Spoke-squeezing tension effects

For a few months now, I've been squeezing spoke pairs on various
aluminum 700c wheels and measuring spoke tension changes for this
often-recommended procedure.

Most of the testing was just to convince myself that my results were
accurate, so I'll spare you page after page of tables and graphs.

The test is simple. Surprisingly, so were the results.

Measure the initial tension of the four spokes.

Hang a 60-lb weight from the midspan of the upper spoke of a
horizontal pair of parallel spokes, and support the lower spoke with a
ceiling rope.

Squeeze the parallel pair of spokes on the other side of the bike
together with a screw-clamp until the gap matches the gap measured
when that pair was roped and weighted with 60 lbs.

Measure the tension of the four spokes again.

Repeat for more spoke quartets around the wheel.

Raise the spoke tension. Match initial tensions. Let tensions vary.

Measure rising tension as the screw-clamp is tightened, turn by turn.

Do wheels with straight or butted spokes.

Test dished rear wheels and symmetrical front wheels, obviously with
different hub widths. The spoke pairs on the dished wheels start out
with different initial tensions.

Different flange heights.

Include 32 and 36 spoke rims.

Cheap rims with no box section (just a U-curve), sockets, or eyelets.

Better rims with box section and eyelets, but no sockets.

Rims with box section, eyelets, and sockets.

None of these variations made any significant difference.

Spoke tension kept rising roughly as much as the squeeze force.

That is, a 60-lb squeeze on each pair raises the tension in each of
the 4 spokes about 55 to 65 pounds.

So I'm curious if anyone can find and measure a reasonably normal 700c
aluminum wheel whose 32-36 spokes behave significantly differently,
say a 120-lb tension increase for a 60-lb squeeze.

(A few months ago, I was the only poster to greet with skepticism a
post that estimated in good faith a 156-lb tension increase for a
30-lb squeeze, while my initial tests with much lower results provoked
comments such as "impossible" and "cheesy wheels" from engineers.)

Steel rims, really deep aero rims, extra-wide hubs, super-tall
flanges, and higher and lower spoke counts are obviously not normal,
though I don't know if they'd behave differently. I tested only
cross-3 lacing, but I'm willing to call even radial lacing normal.

If you're curious and want some details about how to test, feel free
to ask--it's a little trickier than weighing a sack of wheat.

(Don't waste time trying to use the wheel itself as a makeshift
tension gauge. You can't calculate tension by spoke displacement or
bend angles. That approach assumes no rim distortion and leads to
wildly exaggerated tension estimates that suggest that a 90-lb squeeze
would stretch a spoke like taffy. Measure the actual tension by
slapping a tension gauge on the spoke between the rim and the squeeze
point.)

If you're curious and lazy--er, efficient . . . Wheels don't cost much
to ship and I'm willing to test a few wheels whose owners think that
they might produce different results. You'd pay to ship it to me, and
I'd pay to ship it back.

As for why the mechanical advantage seems to be only 1-to-1, it may be
that squeezing two pairs of parallel spokes distorts the circular rim
weirdly.

Viewed in a truing stand, the rim and the four spoke nipples A-B-C-D
look like the exaggerated view below.

When squeezed together, A & C pull the rim to the left. When squeezed
together, B & D pull the rim to the right. Obviously, the forces are
unbalanced and pull the rim into a zig-zag.

rim
|
|
y
/
|
z
/
|
spoke ----A
\
\
B----- spoke
|
|
|
spoke -------C
\
\
D--- spoke
|
/
e
|
/
f
|
|
rim

The crude ASCII diagram doesn't show things well, but the point is
that the upper rim is pulled left, while the lower rim is pulled to
the right.

The rim distortion can be emphasized by squeezing just one pair of
spokes on one side (not two pairs) with 60 lbs of force.

Tension increases in both spokes in the squeezed pair, but the
increase is noticeably less than 55-65 lbs.

Tension also increases in one spoke of the unsqueezed pair, but again
the increase is noticeably less than 55-65 lbs.

Weirdly, the other spoke in the unsqueezed pair loses significant
tension.

If you start squeezing the unsqueezed pair, spoke tension slowly
rises.

Tension rises slowly for the two spokes in the already-squeezed pair,
faster for unsqueezed spoke that showed the initial rise in tension,
and even faster for the spoke that originally lost tension.

When both pairs are squeezed with 60 lbs of force, all four spokes
have gained about 55-65 lbs of tension.

Cheers,

Carl Fogel

snippage of a lotta stuff

Cheers,

Carl Fogel

You have way too much time on your hands. Perhaps a more productive
pursuit?

This commentary from an inveterate futzer ;-)

D'ohBoy

Gentle Readers,

Reading posts such as the one below means you won't be needing Sominex
to sleep, sleep, sleep.

Cheers from the nether world,
L. Welk


wrote:
For a few months now, I've been squeezing spoke pairs on various
aluminum 700c wheels and measuring spoke tension changes for this
often-recommended procedure.

Most of the testing was just to convince myself that my results were
accurate, so I'll spare you page after page of tables and graphs.

The test is simple. Surprisingly, so were the results.

Measure the initial tension of the four spokes.

Hang a 60-lb weight from the midspan of the upper spoke of a
horizontal pair of parallel spokes, and support the lower spoke with a
ceiling rope.

Squeeze the parallel pair of spokes on the other side of the bike
together with a screw-clamp until the gap matches the gap measured
when that pair was roped and weighted with 60 lbs.

Measure the tension of the four spokes again.

Repeat for more spoke quartets around the wheel.

Raise the spoke tension. Match initial tensions. Let tensions vary.

Measure rising tension as the screw-clamp is tightened, turn by turn.

Do wheels with straight or butted spokes.

Test dished rear wheels and symmetrical front wheels, obviously with
different hub widths. The spoke pairs on the dished wheels start out
with different initial tensions.

Different flange heights.

Include 32 and 36 spoke rims.

Cheap rims with no box section (just a U-curve), sockets, or eyelets.

Better rims with box section and eyelets, but no sockets.

Rims with box section, eyelets, and sockets.

None of these variations made any significant difference.

Spoke tension kept rising roughly as much as the squeeze force.

That is, a 60-lb squeeze on each pair raises the tension in each of
the 4 spokes about 55 to 65 pounds.

So I'm curious if anyone can find and measure a reasonably normal 700c
aluminum wheel whose 32-36 spokes behave significantly differently,
say a 120-lb tension increase for a 60-lb squeeze.

(A few months ago, I was the only poster to greet with skepticism a
post that estimated in good faith a 156-lb tension increase for a
30-lb squeeze, while my initial tests with much lower results provoked
comments such as "impossible" and "cheesy wheels" from engineers.)

Steel rims, really deep aero rims, extra-wide hubs, super-tall
flanges, and higher and lower spoke counts are obviously not normal,
though I don't know if they'd behave differently. I tested only
cross-3 lacing, but I'm willing to call even radial lacing normal.

If you're curious and want some details about how to test, feel free
to ask--it's a little trickier than weighing a sack of wheat.

(Don't waste time trying to use the wheel itself as a makeshift
tension gauge. You can't calculate tension by spoke displacement or
bend angles. That approach assumes no rim distortion and leads to
wildly exaggerated tension estimates that suggest that a 90-lb squeeze
would stretch a spoke like taffy. Measure the actual tension by
slapping a tension gauge on the spoke between the rim and the squeeze
point.)

If you're curious and lazy--er, efficient . . . Wheels don't cost much
to ship and I'm willing to test a few wheels whose owners think that
they might produce different results. You'd pay to ship it to me, and
I'd pay to ship it back.

As for why the mechanical advantage seems to be only 1-to-1, it may be
that squeezing two pairs of parallel spokes distorts the circular rim
weirdly.

Viewed in a truing stand, the rim and the four spoke nipples A-B-C-D
look like the exaggerated view below.

When squeezed together, A & C pull the rim to the left. When squeezed
together, B & D pull the rim to the right. Obviously, the forces are
unbalanced and pull the rim into a zig-zag.

rim
|
|
y
/
|
z
/
|
spoke ----A
\
\
B----- spoke
|
|
|
spoke -------C
\
\
D--- spoke
|
/
e
|
/
f
|
|
rim

The crude ASCII diagram doesn't show things well, but the point is
that the upper rim is pulled left, while the lower rim is pulled to
the right.

The rim distortion can be emphasized by squeezing just one pair of
spokes on one side (not two pairs) with 60 lbs of force.

Tension increases in both spokes in the squeezed pair, but the
increase is noticeably less than 55-65 lbs.

Tension also increases in one spoke of the unsqueezed pair, but again
the increase is noticeably less than 55-65 lbs.

Weirdly, the other spoke in the unsqueezed pair loses significant
tension.

If you start squeezing the unsqueezed pair, spoke tension slowly
rises.

Tension rises slowly for the two spokes in the already-squeezed pair,
faster for unsqueezed spoke that showed the initial rise in tension,
and even faster for the spoke that originally lost tension.

When both pairs are squeezed with 60 lbs of force, all four spokes
have gained about 55-65 lbs of tension.

Cheers,

Carl Fogel

wrote:
For a few months now, I've been squeezing spoke pairs on various
aluminum 700c wheels and measuring spoke tension changes for this
often-recommended procedure.

Most of the testing was just to convince myself that my results were
accurate, so I'll spare you page after page of tables and graphs.

The test is simple. Surprisingly, so were the results.

Measure the initial tension of the four spokes.

Hang a 60-lb weight from the midspan of the upper spoke of a
horizontal pair of parallel spokes, and support the lower spoke with a
ceiling rope.

Squeeze the parallel pair of spokes on the other side of the bike
together with a screw-clamp until the gap matches the gap measured
when that pair was roped and weighted with 60 lbs.

Measure the tension of the four spokes again.

Repeat for more spoke quartets around the wheel.

Raise the spoke tension. Match initial tensions. Let tensions vary.

Measure rising tension as the screw-clamp is tightened, turn by turn.

Do wheels with straight or butted spokes.

Test dished rear wheels and symmetrical front wheels, obviously with
different hub widths. The spoke pairs on the dished wheels start out
with different initial tensions.

Different flange heights.

Include 32 and 36 spoke rims.

Cheap rims with no box section (just a U-curve), sockets, or eyelets.

Better rims with box section and eyelets, but no sockets.

Rims with box section, eyelets, and sockets.

None of these variations made any significant difference.

Spoke tension kept rising roughly as much as the squeeze force.

That is, a 60-lb squeeze on each pair raises the tension in each of
the 4 spokes about 55 to 65 pounds.

So I'm curious if anyone can find and measure a reasonably normal 700c
aluminum wheel whose 32-36 spokes behave significantly differently,
say a 120-lb tension increase for a 60-lb squeeze.

(A few months ago, I was the only poster to greet with skepticism a
post that estimated in good faith a 156-lb tension increase for a
30-lb squeeze, while my initial tests with much lower results provoked
comments such as "impossible" and "cheesy wheels" from engineers.)

Steel rims, really deep aero rims, extra-wide hubs, super-tall
flanges, and higher and lower spoke counts are obviously not normal,
though I don't know if they'd behave differently. I tested only
cross-3 lacing, but I'm willing to call even radial lacing normal.

If you're curious and want some details about how to test, feel free
to ask--it's a little trickier than weighing a sack of wheat.

(Don't waste time trying to use the wheel itself as a makeshift
tension gauge. You can't calculate tension by spoke displacement or
bend angles. That approach assumes no rim distortion and leads to
wildly exaggerated tension estimates that suggest that a 90-lb squeeze
would stretch a spoke like taffy. Measure the actual tension by
slapping a tension gauge on the spoke between the rim and the squeeze
point.)

If you're curious and lazy--er, efficient . . . Wheels don't cost much
to ship and I'm willing to test a few wheels whose owners think that
they might produce different results. You'd pay to ship it to me, and
I'd pay to ship it back.

As for why the mechanical advantage seems to be only 1-to-1, it may be
that squeezing two pairs of parallel spokes distorts the circular rim
weirdly.

Viewed in a truing stand, the rim and the four spoke nipples A-B-C-D
look like the exaggerated view below.

When squeezed together, A & C pull the rim to the left. When squeezed
together, B & D pull the rim to the right. Obviously, the forces are
unbalanced and pull the rim into a zig-zag.

rim
|
|
y
/
|
z
/
|
spoke ----A
\
\
B----- spoke
|
|
|
spoke -------C
\
\
D--- spoke
|
/
e
|
/
f
|
|
rim

The crude ASCII diagram doesn't show things well, but the point is
that the upper rim is pulled left, while the lower rim is pulled to
the right.

The rim distortion can be emphasized by squeezing just one pair of
spokes on one side (not two pairs) with 60 lbs of force.

Tension increases in both spokes in the squeezed pair, but the
increase is noticeably less than 55-65 lbs.

Tension also increases in one spoke of the unsqueezed pair, but again
the increase is noticeably less than 55-65 lbs.

Weirdly, the other spoke in the unsqueezed pair loses significant
tension.

If you start squeezing the unsqueezed pair, spoke tension slowly
rises.

Tension rises slowly for the two spokes in the already-squeezed pair,
faster for unsqueezed spoke that showed the initial rise in tension,
and even faster for the spoke that originally lost tension.

When both pairs are squeezed with 60 lbs of force, all four spokes
have gained about 55-65 lbs of tension.

Cheers,

Carl Fogel

the science of wheels never cease to amaze.

Well I probably missed out out all the discussions of this a while back. So
I don't have the whole story.
But although I can see how this is causing what you discuss below, I am
still sort of curious as to how this applies to me riding a bike along on
the wheels like this.
Maybe I need more coffee of something.
I do not see the correlation yet.
As I understand it, when you are riding along, the spokes on the bottom go
through a compression phase (getting more slack), with the spokes on top
going though a tension phase (getting more taut) at the same time.
I don't see anything causing a squeezing phase on the spokes when I ride
along.

wrote in message
...
For a few months now, I've been squeezing spoke pairs on various
aluminum 700c wheels and measuring spoke tension changes for this
often-recommended procedure.

Most of the testing was just to convince myself that my results were
accurate, so I'll spare you page after page of tables and graphs.

The test is simple. Surprisingly, so were the results.

Measure the initial tension of the four spokes.

Hang a 60-lb weight from the midspan of the upper spoke of a
horizontal pair of parallel spokes, and support the lower spoke with a
ceiling rope.

Squeeze the parallel pair of spokes on the other side of the bike
together with a screw-clamp until the gap matches the gap measured
when that pair was roped and weighted with 60 lbs.

Measure the tension of the four spokes again.

Repeat for more spoke quartets around the wheel.

Raise the spoke tension. Match initial tensions. Let tensions vary.

Measure rising tension as the screw-clamp is tightened, turn by turn.

Do wheels with straight or butted spokes.

Test dished rear wheels and symmetrical front wheels, obviously with
different hub widths. The spoke pairs on the dished wheels start out
with different initial tensions.

Different flange heights.

Include 32 and 36 spoke rims.

Cheap rims with no box section (just a U-curve), sockets, or eyelets.

Better rims with box section and eyelets, but no sockets.

Rims with box section, eyelets, and sockets.

None of these variations made any significant difference.

Spoke tension kept rising roughly as much as the squeeze force.

That is, a 60-lb squeeze on each pair raises the tension in each of
the 4 spokes about 55 to 65 pounds.

So I'm curious if anyone can find and measure a reasonably normal 700c
aluminum wheel whose 32-36 spokes behave significantly differently,
say a 120-lb tension increase for a 60-lb squeeze.

(A few months ago, I was the only poster to greet with skepticism a
post that estimated in good faith a 156-lb tension increase for a
30-lb squeeze, while my initial tests with much lower results provoked
comments such as "impossible" and "cheesy wheels" from engineers.)

Steel rims, really deep aero rims, extra-wide hubs, super-tall
flanges, and higher and lower spoke counts are obviously not normal,
though I don't know if they'd behave differently. I tested only
cross-3 lacing, but I'm willing to call even radial lacing normal.

If you're curious and want some details about how to test, feel free
to ask--it's a little trickier than weighing a sack of wheat.

(Don't waste time trying to use the wheel itself as a makeshift
tension gauge. You can't calculate tension by spoke displacement or
bend angles. That approach assumes no rim distortion and leads to
wildly exaggerated tension estimates that suggest that a 90-lb squeeze
would stretch a spoke like taffy. Measure the actual tension by
slapping a tension gauge on the spoke between the rim and the squeeze
point.)

If you're curious and lazy--er, efficient . . . Wheels don't cost much
to ship and I'm willing to test a few wheels whose owners think that
they might produce different results. You'd pay to ship it to me, and
I'd pay to ship it back.

As for why the mechanical advantage seems to be only 1-to-1, it may be
that squeezing two pairs of parallel spokes distorts the circular rim
weirdly.

Viewed in a truing stand, the rim and the four spoke nipples A-B-C-D
look like the exaggerated view below.

When squeezed together, A & C pull the rim to the left. When squeezed
together, B & D pull the rim to the right. Obviously, the forces are
unbalanced and pull the rim into a zig-zag.

rim
|
|
y
/
|
z
/
|
spoke ----A
\
\
B----- spoke
|
|
|
spoke -------C
\
\
D--- spoke
|
/
e
|
/
f
|
|
rim

The crude ASCII diagram doesn't show things well, but the point is
that the upper rim is pulled left, while the lower rim is pulled to
the right.

The rim distortion can be emphasized by squeezing just one pair of
spokes on one side (not two pairs) with 60 lbs of force.

Tension increases in both spokes in the squeezed pair, but the
increase is noticeably less than 55-65 lbs.

Tension also increases in one spoke of the unsqueezed pair, but again
the increase is noticeably less than 55-65 lbs.

Weirdly, the other spoke in the unsqueezed pair loses significant
tension.

If you start squeezing the unsqueezed pair, spoke tension slowly
rises.

Tension rises slowly for the two spokes in the already-squeezed pair,
faster for unsqueezed spoke that showed the initial rise in tension,
and even faster for the spoke that originally lost tension.

When both pairs are squeezed with 60 lbs of force, all four spokes
have gained about 55-65 lbs of tension.

Cheers,

Carl Fogel

On Mon, 17 Jul 2006 05:48:19 -0500, "Earl Bollinger"
wrote:

Well I probably missed out out all the discussions of this a while back. So
I don't have the whole story.
But although I can see how this is causing what you discuss below, I am
still sort of curious as to how this applies to me riding a bike along on
the wheels like this.
Maybe I need more coffee of something.
I do not see the correlation yet.
As I understand it, when you are riding along, the spokes on the bottom go
through a compression phase (getting more slack), with the spokes on top
going though a tension phase (getting more taut) at the same time.
I don't see anything causing a squeezing phase on the spokes when I ride
along.

Dear Earl,

When ridden, a wheel's pre-tensioned spokes lose tension as they roll
under the hub, just as you say. Theory and computer models predict
this, and strain gauges confirm it.

The wheels are designed for this.

Spoke-squeezing is an entirely different matter, a step often
recommended in the wheel-building process.

Sitting in front of the wheel at a truing stand, grab two pairs of
parallel spokes, one pair in each hand, and "vigorously" squeeze each
pair of spokes together. (Think of a harp player.)

The wheels are not designed for this, and it doesn't happen in any
remotely normal riding--it's purely a wheelbuilding procedure in which
spokes are squeezed to bend sideways toward each other.

A large number of posters claim that squeezing spokes like this has
beneficial results due solely to the increase in tension in the
spokes. (Other posters add that there are other benefits, such as
seating nipples and spoke elbows.)

My question was how much the tension in the squeezed spokes actually
rises. Curiously, there were no measurements, just offhand
calculations scattered through posts over many years.

Unfortunately, the calculations assumed that the rim-spoke-hub
structure worked like a giant makeshift tension gauge. If a known
force deflected the spoke sideways a known distance, and the details
of the spoke's elasticity were known, then the tension could be
calculated by some simple geometry.

That's how a spoke tension gauge works--two posts support the spoke,
which is bent sideways by a spring exerting a known force, and the
sideways deflection is measured by a dial gauge.

But a spoke tension gauge works only because its two supporting posts
are massively braced and are distorted in the frame by only a tiny
spring force. The geometrical calculation depends on the two posts
staying the same distance apart.

When spoke pairs are squeezed together, the force is massive (compared
to what a tension gauge uses), and one of the two "posts" (the rim)
hub) distorts wildly.

You can see this distortion by simply squeezing a parallel pair of
spokes lined up with the brake pads--the rim will move quite visibly.
Squeezing two spoke pairs, one on each side of the bike, complicates
things. The two inner spokes kinda-sorta balance each other in the
middle, but the two outer spokes are obviously pulling the rim into a
zig-zag.

So calculations made in good faith on a bad model invariably produced
very high tension increases. Perhaps because these results suited the
theory, they weren't questioned.

When I began squeezing spokes together with known forces and measuring
the tension changes with an actual spoke tension gauge, I found what I
described--squeeze two parallel spoke pairs together, each with 60 lbs
of force, one one each side of the bike, and the tension rises 55-65
lbs for each spoke. The rim design, hub flange, dishing, initial
tension, and so forth just didn't seem to matter.

Compare this measured 60 lb tension increase for a 60 lb squeeze to
the most recent calculation (made in good faith) that a 30 lb squeeze
would produce a 156 lb tension increase. So far, measurements show
only a 1-to-1 mechanical advantage, while calculations that went
unchallenged predicted a 5-to-1 mechanical advantage.

"So far" is the key point. I'm asking if anyone can measure a
significantly higher tension increase for a 60 lb squeeze on a
reasonably normal wheel.

I tested a number of wheels, since my earlier results were called
"impossible" or just the result of a "cheesy wheel" by engineers who
never actually measured any tension increases.

So I'm trying to find out if someone else has a wheel whose spokes
behave differently when squeezed. The wheels that I tested might
somehow be "cheesy," meaning the rims and hubs are unusually flexible,
or they might be built or laced in some strangely weak fashion, or my
testing could just be wildly inaccurate on wheel after wheel and spoke
after spoke over a wide range of squeeze forces and initial tensions.

I doubt that I'm wrong, but the wheels and spokes don't care what I
think, so I want to know what tension changes other people measure
when they squeeze spokes on other wheels.

Cheers,

Carl Fogel

Earl Bollinger wrote:
As I understand it, when you are riding along, the spokes on the bottom go
through a compression phase (getting more slack), with the spokes on top
going though a tension phase (getting more taut) at the same time.

Surprisingly, the spokes at the top don't show that much tension
increase, certainly nothing to match the loss of tension underneath the
hub. Also surprisingly, the spokes at 90o and 180o show a notable
increase in tension.

wrote:
On Mon, 17 Jul 2006 05:48:19 -0500, "Earl Bollinger"
wrote:

Well I probably missed out out all the discussions of this a while back. So
I don't have the whole story.
But although I can see how this is causing what you discuss below, I am
still sort of curious as to how this applies to me riding a bike along on
the wheels like this.
Maybe I need more coffee of something.
I do not see the correlation yet.
As I understand it, when you are riding along, the spokes on the bottom go
through a compression phase (getting more slack), with the spokes on top
going though a tension phase (getting more taut) at the same time.
I don't see anything causing a squeezing phase on the spokes when I ride
along.

Dear Earl,

When ridden, a wheel's pre-tensioned spokes lose tension as they roll
under the hub, just as you say. Theory and computer models predict
this, and strain gauges confirm it.

The wheels are designed for this.

Spoke-squeezing is an entirely different matter, a step often
recommended in the wheel-building process.

Sitting in front of the wheel at a truing stand, grab two pairs of
parallel spokes, one pair in each hand, and "vigorously" squeeze each
pair of spokes together. (Think of a harp player.)

The wheels are not designed for this, and it doesn't happen in any
remotely normal riding--it's purely a wheelbuilding procedure in which
spokes are squeezed to bend sideways toward each other.

A large number of posters claim that squeezing spokes like this has
beneficial results due solely to the increase in tension in the
spokes. (Other posters add that there are other benefits, such as
seating nipples and spoke elbows.)

My question was how much the tension in the squeezed spokes actually
rises. Curiously, there were no measurements, just offhand
calculations scattered through posts over many years.

Unfortunately, the calculations assumed that the rim-spoke-hub
structure worked like a giant makeshift tension gauge. If a known
force deflected the spoke sideways a known distance, and the details
of the spoke's elasticity were known, then the tension could be
calculated by some simple geometry.

That's how a spoke tension gauge works--two posts support the spoke,
which is bent sideways by a spring exerting a known force, and the
sideways deflection is measured by a dial gauge.

But a spoke tension gauge works only because its two supporting posts
are massively braced and are distorted in the frame by only a tiny
spring force. The geometrical calculation depends on the two posts
staying the same distance apart.

When spoke pairs are squeezed together, the force is massive (compared
to what a tension gauge uses), and one of the two "posts" (the rim)
hub) distorts wildly.

You can see this distortion by simply squeezing a parallel pair of
spokes lined up with the brake pads--the rim will move quite visibly.
Squeezing two spoke pairs, one on each side of the bike, complicates
things. The two inner spokes kinda-sorta balance each other in the
middle, but the two outer spokes are obviously pulling the rim into a
zig-zag.

So calculations made in good faith on a bad model invariably produced
very high tension increases. Perhaps because these results suited the
theory, they weren't questioned.

When I began squeezing spokes together with known forces and measuring
the tension changes with an actual spoke tension gauge, I found what I
described--squeeze two parallel spoke pairs together, each with 60 lbs
of force, one one each side of the bike, and the tension rises 55-65
lbs for each spoke. The rim design, hub flange, dishing, initial
tension, and so forth just didn't seem to matter.

Compare this measured 60 lb tension increase for a 60 lb squeeze to
the most recent calculation (made in good faith) that a 30 lb squeeze
would produce a 156 lb tension increase. So far, measurements show
only a 1-to-1 mechanical advantage, while calculations that went
unchallenged predicted a 5-to-1 mechanical advantage.

"So far" is the key point. I'm asking if anyone can measure a
significantly higher tension increase for a 60 lb squeeze on a
reasonably normal wheel.

I tested a number of wheels, since my earlier results were called
"impossible" or just the result of a "cheesy wheel" by engineers who
never actually measured any tension increases.

So I'm trying to find out if someone else has a wheel whose spokes
behave differently when squeezed. The wheels that I tested might
somehow be "cheesy," meaning the rims and hubs are unusually flexible,
or they might be built or laced in some strangely weak fashion, or my
testing could just be wildly inaccurate on wheel after wheel and spoke
after spoke over a wide range of squeeze forces and initial tensions.

I doubt that I'm wrong, but the wheels and spokes don't care what I
think, so I want to know what tension changes other people measure
when they squeeze spokes on other wheels.

Cheers,

Carl Fogel

carl, i've just done a rough and ready test and confirm that a ma3
rim/105 hub combo has spoke tension rise roughly in accordance with your
measurements, i.e. that tension increase ~= squeeze force. there
doesn't seem to be a geometric increase in tension as rigid rim theory
would predict.

On Mon, 17 Jul 2006 20:43:56 -0700, jim beam
wrote:

wrote:
On Mon, 17 Jul 2006 05:48:19 -0500, "Earl Bollinger"
wrote:

Well I probably missed out out all the discussions of this a while back. So
I don't have the whole story.
But although I can see how this is causing what you discuss below, I am
still sort of curious as to how this applies to me riding a bike along on
the wheels like this.
Maybe I need more coffee of something.
I do not see the correlation yet.
As I understand it, when you are riding along, the spokes on the bottom go
through a compression phase (getting more slack), with the spokes on top
going though a tension phase (getting more taut) at the same time.
I don't see anything causing a squeezing phase on the spokes when I ride
along.

Dear Earl,

When ridden, a wheel's pre-tensioned spokes lose tension as they roll
under the hub, just as you say. Theory and computer models predict
this, and strain gauges confirm it.

The wheels are designed for this.

Spoke-squeezing is an entirely different matter, a step often
recommended in the wheel-building process.

Sitting in front of the wheel at a truing stand, grab two pairs of
parallel spokes, one pair in each hand, and "vigorously" squeeze each
pair of spokes together. (Think of a harp player.)

The wheels are not designed for this, and it doesn't happen in any
remotely normal riding--it's purely a wheelbuilding procedure in which
spokes are squeezed to bend sideways toward each other.

A large number of posters claim that squeezing spokes like this has
beneficial results due solely to the increase in tension in the
spokes. (Other posters add that there are other benefits, such as
seating nipples and spoke elbows.)

My question was how much the tension in the squeezed spokes actually
rises. Curiously, there were no measurements, just offhand
calculations scattered through posts over many years.

Unfortunately, the calculations assumed that the rim-spoke-hub
structure worked like a giant makeshift tension gauge. If a known
force deflected the spoke sideways a known distance, and the details
of the spoke's elasticity were known, then the tension could be
calculated by some simple geometry.

That's how a spoke tension gauge works--two posts support the spoke,
which is bent sideways by a spring exerting a known force, and the
sideways deflection is measured by a dial gauge.

But a spoke tension gauge works only because its two supporting posts
are massively braced and are distorted in the frame by only a tiny
spring force. The geometrical calculation depends on the two posts
staying the same distance apart.

When spoke pairs are squeezed together, the force is massive (compared
to what a tension gauge uses), and one of the two "posts" (the rim)
hub) distorts wildly.

You can see this distortion by simply squeezing a parallel pair of
spokes lined up with the brake pads--the rim will move quite visibly.
Squeezing two spoke pairs, one on each side of the bike, complicates
things. The two inner spokes kinda-sorta balance each other in the
middle, but the two outer spokes are obviously pulling the rim into a
zig-zag.

So calculations made in good faith on a bad model invariably produced
very high tension increases. Perhaps because these results suited the
theory, they weren't questioned.

When I began squeezing spokes together with known forces and measuring
the tension changes with an actual spoke tension gauge, I found what I
described--squeeze two parallel spoke pairs together, each with 60 lbs
of force, one one each side of the bike, and the tension rises 55-65
lbs for each spoke. The rim design, hub flange, dishing, initial
tension, and so forth just didn't seem to matter.

Compare this measured 60 lb tension increase for a 60 lb squeeze to
the most recent calculation (made in good faith) that a 30 lb squeeze
would produce a 156 lb tension increase. So far, measurements show
only a 1-to-1 mechanical advantage, while calculations that went
unchallenged predicted a 5-to-1 mechanical advantage.

"So far" is the key point. I'm asking if anyone can measure a
significantly higher tension increase for a 60 lb squeeze on a
reasonably normal wheel.

I tested a number of wheels, since my earlier results were called
"impossible" or just the result of a "cheesy wheel" by engineers who
never actually measured any tension increases.

So I'm trying to find out if someone else has a wheel whose spokes
behave differently when squeezed. The wheels that I tested might
somehow be "cheesy," meaning the rims and hubs are unusually flexible,
or they might be built or laced in some strangely weak fashion, or my
testing could just be wildly inaccurate on wheel after wheel and spoke
after spoke over a wide range of squeeze forces and initial tensions.

I doubt that I'm wrong, but the wheels and spokes don't care what I
think, so I want to know what tension changes other people measure
when they squeeze spokes on other wheels.

Cheers,

Carl Fogel

carl, i've just done a rough and ready test and confirm that a ma3
rim/105 hub combo has spoke tension rise roughly in accordance with your
measurements, i.e. that tension increase ~= squeeze force. there
doesn't seem to be a geometric increase in tension as rigid rim theory
would predict.

Dear Jim,

Neither did a 32-spoke MA3 front wheel that I tested.

Again, anyone can see that the rim that is impressively rigid in
normal use is not rigid in opposition to spoke squeezing.

Simply squeeze a single pair of spokes with one hand and watch as that
section of the rim moves toward the brake pad.

Squeeze two pairs of spokes, and the ends of the 4-spoke section move
in opposite directions. The movement is less obvious, but suggests a
curve over twice as much of the rim.

Cheers,

Carl Fogel

wrote:
snip
Dear Jim,

Neither did a 32-spoke MA3 front wheel that I tested.

Again, anyone can see that the rim that is impressively rigid in
normal use is not rigid in opposition to spoke squeezing.

it's not that rigid in normal use. that's why we hear brake pads scrape
in wet weather - there's quite a lot of movement going on!


Simply squeeze a single pair of spokes with one hand and watch as that
section of the rim moves toward the brake pad.

Squeeze two pairs of spokes, and the ends of the 4-spoke section move
in opposite directions. The movement is less obvious, but suggests a
curve over twice as much of the rim.

Cheers,

Carl Fogel