Wheel building again today (and so on).
A customer (supposedly) entrusted me with a rear wheel built with an EDGE 1-45 rim.
The hub is an Tni Evolution hub (the predecessor to the Evolite), with 24H, all CX-RAY 4-crossing lacing.
It's clearly not a buildable lacing pattern,
but getting asked to rebuild it is such a hassle,
so I tried to brush past it just with an inspection a while back,
but that didn't work out.
The serial number is 6 digits, in the 1,030,000 range—right at the tail end of the EDGE era before it became ENVE.
As a replacement rear hub, I received a Hyperon (ハイペロン) hub.
But this is tough.
It's not about spoke length calculations per se—
it's that even if I rebuild with this hub, the equal-count left-right and antifreewheel-side radial lacing pattern remains unchanged.
↑ Near the valve hole, viewed from above
↑ From the right
↑ From the left
This is a Hyperon rear wheel.
The Hyperon rear wheel, with its offset rim and extremely asymmetric left-right flange heights and high-low flanges,
manages to bring antifreewheel-side radial lacing to a point where it's just barely superior
compared to a non-offset rim, equal-count 24H, all CX-RAY 4-crossing rear wheel.
It's a rare rear wheel in that respect.
Within normal parameters, it's impossible for an antifreewheel-side radial wheel to achieve
spoke deflection resistance comparable to an antifreewheel-side tangent wheel.
The ENVE rim is not an offset rim, so even if you dramatically increase the left-right spoke weight difference,
you can't overcome the major factor of being an offset rim.
For the same Hyperon rear hub,
comparing a wheel with original Hyperon specs (meaning stock Hyperon)
versus one with half-comp spokes and a 1-45 rim—
both with wheel center checked, obviously—
when you grip the finished wheel's spokes,
the stock Hyperon should show less deformation in the antifreewheel-side spokes.
But there was one factor that could flip the outcome:
The ENVE rim has a higher spoke tension tolerance limit.
It can tension right up to the allowable limit (naturally the freewheel side only),
and if the antifreewheel side follows along, minus the offset amount...
This Hyperon hub, however, had ceramic ball retainers.
That's chronologically odd.
Because the ball race was black.
In 2004, Hyperon Ultra came out as the first WO carbon rim wheel,
but at that point Hyperon was still tubular spec only,
with steel ball bearing spec.
Black ball race + ceramic balls is the same as early USB, but
this combination has heavy ball race wear and is prone to pitting.
After wiping the grease and checking, this hub showed
clean wear marks with no pitting.
Let me review the Hyperon Ultra timeline.
This is the 2003 spare parts catalog.
Only Hyperon is listed.
Carbon WO rims didn't exist yet, so
naturally there's only tubular rim spec without clarification.
The bearing balls are HB-RE023, steel balls.
HB means hub parts, RE means Record grade,
and numbers like 023 with X23 endings are the designation for bearing balls.
I couldn't find the 2004 catalog right away, so here's the 2005 one.
The first generation Hyperon Ultra
only appeared in 2004 and 2005,
and Hyperon specs were the same for both years.
This is the hub exploded view page for five models:
Hyperon, Hyperon Ultra, Bora,
Neutron, and Eurus (which I unofficially call that to distinguish the steel spoke version from Urus)—
Only Hyperon Ultra has ceramic ball spec,
so parts are also HY, the Hyperon part number.
The hub dust cap's trailing number is 011, but
whereas the conventional black cap is Bora part number BO011,
the first generation Hyperon Ultra's red cap alone
is part number HY011.
Next are the 2006 and 2007 catalogs.
The Hyperon section content was the same.
Hyperon and Hyperon Ultra are together on the same page, but...
Steel ball bearing spec only—
Black cap spec only.
This second-generation Hyperon Ultra
"discontinued" the ceramic bearings and red cap of the first generation—
content-wise it's just
the WO rim version of the traditional tubular rim Hyperon.
From this embarrassing failure regarding ceramic bearings,
there followed a period with only steel ball bearings,
and CULT wouldn't appear until much later.
Then USB came out even later, but
initially it was black ball race + ceramic balls,
the same specs as first-gen Hyperon Ultra.
I thought, "It'll pit just as quick as that!" But sure enough,
pitting incidents became frequent,
and through an unannounced, even part-number-unchanged spec revision,
it switched to silver ball race (in-house-manufactured pseudo-CULT),
after which pitting incidents dropped dramatically.
Incidentally, the first-gen Hyperon Ultra's
ceramic ball retainer part number is HY023, but
the same part on current CULT and USB is HY123—
the hundreds digit advanced, a different part.
So the Hyperon hub I received had
ceramic balls and a black cap, which is why the dating seemed off.
Plausible scenarios are:
·First-gen Hyperon Ultra hub with cap swapped to black
·2004–2005 era hub with HB-HY123 retrofit later
These two are most likely.
Unlike CULT, which requires grease
and can't achieve CULT's characteristic smooth rotation,
swapping to steel ball bearing doesn't much change rotation performance when it comes to pitting risk reduction.
Alternatively, there's the option of changing the ball race itself to CULT spec, but...
On Hyperon and Bora carbon hubs from this era,
the front left and right, and rear left side have a design where
if spokes are laced through, the cone can't be replaced.
I didn't suggest upgrading to CULT cones
before building the wheel, so CULT upgrade is off the table!
On the original wheel, I loosened every nipple several turns to fully relax spoke tension before cutting the spokes.
The image above is those spokes.
Current ENVE is the 4th generation of integrated nipples, but
since this rim is late EDGE,
the 2nd generation integrated nipples are used.
But there's a mistake common to 2nd gen: the orientation was reversed during assembly.
↑ This side with the wrench flat is the inner side.
↑ This side of the nipple has no wrench flat
and thread starts abruptly,
so if spoke length is even slightly long,
you run out of thread and bottom out.
From carbon contamination at the thread root,
it looks like there were 2–3 more turns available.
The first-gen integrated nipples are also aluminum hexagonal posts,
but have nylon locking compound embedded at the outer end,
so it's nearly impossible to reverse even by accident.
Thus reversed nipples are a mistake unique to 2nd gen.
At... last I got there...
The moment center was achieved felt like catching a fingertip on a mountain peak.
It took forever...
Built.
The freewheel side is black Champion.
I'm doing the final crossing,
and this is the same on the stock Hyperon too.
The antifreewheel side is black CX-RAY.
The left-right spoke weight ratio difference is slightly larger than stock Hyperon
(freewheel side is larger, antifreewheel side about the same).
Since the left flange is wider than the original Evolution hub and the offset is larger,
lateral stiffness is achievable, but the antifreewheel side should go limp.
Yet the antifreewheel side tensioned better than the original Evolution + all CX-RAY.
Originally the antifreewheel side deflected mushily, but
after the rebuild it's almost nonexistent.
To be clear, the original state refers to after I inspected,
did a true check and some tightening.
Even so, the antifreewheel-side stiffness trails stock Hyperon
and also trails ordinary 24H hub half-comp 4-crossing lacing
without cross-threading.
But among "1-45 rim rear wheels with CX-RAY radial antifreewheel lacing,"
this might be the stiffer antifreewheel-side rear wheel in the world.
Hyperon is a reverse-rim wheel on both front and rear.
So when viewed from the right side,
the rim hole clockwise next to the valve hole
becomes a freewheel-side spoke
(the spoke from the near flange from the viewer's perspective).
For this rebuild, I built it as a normal rim.
The rim-side holes have no offset.
When viewed from the right, the rim hole clockwise next to the valve hole
becomes an antifreewheel-side spoke
(the spoke from the far flange from the viewer's perspective).
I drew the left-right tangent 4-spoke final crossing pair when viewing a normal-rim wheel from the right
(from one side, but let's say from the right).
Normal rim, put another way, means
"the front final crossing as seen by the viewer
is twisted ahead of the rear final crossing toward the clockwise direction."
I straighten the arc of the rim inner edge and the hub flange arc into equal-length parallel lines.
From here, opening the left spoke diagram
in the reverse procedure of folding it downward yields
A rim-hub-rim type wheel unfolding.
I add one more cycle.
When determining the valve hole phase,
the only place where a perpendicular line can be drawn
without touching the spokes in the unfolding is here.
The right rim final crossing shifts right in the diagram
because it's a normal rim.
Reverse rim, in a wheel unfolding,
is mirror-reversed relative to normal rim.
If setting the phase so the valve hole doesn't land
within the final crossing 4-spoke pair, there's only one place.
I added detail to the reverse rim diagram, but
a straight-spoke hub flange that enforces left-right tangent lacing on both sides
means each flange hole has directional intent:
some holes are dedicated to spoke heads-in spoke direction
and others to outpoke direction.
So combining a forced left-right tangent lacing hub for normal rim
with a reverse-rim-offset rim doesn't allow both
"getting the offset right" and
"valve hole phase within the final crossing 4-spoke pair"
to coexist.
But such wheels do exist normally...actually they did exist (here).
This Hyperon hub has one side (the antifreewheel side)
forced to radial lacing, so it's not left-right tangent lacing.
When radial spoke lacing is drawn in a wheel unfolding,
it becomes a line perpendicular to the parallel lines.
By the way, like the few-spoke side of Shimano's Optibral,
zero-crossing lacing with compressed hub-phase range
deviates the spoke path from a radial line from hub center,
so it's not radial lacing in the strict sense,
and the unfolding doesn't show a perpendicular to the parallel lines either.
So originally a forced left-right tangent for reverse rim hub,
but I redrew the antifreewheel-side flange with radial forced lacing intent.
↑ Like this
For a hub with one radial side,
the valve hole phases avoiding the final crossing tangent spokes
in the unfolding are
Like this, one more place.
In other words, "I chose to treat it as a normal rim."
With one side radial,
it's no longer exclusive to normal-rim (reverse-rim) hubs.
Since this EDGE rim had no offset,
either would work, but I chose normal rim.
If the rim had a normal offset, it must be built as normal rim.
If the Hyperon hub happened to be
a forced left-right tangent reverse-rim-only hub,
I would've asked the customer, "The valve hole position will look awkward—okay?"
A customer (supposedly) entrusted me with a rear wheel built with an EDGE 1-45 rim.
The hub is an Tni Evolution hub (the predecessor to the Evolite), with 24H, all CX-RAY 4-crossing lacing.
It's clearly not a buildable lacing pattern,
but getting asked to rebuild it is such a hassle,
so I tried to brush past it just with an inspection a while back,
but that didn't work out.
The serial number is 6 digits, in the 1,030,000 range—right at the tail end of the EDGE era before it became ENVE.
As a replacement rear hub, I received a Hyperon (ハイペロン) hub.
But this is tough.
It's not about spoke length calculations per se—
it's that even if I rebuild with this hub, the equal-count left-right and antifreewheel-side radial lacing pattern remains unchanged.
↑ Near the valve hole, viewed from above
↑ From the right
↑ From the left
This is a Hyperon rear wheel.
The Hyperon rear wheel, with its offset rim and extremely asymmetric left-right flange heights and high-low flanges,
manages to bring antifreewheel-side radial lacing to a point where it's just barely superior
compared to a non-offset rim, equal-count 24H, all CX-RAY 4-crossing rear wheel.
It's a rare rear wheel in that respect.
Within normal parameters, it's impossible for an antifreewheel-side radial wheel to achieve
spoke deflection resistance comparable to an antifreewheel-side tangent wheel.
The ENVE rim is not an offset rim, so even if you dramatically increase the left-right spoke weight difference,
you can't overcome the major factor of being an offset rim.
For the same Hyperon rear hub,
comparing a wheel with original Hyperon specs (meaning stock Hyperon)
versus one with half-comp spokes and a 1-45 rim—
both with wheel center checked, obviously—
when you grip the finished wheel's spokes,
the stock Hyperon should show less deformation in the antifreewheel-side spokes.
But there was one factor that could flip the outcome:
The ENVE rim has a higher spoke tension tolerance limit.
It can tension right up to the allowable limit (naturally the freewheel side only),
and if the antifreewheel side follows along, minus the offset amount...
This Hyperon hub, however, had ceramic ball retainers.
That's chronologically odd.
Because the ball race was black.
In 2004, Hyperon Ultra came out as the first WO carbon rim wheel,
but at that point Hyperon was still tubular spec only,
with steel ball bearing spec.
Black ball race + ceramic balls is the same as early USB, but
this combination has heavy ball race wear and is prone to pitting.
After wiping the grease and checking, this hub showed
clean wear marks with no pitting.
Let me review the Hyperon Ultra timeline.
This is the 2003 spare parts catalog.
Only Hyperon is listed.
Carbon WO rims didn't exist yet, so
naturally there's only tubular rim spec without clarification.
The bearing balls are HB-RE023, steel balls.
HB means hub parts, RE means Record grade,
and numbers like 023 with X23 endings are the designation for bearing balls.
I couldn't find the 2004 catalog right away, so here's the 2005 one.
The first generation Hyperon Ultra
only appeared in 2004 and 2005,
and Hyperon specs were the same for both years.
This is the hub exploded view page for five models:
Hyperon, Hyperon Ultra, Bora,
Neutron, and Eurus (which I unofficially call that to distinguish the steel spoke version from Urus)—
Only Hyperon Ultra has ceramic ball spec,
so parts are also HY, the Hyperon part number.
The hub dust cap's trailing number is 011, but
whereas the conventional black cap is Bora part number BO011,
the first generation Hyperon Ultra's red cap alone
is part number HY011.
Next are the 2006 and 2007 catalogs.
The Hyperon section content was the same.
Hyperon and Hyperon Ultra are together on the same page, but...
Steel ball bearing spec only—
Black cap spec only.
This second-generation Hyperon Ultra
"discontinued" the ceramic bearings and red cap of the first generation—
content-wise it's just
the WO rim version of the traditional tubular rim Hyperon.
From this embarrassing failure regarding ceramic bearings,
there followed a period with only steel ball bearings,
and CULT wouldn't appear until much later.
Then USB came out even later, but
initially it was black ball race + ceramic balls,
the same specs as first-gen Hyperon Ultra.
I thought, "It'll pit just as quick as that!" But sure enough,
pitting incidents became frequent,
and through an unannounced, even part-number-unchanged spec revision,
it switched to silver ball race (in-house-manufactured pseudo-CULT),
after which pitting incidents dropped dramatically.
Incidentally, the first-gen Hyperon Ultra's
ceramic ball retainer part number is HY023, but
the same part on current CULT and USB is HY123—
the hundreds digit advanced, a different part.
So the Hyperon hub I received had
ceramic balls and a black cap, which is why the dating seemed off.
Plausible scenarios are:
·First-gen Hyperon Ultra hub with cap swapped to black
·2004–2005 era hub with HB-HY123 retrofit later
These two are most likely.
Unlike CULT, which requires grease
and can't achieve CULT's characteristic smooth rotation,
swapping to steel ball bearing doesn't much change rotation performance when it comes to pitting risk reduction.
Alternatively, there's the option of changing the ball race itself to CULT spec, but...
On Hyperon and Bora carbon hubs from this era,
the front left and right, and rear left side have a design where
if spokes are laced through, the cone can't be replaced.
I didn't suggest upgrading to CULT cones
before building the wheel, so CULT upgrade is off the table!
On the original wheel, I loosened every nipple several turns to fully relax spoke tension before cutting the spokes.
The image above is those spokes.
Current ENVE is the 4th generation of integrated nipples, but
since this rim is late EDGE,
the 2nd generation integrated nipples are used.
But there's a mistake common to 2nd gen: the orientation was reversed during assembly.
↑ This side with the wrench flat is the inner side.
↑ This side of the nipple has no wrench flat
and thread starts abruptly,
so if spoke length is even slightly long,
you run out of thread and bottom out.
From carbon contamination at the thread root,
it looks like there were 2–3 more turns available.
The first-gen integrated nipples are also aluminum hexagonal posts,
but have nylon locking compound embedded at the outer end,
so it's nearly impossible to reverse even by accident.
Thus reversed nipples are a mistake unique to 2nd gen.
At... last I got there...
The moment center was achieved felt like catching a fingertip on a mountain peak.
It took forever...
Built.
The freewheel side is black Champion.
I'm doing the final crossing,
and this is the same on the stock Hyperon too.
The antifreewheel side is black CX-RAY.
The left-right spoke weight ratio difference is slightly larger than stock Hyperon
(freewheel side is larger, antifreewheel side about the same).
Since the left flange is wider than the original Evolution hub and the offset is larger,
lateral stiffness is achievable, but the antifreewheel side should go limp.
Yet the antifreewheel side tensioned better than the original Evolution + all CX-RAY.
Originally the antifreewheel side deflected mushily, but
after the rebuild it's almost nonexistent.
To be clear, the original state refers to after I inspected,
did a true check and some tightening.
Even so, the antifreewheel-side stiffness trails stock Hyperon
and also trails ordinary 24H hub half-comp 4-crossing lacing
without cross-threading.
But among "1-45 rim rear wheels with CX-RAY radial antifreewheel lacing,"
this might be the stiffer antifreewheel-side rear wheel in the world.
Hyperon is a reverse-rim wheel on both front and rear.
So when viewed from the right side,
the rim hole clockwise next to the valve hole
becomes a freewheel-side spoke
(the spoke from the near flange from the viewer's perspective).
For this rebuild, I built it as a normal rim.
The rim-side holes have no offset.
When viewed from the right, the rim hole clockwise next to the valve hole
becomes an antifreewheel-side spoke
(the spoke from the far flange from the viewer's perspective).
I drew the left-right tangent 4-spoke final crossing pair when viewing a normal-rim wheel from the right
(from one side, but let's say from the right).
Normal rim, put another way, means
"the front final crossing as seen by the viewer
is twisted ahead of the rear final crossing toward the clockwise direction."
I straighten the arc of the rim inner edge and the hub flange arc into equal-length parallel lines.
From here, opening the left spoke diagram
in the reverse procedure of folding it downward yields
A rim-hub-rim type wheel unfolding.
I add one more cycle.
When determining the valve hole phase,
the only place where a perpendicular line can be drawn
without touching the spokes in the unfolding is here.
The right rim final crossing shifts right in the diagram
because it's a normal rim.
Reverse rim, in a wheel unfolding,
is mirror-reversed relative to normal rim.
If setting the phase so the valve hole doesn't land
within the final crossing 4-spoke pair, there's only one place.
I added detail to the reverse rim diagram, but
a straight-spoke hub flange that enforces left-right tangent lacing on both sides
means each flange hole has directional intent:
some holes are dedicated to spoke heads-in spoke direction
and others to outpoke direction.
So combining a forced left-right tangent lacing hub for normal rim
with a reverse-rim-offset rim doesn't allow both
"getting the offset right" and
"valve hole phase within the final crossing 4-spoke pair"
to coexist.
But such wheels do exist normally...actually they did exist (here).
This Hyperon hub has one side (the antifreewheel side)
forced to radial lacing, so it's not left-right tangent lacing.
When radial spoke lacing is drawn in a wheel unfolding,
it becomes a line perpendicular to the parallel lines.
By the way, like the few-spoke side of Shimano's Optibral,
zero-crossing lacing with compressed hub-phase range
deviates the spoke path from a radial line from hub center,
so it's not radial lacing in the strict sense,
and the unfolding doesn't show a perpendicular to the parallel lines either.
So originally a forced left-right tangent for reverse rim hub,
but I redrew the antifreewheel-side flange with radial forced lacing intent.
↑ Like this
For a hub with one radial side,
the valve hole phases avoiding the final crossing tangent spokes
in the unfolding are
Like this, one more place.
In other words, "I chose to treat it as a normal rim."
With one side radial,
it's no longer exclusive to normal-rim (reverse-rim) hubs.
Since this EDGE rim had no offset,
either would work, but I chose normal rim.
If the rim had a normal offset, it must be built as normal rim.
If the Hyperon hub happened to be
a forced left-right tangent reverse-rim-only hub,
I would've asked the customer, "The valve hole position will look awkward—okay?"
Products Mentioned in This Article
- 🛒 EDGE 1-45 bicycle rim
- 🛒 Sapim CX-RAY bicycle spoke
- 🛒 Campagnolo Hyperon hub
- 🛒 Campagnolo Hyperon Ultra wheelset
* Amazon affiliate links — prices may vary
Original Japanese post: のむラボ日記 #6804