke length determined...?).
"Removing all freewheel nipples" corresponds to
"securing anti-freewheel radial lacing" in the earlier diagram.
I did the mock-up.
Zh lacing.
No phase offset,
and the anti-freewheel 0-cross is radial.
Got it now, you damn fool?
If I wanted to convert this to XI lacing from here,
for example with 4-cross,
from the marked flange hole on the tape,
you can do it this way.
As I mentioned, I prefer the obtuse final-cross angle,
so if I did this, I'd use 6-cross Zh.
Also, the original pattern and Zh conversion use
Italian-style nu-spoke and anti-nu-spoke,
but the XI lacing in the image reverses them,
so to avoid that, you'd need to remove spokes from the hub.
This is a Racing Zero rear wheel with offset phase,
and it's XI laced.
When Fulcrum had evenly-spaced rim holes, Zh was also used,
but with offset phases, it's basically XI.
Unlike this case, the offset phase is on the rim side,
so whether the anti-freewheel spokes create Zh or XI
doesn't change the freewheel final-cross angle.
This rear wheel has a "28H anti-freewheel radial lacing
equal-number lacing wheel
with every other anti-freewheel spoke removed" structure—21H.
If you reverse the removal phase on the anti-freewheel side,
the spoke path becomes either this,
or this.
The circumstances differ somewhat, but
today's situation is similar.
I received a rear wheel from a customer built with a 120mm-high tubular rim.
It's a trike rear wheel, but the trike (three-wheeler) I'm referring to here
is a recumbent with two front wheels and one rear wheel.
The customer has given me permission—or rather, specifically requested—so I'll write this,
but this was built by HC Works, a damn fool shop
that mainly deals with recumbents and trikes.
I'll explain what makes it damn foolish as I go,
but since there's a possibility of similar mistakes
on other wheels out there, this also serves as a warning.
A BCT brand 2:1 spoke pattern hub, 24H,
all-black Champion 14-gauge, which reads as 40-cross lacing.
With this rim being high-profile, the spokes are short,
but the structure on the inner side of the rim holes
is much the same as a 20-inch HE low-profile rim (spoke length is nearly identical).
With spokes this short, once tensioned to a certain point,
spoke deflection when squeezed becomes almost completely rigid,
but as you can see, this is abnormally loose on the freewheel side.
There's almost no runout or centering issues.
It was built with just a soft tug from the preliminary assembly stage.
As I'll mention later, when I half-disassembled it,
loosening all nipples two full turns allowed me to
loosen them by hand after that.
When I press the measurement probe of a Hozan tension meter...
The spokes deflect like this.
I checked all freewheel-side spokes.
H1ST here is 74. The next one was 72.
Most locations are under 80, with a few reaching around 90.
The highest reading was 109.5.
The lower the tension, the more variation in spoke tension.
The relationship between H1ST and second ST for plain spokes
isn't particularly secret knowledge, so I'll show the chart too,
but when H1ST is under 80, the second ST is under 150N,
which becomes "too low to measure."
At the highest H1ST of 109.5, it's around 500N,
but only the final-cross pair exceeded 100.
Just to be clear, this isn't the main point of this post.
The anti-freewheel side 0-cross lacing on this rear wheel isn't radial.
It has a phase offset—it's twisted.
It might be easier to see if tilted.
It's hard to believe this wasn't noticed before finishing the wheel,
so this must be "normal" for HC Works.
I expect to hear excuses like "this condition is unavoidable with this hub and rim,"
so I'll provide counterevidence below.
With small-wheel builds, there are cases where a 36H hub
must be laced to a 24H rim with equal numbers left and right
(such as when building with Alfine internal hub),
and in those cases phase offset does occur.
But this rear wheel can be built without phase offset.
Colima makes a wheel called the MCC,
which uses 0-cross carbon spokes on both front and rear, left and right.
The front wheel is radial, but
the rear wheel has both sides tilted in the porcupine direction.
Since it's a drive wheel, this must reflect consciousness of
the force direction when pedaling twists the freewheel body hard.
If this rear wheel absolutely had to be laced this way,
the anti-freewheel 0-cross should tilt porcupine-direction,
but it's actually the opposite.
Initially I thought the anti-freewheel spokes
had just been shifted one hole at a time,
but that's not the case.
When I place a silver spoke on the radial line
from the 0-cross flange hole...
The rim hole along that extension already has a freewheel-side spoke.
Up to this point you might think "maybe this wheel can only be laced this way,"
but that's not true at all.
The freewheel side uses 4-cross (2 crosses), and if I place a silver spoke
parallel to the hub axle from directly below the final cross,
the anti-freewheel phase aligns with a flange hole.
Doing the same from directly below the first cross,
the anti-freewheel phase falls exactly halfway between flange holes.
The spoke in the image above isn't red—it's the hub color reflecting off the silver spoke.
This means the hub has a phase equivalent to
a generic 32H hub with every other anti-freewheel flange hole blanked off.
My shop whiteboard has a blank side and a grid side,
and with the grid I can draw fairly accurate diagrams freehand.
If I make each grid square equal to 45° on the wheel diagram,
45° of a 32H rim equals four rim holes.
I extended that to 180°.
From there I drew a wheel diagram
in hub-rim-hub format using generic 32H hub lacing.
Let me try 0-cross lacing on both sides.
When spoke path doesn't create phase offset and
is perpendicular to the hub and rim lines in the diagram,
that 0-cross lacing becomes radial.
I erased the spokes.
This hub has every other anti-freewheel flange hole blocked off,
so I'll block them in the diagram too.
This makes the hub 16:8H at 24H
(though the diagram only shows 180° of it).
I changed the rim to 24H. 45° contains three rim holes.
I marked the rim hole distribution for the 2:1 pattern on the 24H rim.
Here, the fundamental mistake is not aligning the phases between the rim and left flange holes
while prioritizing anti-freewheel radial lacing.
I laced the freewheel side 0-cross.
All spokes deviate from the radial line, but
they alternate, deviating at equal angles in opposite directions,
so with tangent lacing, the combined direction of the final-cross pair
aligns with the radial line, preventing phase offset.
Connecting the remaining anti-freewheel hub and rim holes...
gives you this.
Or this.
Only the porcupine direction reverses.
This is the initial condition.
If only the freewheel tangent lacing is in place without the anti-freewheel spokes,
there's no phase offset,
but threading the anti-freewheel spokes
causes the freewheel tangent to twist slightly along with it.
In both the diagram and the actual initial condition,
the 2:1 was XI (X-eye) lacing,
but a wheel can be laced without phase offset with either XI or Zh (Zhe) lacing,
where the anti-freewheel radial spoke passes through the freewheel's final cross when viewed from the side.
Of the three parallel lines, the top and bottom hub lines are fixed,
but the middle rim line can be shifted left or right arbitrarily.
For a 24H rim with "anti-freewheel radial lacing" prioritized,
I drew spokes first.
At that point, I'd only drawn
the rim holes for those anti-freewheel spokes.
I added the freewheel spoke rim holes.
Two of them, dividing the anti-freewheel spoke holes into thirds.
I drew the freewheel spokes.
No phase offset, anti-freewheel spokes remain radial.
The above shows Zh lacing, but
it can also be XI lacing.
I prefer the obtuse angle at the final cross,
so I'd probably go with Zh.
I removed all spoke tension.
With tension gone, the freewheel spokes
take on the anti-freewheel twist more noticeably,
and you can see the final-cross tip is twisted.
If I remove the anti-freewheel spokes from here,
the freewheel tangent lacing shows no twist.
From here, I'll remove the freewheel nipples but
leave all the spokes threaded through the flange holes untouched,
converting to a phase-offset-free 2:1 pattern.
The required spoke length changes,
but since this is a mock-up, that's fine.
As I'll explain later, the spoke length itself was wrong anyway
(or how exactly was the anti-freewheel twisted spoke length determined...?).
"Removing all freewheel nipples" corresponds to
"securing anti-freewheel radial lacing" in the earlier diagram.
I did the mock-up.
Zh lacing.
No phase offset,
and the anti-freewheel 0-cross is radial.
Got it now, you damn fool?
If I wanted to convert this to XI lacing from here,
for example with 4-cross,
from the marked flange hole on the tape,
you can do it this way.
As I mentioned, I prefer the obtuse final-cross angle,
so if I did this, I'd use 6-cross Zh.
Also, the original pattern and Zh conversion use
Italian-style nu-spoke and anti-nu-spoke,
but the XI lacing in the image reverses them,
so to avoid that, you'd need to remove spokes from the hub.
This is a Racing Zero rear wheel with offset phase,
and it's XI laced.
When Fulcrum had evenly-spaced rim holes, Zh was also used,
but with offset phases, it's basically XI.
Unlike this case, the offset phase is on the rim side,
so whether the anti-freewheel spokes create Zh or XI
doesn't change the freewheel final-cross angle.
This rear wheel has a "28H anti-freewheel radial lacing
equal-number lacing wheel
with every other anti-freewheel spoke removed" structure—21H.
If you reverse the removal phase on the anti-freewheel side,
the spoke path becomes either this,
or this.
The circumstances differ somewhat, but
today's situation is similar.
Related Products on Amazon
* Amazon affiliate links — prices may vary