Better than spoke squeezing?
carl...@comcast.net
with the axle on a wooden block and leaning on the wheel with one hand
on either side of the rim.
The Mavic site is hard to navigate, so here's a copy of the Mavic pdf,
which has a picture:
http://home.comcast.net/~carlfogel/download/SpokeTension.pdf
"Step 6 When there is no lateral or vertical movement, put the hub
axle of the wheel on a wooden block and push on opposite sides of the
rim with your hands. The spokes and nipples will now 'seat' in
position (stress relieving)."
The Mavic page indicates that this is first done at about 40-60 kgf
spoke tension on a wheel that will end up at 90-110 kgf, and repeated
as the tension is increased "until the wheel is perfect and the spokes
make no more cracking sounds."
Various posters commented, but no one had tested things, so . . .
I leaned heavily on a 36-spoke steel front wheel whose spokes had been
squeezed in the often-recommended fashion. The wheel made cracking
noises, probably spokes and nipples unwinding on the rim.
Another poster reported similar results.
I leaned on a bathroom scale, which indicated that at 195 lbs I can
put about 100 lbs onto a rim laid flat on a workbench.
I set up a test like a roulette wheel by bolting a plain no-box steel
MTB rim from a Fury RoadMaster with 36 2mm-straight spokes to a long
metal bar and screwing the bar onto some 2x4's for lateral stability.
To mimic leaning with my hands on either side of the flat rim, I laid
a 2x4 across the wheel, with a hole bored in the middle to clear the
end of the axle that was sticking up and some small 2x4 pieces nailed
under the board to lift it higher off the rim.
This rig made it easy to apply a Park tension gauge from underneath:
http://i15.tinypic.com/2jcdcnt.jpg
I measured spoke tension before and after I loaded the board and 100
lbs of weights onto the horizontal wheel. Details appear at the end of
this post.
Here's a graph that shows the symmetrical effect, with one line
showing the tension change for the 18 upper spokes around the wheel
and another line showing the tension change for the 18 lower spokes:
http://i10.tinypic.com/2njccc9.jpg
Tension rise for the two top-side spokes under the ends of the board
was around 85-95 lbs, a 50-55% increase, more than tension increases
measured with the familiar two-handed method.
A typical poster is unlikey to exceed 60 lbs of squeeze force with
either hand. No spokes on any wheels measured in previous tests with
60 lb squeeze forces ever gained more than 55~65 lbs in tension.
(The bend angles are impressive, but most of the bend is due to slack
gained as the four spokes distort the rim into a faint Z or N shape.)
Whether the higher tension available with the Mavic block-of-wood
technique is worthwhile is a separate matter. Tests involving heating
tensioned spoke bends and measuring how much the bend angles changed
suggest that practically all residual stresses are relieved before
tension rises to ordinary wheel-building levels.
Details of the block-of-wood measurements appear below.
Cheers,
Carl Fogel
A simple no-box steel MTB rim was used, with the tire deflated.
There were 36 straight 2mm-spokes. The rim was horizontal and free to
spin like roulette wheel.
A 2x4 was laid flat across rim, with one end on the 1st top spoke
counter-clockwise from the valve hole. Tension was measured before
andafter ~100 lbs of weights were loaded onto the board.
Spoke #36, a lower spoke, lost enormous tension and dropped to Park
mark 13, off Park scale, so it was extrapolated.
Figures are rounded in some columns
Park marks are estimated in tenths, giving a spurious air of accuracy,
but not as bad as estimating in quarters, which would give an even
more spurious impression of 2-decimal accuracy. The graph shows a
clear pattern.
Below, spokes are numbered 1-36, with #1 being a top spoke under one
end of the board, and spokes 17 and 19 being the top spokes under the
other end of the board.
no no
load load load load
| | | |
spoke Park Park | | change %
# mark mark lbs lbs lbs change
----- ----------- ------------ ---- ----
1top 21 24.5 167.2 250.8 83.6 50.0% board
2 23 19.4 209.0 141.7 -67.3 -32.2%
3t 21 23.5 167.2 222.2 55.0 32.9%
4 20.8 18.9 163.7 135.1 -28.6 -17.5%
5t 21.9 22.5 185.0 198.0 13.0 7.0%
6 20.4 20.9 156.6 165.4 8.8 5.6%
7t 21 20.1 167.2 151.4 -15.8 -9.5%
8 20.6 21.8 160.2 183.0 22.9 14.3%
9t 22.1 21.5 189.2 177.1 -12.1 -6.4%
10 22.4 23 195.8 209.0 13.2 6.7%
11t 18.8 17 133.8 112.2 -21.6 -16.1%
12 20.8 21.9 163.7 185.0 21.3 13.0%
13t 22.1 22 189.2 187.0 -2.2 -1.2%
14 21.1 20.5 169.2 158.4 -10.8 -6.4%
15t 21.5 22.9 177.1 206.8 29.7 16.8%
16 20.5 17 158.4 112.2 -46.2 -29.2%
17t 20.8 24 163.7 235.4 71.7 43.8% board
18 20.9 16.5 165.4 106.7 -58.7 -35.5% x
19t 21 24.9 167.2 263.1 95.9 57.4% board
20 22.5 19.2 198.0 139.0 -59.0 -29.8%
21t 19.4 21.5 141.7 177.1 35.4 25.0%
22 19.2 18.2 139.0 125.8 -13.2 -9.5%
23t 22.9 23.6 206.8 224.8 18.0 8.7%
24 20.2 20.7 153.1 161.9 8.8 5.7%
25t 22 21.1 187.0 169.2 -17.8 -9.5%
26 21.1 21.9 169.2 185.0 15.8 9.4%
27t 22.1 20 189.2 149.6 -39.6 -20.9%
28 20.1 21.1 151.4 169.2 17.8 11.8%
29t 21.9 19.3 185.0 140.4 -44.7 -24.1%
30 21.9 22 185.0 187.0 2.0 1.1%
31t 22 21.2 187.0 171.2 -15.8 -8.5%
32 20.9 21 165.4 167.2 1.8 1.1%
33t 22.5 23.4 198.0 219.6 21.6 10.9%
34 21.2 18.8 171.2 133.8 -37.4 -21.9%
35t 21.3 23.2 173.1 214.3 41.1 23.8%
36 19 13 136.4 72.6 -63.8 -46.8%
data ends
Ben C
> Mavic used to recommend stress-relieving a wheel by laying it flat
> with the axle on a wooden block and leaning on the wheel with one hand
> on either side of the rim.
> Whether the higher tension available with the Mavic block-of-wood
> technique is worthwhile is a separate matter. Tests involving heating
> tensioned spoke bends and measuring how much the bend angles changed
> suggest that practically all residual stresses are relieved before
> tension rises to ordinary wheel-building levels.
Thank you for another outstanding experiment.
The way I see it is that if you're getting cracking sounds, you are
probably doing something useful (pace Jobst who suggested it was as much
use as cracking your knuckles, and may be right), whether this is
relieving residual stresses, spoke windup, or seating the bends in the
hub flange.
I don't think Mavic made any claims about residual stresses per se,
they seem to be more of an r.b.t. theme.
carl...@comcast.net
Dear Ben,
It's confusing. Mavic doesn't mention residual stresses.
But Mavic does call leaning on the rim "stress relieving."
Except that they do it to "seat" the spokes and nipples.
"Step 6 When there is no lateral or vertical movement, put the hub
axle of the wheel on a wooden block and push on opposite sides of the
rim with your hands. The spokes and nipples will now 'seat' in
position (stress relieving)."
http://home.comcast.net/~carlfogel/download/SpokeTension.pdf
Both block-of-wood and spoke-squeezing increase spoke tension, which
on RBT leads to debates about how much residual stress can be relieved
where the spokes are bent or formed.
Both techniques also lead to cracking noises as the spokes and nipples
unwind.
As far as I know, there's no debate about whether eliminating wind-up
is possible and is a good thing. Wind-up is bad because it adds a
torsional stress and because eventually the spoke will unwind and
loosen, so it should be eliminated by backing off after turning the
nipple and by increasing and then releasing the spoke tension, which
may give a noise described as cracking, pinging, or popping.
Both techniques also seat the spokes and nipples in the hubs and rims.
Like spoke wind-up, there's no debate (as far as I know) about whether
seating things is a good idea.
Unike spoke-squeezing, the block-of-wood technique doesn't bend the
spokes at any noticeable angle, so it may not seat things as much. But
it does produce much greater tension decreases, so it will probably
produce more cracking as spokes unwind. It also produces greater
tension increases, whose value is open to question, but which are
presumably better than smaller tension increases.
Cheers,
Carl Fogel
Ben C
> On Tue, 14 Nov 2006 03:59:38 -0600, Ben C <spam...@spam.eggs> wrote:
> Both block-of-wood and spoke-squeezing increase spoke tension, which
> on RBT leads to debates about how much residual stress can be relieved
> where the spokes are bent or formed.
>
> Both techniques also lead to cracking noises as the spokes and nipples
> unwind.
I see, so cracking = unwinding. That explains why I got much less
cracking when I was working on a nice new rim with linseed oil on the
spokes and being careful to turn a bit too far and come back again.
> As far as I know, there's no debate about whether eliminating wind-up
> is possible and is a good thing. Wind-up is bad because it adds a
> torsional stress and because eventually the spoke will unwind and
> loosen, so it should be eliminated by backing off after turning the
> nipple and by increasing and then releasing the spoke tension, which
> may give a noise described as cracking, pinging, or popping.
>
> Both techniques also seat the spokes and nipples in the hubs and rims.
> Like spoke wind-up, there's no debate (as far as I know) about whether
> seating things is a good idea.
>
> Unike spoke-squeezing, the block-of-wood technique doesn't bend the
> spokes at any noticeable angle, so it may not seat things as much. But
> it does produce much greater tension decreases, so it will probably
> produce more cracking as spokes unwind. It also produces greater
> tension increases, whose value is open to question, but which are
> presumably better than smaller tension increases.
The tension increases presumably help to seat the spokes.
jobst....@stanfordalumni.org
>> Both block-of-wood and spoke-squeezing increase spoke tension,
>> which on RBT leads to debates about how much residual stress can be
>> relieved where the spokes are bent or formed.
>> Both techniques also lead to cracking noises as the spokes and
>> nipples unwind.
> I see, so cracking = unwinding. That explains why I got much less
> cracking when I was working on a nice new rim with linseed oil on
> the spokes and being careful to turn a bit too far and come back
> again.
If you didn't back off with the spoke wrench after adjusting tension,
there was windup and this windup gets released when pressing the dim
down with the hub supported. If you didn't hear as much noise, it may
be because linseed oil suppressed the noise that is classically a
metal to metal stick-slip phenomenon.
>> As far as I know, there's no debate about whether eliminating
>> wind-up is possible and is a good thing. Wind-up is bad because it
>> adds a torsional stress and because eventually the spoke will
>> unwind and loosen, so it should be eliminated by backing off after
>> turning the nipple and by increasing and then releasing the spoke
>> tension, which may give a noise described as cracking, pinging, or
>> popping.
>> Both techniques also seat the spokes and nipples in the hubs and
>> rims. Like spoke wind-up, there's no debate (as far as I know)
>> about whether seating things is a good idea.
>> Unlike spoke-squeezing, the block-of-wood technique doesn't bend the
>> spokes at any noticeable angle, so it may not seat things as much.
>> But it does produce much greater tension decreases, so it will
>> probably produce more cracking as spokes unwind. It also produces
>> greater tension increases, whose value is open to question, but
>> which are presumably better than smaller tension increases.
> The tension increases presumably help to seat the spokes.
Just because you didn't hear it, doesn't mean it wasn't there, unless
you backed off after final adjustments. DT Revolution spokes and most
flat spokes are so torsionally challenged that they require special
handling while turning spoke nipples. With those spokes it is
obvious, the revolution spokes just twisting as the thread no longer
advances and flat spokes becoming helical ornaments while doing so.
Jobst Brandt
jim beam
one more thing to look into - the difference in tension increase
depending on side for a dished rear...