Spoke tension dropped when tire deflated
carl...@comcast.net
Spoke tension is supposed to drop as the tire inflates and the cord
angle causes it to contract and squeeze the rim inward.
Dianne tested a 700c tire and posted her results, which showed that
spoke tension and inflation were inversely related:
http://www.geocities.com/dianne_1234/bikes/tension-inflation/tension-inflation.htm
Dianne tested, re-tested, provided pictures of her Park tension gauge,
and confirmed the theory that the angle of the tire cords causes the
tire to shorten and contract, squeezing the wheel inward and thus
reducing the spoke tension.
I trust Diane's measurements results.
But I've recently been playing with a 700x26 and a 26x1.75 tire on my
workbench, and that fat MTB tire was just hanging there, with my Park
spoke tension gauge handy, so I hung my tension gauge on a spoke and
let 60 psi of air out through the Schrader valve . . .
I expected the spoke tension to rise.
But the damn spoke tension slowly and visibly dropped.
I repeated the test a few times, but the spoke tension stubbornly went
the wrong way, slowly and visibly dropping as I let the air out of the
tire.
The 700x26 tire was sitting nearby at 120 psi, so I tried it.
Nothing.
I hung my Park tension gauge from a spoke, let all the air out through
the Presta valve, and my tension gauge just sat there--no visible rise
or fall on the spoke tension gauge.
This was getting ridiculous, so I put the 700c tire on a different
rim.
The tire hadn't shown any spoke tension change on my no box-section
rim, but maybe it would behave differently on a Mavic MA-3 box-section
rim.
I pumped the 700x26 up to 120 psi again on the box-section MA-3 rim,
hung my tension gauge on a spoke, and let the air out.
At last, the tension gauge moved very slightly in the "right"
direction, showing a spoke tension increase as air pressure dropped.
psi spoke according
tester tire rim design change tension to theory
------ ---- ---------- ------ ------- ---------
carl 26x1.75 no box steel 60..0 big rise wildly wrong
carl 700x26 no box alum 120..0 no change less wrong
carl 700x26 box ma3 alum 120..0 tiny rise barely right
dianne 700x25 aero alum 120..0 larger rise right
Dianne noted that "The rim properties may affect how much the tire
inflation affects spoke tension."
But I wasn't expecting that rim properties would reverse the effect.
The wider 1.75" tire at lower pressure on a wider 26" simple no-box
steel rim lost spoke tension as 60 psi was released.
The narrow 700 x 26 tire at higher pressure on a narrower simple
no-box aluminum rim didn't show any appreciable spoke tension change
as 120 psi was released.
The narrow 700 x 26 tire at higher pressure on a narrow MA-3 aluminum
box-section rim showed a tiny rise in spoke tension as 120 psi was
released.
And Dianne's impressively braced aero 700x25 showed a small but
obvious gain in tension as 120 psi was released.
The tire must constrict as inflates if its cords are at any remotely
normal angle, so deflating the tire should relax the tire
constriction, let the rim expand, and cause the spoke tension to rise.
Since the spoke tension doesn't seem to drop with deflation on
different rims, other forces or changes must be involved. The casing
tension or air pressure may deform a wider or simpler rim in some way
that overwhelms the expected spoke tension change.
Whatever the explanation is, it was startling to see the tension gauge
needle move the wrong way as I let the air out of the 26" tire.
Cheers,
Carl Fogel
jobst....@stanfordalumni.org
> What happens to spoke tension when a tire is inflated?
> Spoke tension is supposed to drop as the tire inflates and the cord
> angle causes it to contract and squeeze the rim inward.
> Dianne tested a 700c tire and posted her results, which showed that
> spoke tension and inflation were inversely related:
> Dianne tested, re-tested, provided pictures of her Park tension
> gauge, and confirmed the theory that the angle of the tire cords
> causes the tire to shorten and contract, squeezing the wheel inward
> and thus reducing the spoke tension.
> I trust Diane's measurements results.
Appending this gratuitous claim doesn't help disperse doubt.
Even if you use your musical ear to sense the difference in tone. I
haven't invented this from mental models, but measured it. I have a
tensiometer and measured the difference. There was less spoke tension
when the tire is inflated than when flat... for both a tubular and
clincher.
I believe what you see is measurement inaccuracy involving. The
instrument shown cannot be zeroed in situ so that not measuring at
exactly the same location can give different readings.
> Nothing.
I suspect your instrument has friction losses that obscure what it is
trying to measure. The larger the rim cross section, the smaller the
difference between inflated and flat. That may be part of your
problem. You might apply the equation from "the Bicycle Wheel" to
assess the constriction force for your tire size and inflation
pressures and see that the forces are small.
Jobst Brandt
carl...@comcast.net
>Carl Fogel writes:
>
>> What happens to spoke tension when a tire is inflated?
>
>> Spoke tension is supposed to drop as the tire inflates and the cord
>> angle causes it to contract and squeeze the rim inward.
>
>> Dianne tested a 700c tire and posted her results, which showed that
>> spoke tension and inflation were inversely related:
>
> http://tinyurl.com/67gwc
>
>> Dianne tested, re-tested, provided pictures of her Park tension
>> gauge, and confirmed the theory that the angle of the tire cords
>> causes the tire to shorten and contract, squeezing the wheel inward
>> and thus reducing the spoke tension.
>
>> I trust Diane's measurements results.
>
>Appending this gratuitous claim doesn't help disperse doubt.
>
>Even if you use your musical ear to sense the difference in tone. I
>haven't invented this from mental models, but measured it. I have a
>tensiometer and measured the difference. There was less spoke tension
>when the tire is inflated than when flat... for both a tubular and
>clincher.
>
>I believe what you see is measurement inaccuracy involving. The
>instrument shown cannot be zeroed in situ so that not measuring at
>exactly the same location can give different readings.
[snip]
Dear Jobst,
I hung the tension gauge on the spoke.
Then I watched the tension gauge needle drop as I pressed on Schrader
valve and let the air out. It moved the wrong way.
Neither zeroing nor friction apply in the slightest.
The question is not whether it's a tubular or a clincher. All tires
were clinchers. The question is what practical explanation is there
for the "wrong" result.
Dianne sensibly mentioned the possibility that different rims might
affect the result.
You just rode off in all directions without reading what I wrote.
Cheers,
Carl Fogel
jobst....@stanfordalumni.org
>>> What happens to spoke tension when a tire is inflated?
>>> Spoke tension is supposed to drop as the tire inflates and the cord
>>> angle causes it to contract and squeeze the rim inward.
>>> Dianne tested a 700c tire and posted her results, which showed that
>>> spoke tension and inflation were inversely related:
>>> Dianne tested, re-tested, provided pictures of her Park tension
>>> gauge, and confirmed the theory that the angle of the tire cords
>>> causes the tire to shorten and contract, squeezing the wheel inward
>>> and thus reducing the spoke tension.
>>> I trust Diane's measurements results.
>> Appending this gratuitous claim doesn't help disperse doubt.
>> Even if you use your musical ear to sense the difference in tone. I
>> haven't invented this from mental models, but measured it. I have a
>> tensiometer and measured the difference. There was less spoke tension
>> when the tire is inflated than when flat... for both a tubular and
>> clincher.
>> I believe what you see is measurement inaccuracy involving. The
>> instrument shown cannot be zeroed in situ so that not measuring at
>> exactly the same location can give different readings.
>>> It's raining, so I had some fun with a tire and my camera on a stand.
>>> I used a fine-point red sharpy pen to cross-hatch the sidewall at
>>> the bead on a Kevlar-bead 700x26 tire. Then I took 4 pictures at 1
>>> psi, 40 psi, 85 psi, 120 psi, and back down around 15 psi.
>>> It looks as if the middle /\ of the lowest cross-hatching is
>>> creeping out into sight as pressure increases, and creeping back
>>> out of sight when pressure is reduced.
>>> Nothing.
> I hung the tension gauge on the spoke.
> Then I watched the tension gauge needle drop as I pressed on Schrader
> valve and let the air out. It moved the wrong way.
> Neither zeroing nor friction apply in the slightest.
> The question is not whether it's a tubular or a clincher. All tires
> were clinchers. The question is what practical explanation is there
> for the "wrong" result.
> Dianne sensibly mentioned the possibility that different rims might
> affect the result.
> You just rode off in all directions without reading what I wrote.
Sorry, I did not see that you left the tensiometer in place when
changing pressure. It is possible to interpret that as a before an
after measurement, which I did. Now that I understand that, I am at a
loss to explain it. What sort of tire is the one in the pictures and
on what kind of rim? Does it have a hooked bead? I wasn't aware that
this is a Kevlar bead tire either.
Jobst Brandt
mike.a...@gmail.com
through the right angle, and lifting at the spoke holes, lifting at the
spoke holes. Can you measure the width inflated?
carl...@comcast.net wrote:
<deleted>
> The wider 1.75" tire at lower pressure on a wider 26" simple no-box
> steel rim lost spoke tension as 60 psi was released.
<deleted>
>
> Carl Fogel
carl...@comcast.net
Dear Jobst,
And I'm sorry, too--we tend to get snippy with each other.
In this case, getting snippy had the happy result of making me try out
the movie feature on my camera, since it's still raining.
Here's a 24-second 3mb video showing my Park tension gauge needle
dropping from over 24 to under 24 (less tension) as 60 psi of air
hisses out of a 26 x 1.75 mtb clincher tire on a no-box-section steel
rim:
http://www.youtube.com/v/NeG5o3WX7Mw
The inflated tire's constriction should be relaxing, so I expected
spoke tension to rise, but the opposite effect is easily observed.
The video is shaky because I'm pushing the Schrader valve (you can
hear the air escaping on the audio) and that jiggles the tire. Like
most such movies, it's clearer in a smaller window.
The tire is a $10 WalMart 26x1.75 semi-slick steel-bead clincher that
I bought to see how much a slick would speed up the originally knobby
Fury RoadMaster. The rim is the Fury RoadMaster's no-box steel rim.
(My 700x26 tire does have a Kevlar bead, which is the one that you
probably saw in the other pictures in the other thread about the bead
possibly expanding with pressure.)
Anyway, I was startled when my generic 26x1.75 tire and rim behaved
differently than the tire and rim that Dianne tested.
Probably the vee of the rim was spread by inflation and is relaxing so
much as the tire deflates that its effect (rim moves closer to hub)
overwhelms the expected effect of the tire's constriction relaxing
(rim expands).
I suspect that the two effects vary according to:
a) the width of the rim (more pressure forcing the vee outward and
increasing spoke tension)
b) the design of the rim and its material (my no-box rim should spread
more than a box rim, which should spread more than Dianne's aero-box
rim, but my rim confuses things by being made of steel, which should
spread less)
c) the constriction (how hard a narrow tire at 120 psi constricts
versus how hard a wide tire at 120 psi constricts)
d) the original spoke tension (which would resist the pressure
widening the rim into a broader vee)
e) how stretchy the spokes are (my 26" has 2.0 plated steel spokes;
surprisingly, Dianne's wheel used 2.0 spokes, too, but I expect that
they were stainless steel)
The opposing forces would explain why I could see so many different
effects--a visible tension drop with my 26x1.75 60 psi no-box, no
apparent tension change with my 700x26 120 psi no-box, a very tiny
rise in tension with an MA-3 box 120 psi, and Dianne's noticeable
tension rise with her aero 700x25 120 psi example.
For a change, there may be a practical effect in our usually esoteric
observations.
Dianne found that "inflating the tire to 120 psi dropped this spoke's
tension from about 99 kgf to about 83 kgf, a decrease of about 16 kgf
or roughly 35 pounds." That's a 17% loss of spoke tension.
But on my annoying 26x1.75, inflating the tire must increase the spoke
tension roughly a similar amount.
That is, if Dianne and I both tension our spokes to 100 kgf in a
stand, hers will drop to 83 kgf when she inflates her 700c tire to 120
psi on her aero-box rim, while mine should rise to around 115-120 kgf
when I inflate my 1.75" tire to 60 psi on my steel no-box rim.
Cheers,
Carl Fogel
Phil Holman
<carl...@comcast.net> wrote in message
news:nd7ji2588hi4maekl...@4ax.com...
Phil H
carl...@comcast.net
<mike.a...@gmail.com> wrote:
Dear Mike
D'oh!
I only thought about measuring the wheel diameter, which is far too
big for my calipers. Your suggestion is far more sensible.
And yes, that looks like the explanation.
raw
rim
widths
26" 700x26 700x26
no-box no-box box ma-3
psi steel aluminum aluminum
---- ------ -------- --------
0 1.082" 0.901" 0.085"
60 1.118" 0.907" 0.086"
120 n/a 0/912" 0.088"
total
rim
spread
26" 700x26 700x26
no-box no-box box ma-3
psi steel aluminum aluminum
---- ------ -------- --------
0 x x x
60 0.036" 0.006" 0.001"
120 n/a 0.011" 0.003"
At the same 60 psi, the wide 26" no-box steel rim spread 6 times as
much as the narrow 700c no-box aluminum rim, and 36 times as much as
the 700c boxed aluminum MA-3 rim.
Of course, no one just pumping up the mtb tire would be likely to
notice the roughly 1/32nd of an inch widening of the steel rim.
Cheers,
Carl Fogel
carl...@comcast.net
>Carl, did you try it on several different (orthogonal) spokes?
>
>Phil H
Dear Phil,
Maybe.
I didn't bother picking a particular spoke, but must have used
different ones between the original test and the fuss setting up to
make the movie. (The original director's cut was the first time that I
tried the movie feature. It had really great focus and lurid footage
of the needle dropping, but it turned out to be 19 mb, too large for
casual viewing. What you see at YouTube is my second effort.)
Anyway, I think the point is moot.
Mike suggested measuring the rim widths before and after inflation,
which I just posted.
Boy, that steel rim suffers from middle-psi spread!
Cheers,
Carl Fogel
jobst....@stanfordalumni.org
> Here's a 24-second 3mb video showing my Park tension gauge needle
> dropping from over 24 to under 24 (less tension) as 60 psi of air
> hisses out of a 26 x 1.75 mtb clincher tire on a no-box-section steel
> rim:
> http://www.youtube.com/v/NeG5o3WX7Mw
I don't don't your findings, what I wonder is what you are discovering
and why that is so. For conventional rims be that for tubulars or
clinchers, the effect is as described in the constriction formula. I
suspect that there are rim and tire configurations where one effect
overrides the other and that is what you are seeing.
What is the upshot of what you found?
Jobst Brandt
jobst....@stanfordalumni.org
> Here's a 24-second 3mb video showing my Park tension gauge needle
> dropping from over 24 to under 24 (less tension) as 60 psi of air
> hisses out of a 26 x 1.75 mtb clincher tire on a no-box-section steel
> rim:
> http://www.youtube.com/v/NeG5o3WX7Mw
I don't doubt your findings, what I wonder is what you are discovering
carl...@comcast.net
Dear Jobst,
I think that Mike's suggestion to measure the rim widths before and
after inflation explains things.
See my reply to him elsewhere in this thread, but basically the wider
steel no-box rim widens 6 to 36 times as much as the narrower aluminum
no-box and box rims when the tires are inflated to 60 psi.
The width, shape, and material must combine to make the steel rim
expand due to pressure on the sidewalls, pulling the spoke-line inward
and tightening the spokes, apparently much more than the tire
constriction from the same pressure is compressing the rim and
loosening the spokes.
Cheers,
Carl Fogel
Alex
carl...@comcast.net wrote:
> But I've recently been playing with a 700x26 and a 26x1.75 tire on my
> workbench, and that fat MTB tire was just hanging there, with my Park
> spoke tension gauge handy, so I hung my tension gauge on a spoke and
> let 60 psi of air out through the Schrader valve . . .
>
> I expected the spoke tension to rise.
>
> But the damn spoke tension slowly and visibly dropped.
I would try the experiment again but remove the tension gauge after
each measurement,
don't just let it hang on the spoke.
----------------
Alex
carl...@comcast.net
Dear Alex,
Basically, the Park tension gauge is just a spring clamp.
One end of the clamp is conveniently marked to show whether it is
being pushed out by more tension on the spoke that it's trying to
bend, or closing as spoke tension reduces.
If anything, leaving the gauge in place is more accurate--any movement
is clearly the result of spoke tension change, not friction, zeroing,
or operator error.
In any case, Mike's suggestion about measuring the rim widths before
and after inflation seems to explain things, as you can see elsewhere
in this thread. The wide no-box steel 26" rim expands 0.036" at 60
psi, far more than the narrower rims.
The widening of the rim-vee must pull the spoke-line away from the hub
and increase the spoke tension. This apparently overwhelms the
counter-force of the cords causing the tire to constrict on the rim
and reduce spoke tension.
Cheers,
Carl Fogel
Joe Riel
> In any case, Mike's suggestion about measuring the rim widths before
> and after inflation seems to explain things, as you can see elsewhere
> in this thread. The wide no-box steel 26" rim expands 0.036" at 60
> psi, far more than the narrower rims.
>
> The widening of the rim-vee must pull the spoke-line away from the hub
> and increase the spoke tension. This apparently overwhelms the
> counter-force of the cords causing the tire to constrict on the rim
> and reduce spoke tension.
I'm confused. If the rim is widening, the spoke holes move laterally
closer to the corresponding hub flanges (left holes move toward left
flange, and vice-versa). That should decrease the spoke tension. Is
the wheel built "backwards"? That is, do the spokes from the left
holes go to the right flange?
--
Joe Riel
carl...@comcast.net
Dear Joe,
I think that two things happen when the tire is inflated.
The angled cords of the tire try to expand the tire's minor diameter
and shorten the tire's major diameter, pulling tight around the rim
the long way and compressing inward toward the hub, which should
reduce spoke tension.
But at the same time the expanding tire's minor diameter is trying to
force the vee of the rim apart, which moves the spoke center-line
outward, away from the hub, which should raise spoke tension.
In a narrow, well-braced box-section rim like the one that Dianne
tested, there's so little flattening of the rim that the major effect
is what we expect--the tire tightens the long way around the rim like
a belt and reduces spoke tension,
But in a wide, no-box rim like the 26" that I tested, the vee of the
rim flattens out so much that its pull on the spokes becomes the major
effect and the damned spokes gain tension when the tire is inflated
(and lose tension quite noticeably when the tire is deflated, which
was what startled me).
unexpected expected
inflate=more tension inflate=less tension
Carl Dianne
uniflated inflated wide rim inflated narrow
widens a lot well-braced box rim
constricts a bit mostly constricts
------------------------------------------------------
\ /
\ / ' . . ' \ /
\ / ' . ' \____ /
\ / 1 spoke \ /
1 2 stretches \ /
2 3 1 spoke
3 4 2 contracts
hub hub hub
A narrow, well-braced rim that scarcely flattens at all will show the
expected tension decrease as the tire inflates and constricts. My MA-3
showed the expected loss of tension with inflation that Dianne found
with a similar or even better braced rim. The tension gauge needle
rose as the tire deflated and the constriction eased off.
A narrow, simple rim may show almost no change. My narrow no-box rim
widened and flattened more than an MA-3 at 60 psi, but the two effects
pretty much canceled each other out. I couldn't see any obvious
movement on the tension gauge when I let the air out.
A wide, simple rim like the 26" that startled me just plain goes the
"wrong" way. When I let the air out, the needle on the tension gauge
went the other way. The vee of the rim was contracting 0.036",
narrowing and moving the nipple inward toward the tire so much that
the loss of constriction was overwhelmed.
Cheers,
Carl Fogel
Joe Riel
--
Joe Riel
jobst....@stanfordalumni.org
I don't see any advantage in that method, it adding a source of error
because the gage is not capable of being zeroed in situ. In that way
it can introduce changes in its readings that do not come from spoke
tension.
Jobst Brandt
scotta...@yahoo.com
I think friction does apply. Small changes in tension could be hidden
by friction within the Park tensionometer. The needle will only be
able to move once the friction internal to the device is overcome by
the change in the spoke tension.
I don't see how one could determine the significance of friction
without performing some controlled testing of the tensionometer.
regards,
Scott Sawyer
carl...@comcast.net
<scotta...@yahoo.com> wrote:
Dear Scott,
How would friction cause the needle to drop instead of rise?
The tension gauge is in place. The needle will not move unless the
central hammer to which it is attached moves inward (spoke bends more,
less tension) or outward (spoke straightens, more tension) against the
steady spring tension pushing the hammer into the spoke
central hammer
pushes down, bending spoke downward
pivot V
x |
x V
x |
x V
_____ x | _________________spoke
O ' . H . ' O
' x
fixed x fixed
post x post
x
x
x
x
x end of needle
deflating tire moved needle downward
spoke bent further downward
same force on central hammer
movement is large for a tension gauge
But the proof is in the pudding, so . . .
I checked a spoke's tension with the Park gauge while the 26" tire was
inflated to 60 psi on the no-box wide steel rim. The needle sat at 23.
I deflated the tire and watched the spoke tension drop to just over 22
(less tension).
Then I removed the gauge, put it back on, and checked the tension
again. When removed and re-checked, the tension gauge needle showed
about 21 & a quarter, even further than the apparent drop when the
gauge was in place. (Writing it as "a quarter" avoids the spurious
accuracy implied by 21.25.)
So when the gauge is left in place and the spoke tension changes, the
in-place friction only minimizes the movement of the needle. Friction
doesn't cause the needle to move the wrong way.
Again, remember that this unexpected drop in spoke tension with tire
deflation is only with the wide no-box steel rim and wide tire.
Narrower rims, box bracing, and narrower tires end up behaving in the
opposite and expected fashion that Dianne measured, namely spoke
tension rising as a 120 psi tire is deflated.
Cheers,
Carl Fogel
jobst....@stanfordalumni.org
> Again, remember that this unexpected drop in spoke tension with tire
> deflation is only with the wide no-box steel rim and wide tire.
> Narrower rims, box bracing, and narrower tires end up behaving in
> the opposite and expected fashion that Dianne measured, namely spoke
> tension rising as a 120 psi tire is deflated.
Thee is an inconsistency that you may not have intended. The two
experiments above, yours and Dianne's seem to both have increasing
tension with decreasing inflation pressrue... either that or I'm
getting my negations confused. Help! Clarify!
Jobst Brandt
carl...@comcast.net
Dear Jobst,
Blame my confusing scribbling, plus the fact that you can test for the
expected effect either way.
If you inflate a tire, the tire constricting around the rim should
cause spoke tension to drop.
If you deflate the same tire, the tire relaxing around the rim should
cause spoke tension to rise.
Dianne inflated a tire to 120 psi and carefully photographed the
tension gauge showing the expected spoke tension drop as the inflating
tire constricted the rim.
I happened to think of her demonstration while fiddling with other
matters, noticed an inflated tire nearby, and then wondered like
Macbeth if that was a tension gauge I saw before me, its handle toward
my hand.
Come, let me clutch thee, I said to the tension gauge.
I slapped the tension gauge on a handy spoke, pushed the Schrader
valve in, and eyed the needle, expecting to see Dianne's results in
reverse.
Tire deflation was supposed to make the tire relax its python-like
grip on the rim. Freed of the tire constriction, the rim was supposed
to expand. The expanding rim was supposed to stretch the spokes
outward and cause spoke tension to rise.
But the damn tension gauge needle dropped instead of rising. It went
slowly and steadily moved the wrong way as the air hissed out of the
tire, showing a drop in spoke tension.
I pumped the tire up and deflated it a few times to confirm the
obvious movement, checked Dianne's page to make sure that I wasn't
just confused, and wondered why I couldn't get even the simplest
demonstration to work.
Inflating or deflating, the tire pressure is supposed to be inversely
related to spoke tension--if the only force involved is the inflating
tire constricting the rim.
My explanation is that the tire pressure must be working at the same
time in another direction.
The tire expanding sideways flattens the vee of the rim. As the rim
bows outward, the spoke centerline is drawn away from the hub,
tightening the spokes--more air pressure, more tension.
If the vee of the rim changes dramatically, then its effect can
overwhelm the expected effect of the inflating tire constricting the
long way around the rim.
At Mike's suggestion, I measured the rim widths before and after
inflation. Sure enough, the misbehaving rim widened a 0.036" at 60
psi, a massive increase compared to two narrow 700c rims, which
widened only 0.006" and 0.001".
Widening that much must have drawn the misbehaving rim's spoke
centerline up so far that it not only canceled out, but actually
overwhelmed the expected tension decrease when the tire was inflated.
Deflating is just a faster and more dramatic method of getting the
tension gauge to show the effect. You push the valve in and watch the
needle move over a few seconds while the tire and tension gauge are
sitting there, nice and steady. Pumping the tire up the other way by
hand wouldn't let you see the gauge actually moving, since hand
pumping is so slow and you can't keep your eye on the scarcely moving
tension gauge needle.
Cheers,
Carl Fogel
dvt
> On Mon, 09 Oct 2006 18:14:45 GMT, Joe Riel <jo...@k-online.com> wrote:
>
>
>>carl...@comcast.net writes:
>>
>>
>>>In any case, Mike's suggestion about measuring the rim widths before
>>>and after inflation seems to explain things, as you can see elsewhere
>>>in this thread. The wide no-box steel 26" rim expands 0.036" at 60
>>>psi, far more than the narrower rims.
>>>
>>>The widening of the rim-vee must pull the spoke-line away from the hub
>>>and increase the spoke tension. This apparently overwhelms the
>>>counter-force of the cords causing the tire to constrict on the rim
>>>and reduce spoke tension.
>>
>>I'm confused. If the rim is widening, the spoke holes move laterally
>>closer to the corresponding hub flanges (left holes move toward left
>>flange, and vice-versa). That should decrease the spoke tension. Is
>>the wheel built "backwards"? That is, do the spokes from the left
>>holes go to the right flange?
>
>
> Dear Joe,
> But at the same time the expanding tire's minor diameter is trying to
> force the vee of the rim apart, which moves the spoke center-line
> outward, away from the hub, which should raise spoke tension.
I think that's true for a non-box rim.
> In a narrow, well-braced box-section rim like the one that Dianne
> tested, there's so little flattening of the rim that the major effect
> is what we expect--the tire tightens the long way around the rim like
> a belt and reduces spoke tension,
I think the boxed rim might have the opposite effect, depending on where
the spoke is supported. Here are a few crude ASCII diagrams to
demonstrate...
Channel section rim, neutral position:
| |
| |
| |
| |
| |
---------------------
Channel section rim, pressurized:
| <----pressure----> |
| |
| |
| |
| ----- |
------ -------
Box section rim, neutral position:
| |
| |
| |
| |
| |
---------------------
| |
| |
| |
---------------------
Box section rim, pressurized:
| <----pressure----> |
| |
| |
| |
| --- |
------- --------
| |
| |
| |
------- --------
---
Sorry, I'm not very accomplished with ASCII art. Or any art form, for
that matter. But I think that the channel section rim will flex in a way
that tightens the spokes. In a box section rim, one of two scenarios is
possible. If the spoke rides on the spoke bed (nearest the tyre), the
box section rim will likely pull tighten the spoke, but to a lesser
degree than the channel section rim.
But if the spoke is supported by the inner wall of the box (nearest the
hub), I think the spoke actually loses tension. I would expect the inner
wall to bow towards the hub.
If the spoke is supported by both inner and outer walls via socket and
eyelet, I don't know which effect would win. And with "aero" section
rims like dianne_1234 tested, I'm not willing to hazard a guess which
way things will move.
In summary, I think the spoke tension change with tyre pressure may have
several variables that you didn't anticipate. You would need someone
with a good model to figure out all of the effects.
--
Dave
dvt at psu dot edu
scotta...@yahoo.com
Carl,
We're having a small misunderstanding. I agree with everything you have
written with the exception of your comment on the insignificance of
friction. I am not saying friction will "cause the needle to drop
instead of rise." I am saying the needle cannot move until the change
in tension is sufficient to over come the friction internal to the
device. That is small changes in tension may give no movement of the
needle.
Specifically, I was wondering if some of the "no change" data points
fall into this catagory (ie the change in tension was not large enough
to overcome the effect of friction).
Regards,
Scott
carl...@comcast.net
<scotta...@yahoo.com> wrote:
Dear Scott,
Yes, I agree that friction could cause the needle not to move when the
tension gauge is just hanging, or not as far.
When measuring actual tension (as opposed just seeing which way the
needle moves in this odd case), it's best to squeeze and release a
Park-style gauge at least half a dozen times and see where the needle
ends up.
A further practical problem is raised by spoke squeezing.
Powerful squeezing can leave faint but noticeable bends in a spoke.
This can distort a Park-style gauge and possibly even goof up a DT
style gauge--there's more resistance pressing against the bend, less
pressing with the bend, and roughly neutral from the side:
-->( versus -->)
I've seen strikingly different readings from a Park gauge attached
first one way to a faintly bent spoke and then the other:
| |
|gauge versus gauge|
| |
Like most theoretically simple measurements, spoke tension can be
beastly trick. People on diets often learn the sneaky trick of
standing on the bathroom spring-scale with their feet as far apart as
possible to shave a bit off the result.
Cheers,
Carl Fogel
Dan
As I thought about this problem, two things occur to me. Most spoke
tensiometers don't look like very precise instruments (I don't have
one) and the method of measurement has influence on the result since it
deforms the structure.
Intuition tells me that increasing tire pressure would increase spoke
tension since the tire and tube want to increase in all dimensions
including diameter as pressure goes up. This may or may not be so.
So I decided to try the experiment using a different method. I tried to
measure spoke strain directly using digital calipers. I opened the
calipers to 100mm, locked them and clamped the tips to a spoke using
little clamps. I released the caliper lock, zeroed the calipers and let
the air out of the tire. The change was below detection and I conclude
that the change in tension is so small so as to be insignificant.
Tire 700x23c Continental folding clincher
Pressure 120 psi
Rim Mavic Open Pro
Low flange hub
Spoke double butted 1.7mm
Caliper reads to 0.01mm
Minimum force detectable: (strain) * (modulus) * (area)
(.01/100) * (29,000,000) * (pi (1.7/25.4)^2 /4) = 10 lbs
Feeling disappointed with the result, I tried it again with the same
result.
27-1/4 clincher
90 psi
Single wall Weinmann rim - presumably more flexible.
High flange hub
Double butted 1.6
Thinking that my method was fatally flawed, I put the caliper back on
the first wheel (after reinflating it) and with the instrumented spoke
at the contact patch, I pressed down on the top if the wheel while
standing on my head to see the reading. I found I could easily generate
-0.03mm or about 30 lbs.
Dan
carl...@comcast.net
Dear Dan,
Intuition is often mistaken.
In this case, the inflation force distorts the rim in at least two
ways.
First, the plies of the tire are angled to cause the tire to shorten
with considerable force as it expands.
A Chinese finger-trap toy illustrates how an expanding tube with
angled plies must shorten. To pull your fingers out of the braided
tube, you have to push your fingers together. The tube expands,
letting go of your fingers, as you force it to shorten.
The tire plies are angled to do the same thing. Dianne's testing
showed how powerful the constriction of the tire on the rim can be. A
spoke tensioned to 100 kgf on her well-braced, narrow aero rim dropped
to 83 kgf when she inflated the tire from 0 to 120 psi.
The accuracy of the Park gauge that Dianne used is open to
question--the real tension could have been 105 kgf, not 100 kgf, and
it could have dropped to only 93 kgf, not 83 kgf, and so forth. But
the drop was easily visible.
When I tried to duplicate Dianne's result on the spur of the moment, I
had an inflated MTB tire on a wide, unbraced rim. I thought it would
be quick and amusing to hang a tension gauge on a spoke, let the air
out of the tire, and watch the tension gauge needle rise.
Instead, the needle slowly and smoothly dropped, heading the wrong
way. Again, the absolute tension and amount of change might be
somewhat inaccurate, but there was no question about the direction and
significance--anyone could see the stupid needle moving the wrong way.
The reason for my needle moving stubbornly in the "wrong" direction
was that the vee-shape my wider unbraced rim was being flattened by
the extra leverage of the inflating tire against the wider rim,
coupled with the lack of bracing to resist it. Measurements showed
that at 60 psi, the wide MTB rim spread out a whopping 0.036", while a
well-braced, narrow 700c rim (similar to what Dianne tested) spread
out only 0.001"--no significant flattening.
When inflation flattens and widens a vee-shaped rim, the spokes
around the center-line of the rim are pulled away from the hub,
increasing the tension. This effect on my MTB rim overwhelmed the
constriction effect.
As for measuring tension by the narrowing of the stretching spoke, I
suspect that the delicacy of the measurement is ill-suited to the
gross situation. Surface grime and irrefularities on the spoke can
amount to a thousandth of an inch. It will be difficult to hold the
calipers steady against the spoke as the wheel moves slightly, and
even a tiny angle change can amount to several thousandths of an inch.
My suspicions, however, are no more than "intuition," so I just tested
things in the basement.
My dial calipers show that a MTB tire spoke next to the valve is
0.078" at 60 psi, with the calipers at an easily repeatable spot and
angle.
The spoke is still 0.078" after I let the air out of the tire.
And the spoke is still 0.078" after I unscrew it completely.
I suspect that your 0.03 mm / 0.001" change was an artifact of the
testing procedure and the limits of the caliper's accuracy.
You can check whether this is the case by measuring a spoke in a wheel
and then measuring it again after you unscrew the spoke, releasing all
tension. Do your calipers show a much larger change in spoke width
than 0.03 mm when the tension changes 200 pounds?
Cheers,
Carl Fogel
Dan
carl...@comcast.net wrote:
>
> My dial calipers show that a MTB tire spoke next to the valve is
> 0.078" at 60 psi, with the calipers at an easily repeatable spot and
> angle.
>
> The spoke is still 0.078" after I let the air out of the tire.
>
> And the spoke is still 0.078" after I unscrew it completely.
>
> I suspect that your 0.03 mm / 0.001" change was an artifact of the
> testing procedure and the limits of the caliper's accuracy.
>
> You can check whether this is the case by measuring a spoke in a wheel
> and then measuring it again after you unscrew the spoke, releasing all
> tension. Do your calipers show a much larger change in spoke width
> than 0.03 mm when the tension changes 200 pounds?
>
> Cheers,
>
> Carl Fogel
Carl-
I think you may have misinterpreted my post. I was not measuring the
change in spoke diameter (Poisson's effect) but the change in length.
I opened the caliper to 100mm (~4") and attached the tips to a spoke
with two small clamps thus creating an axial strain gauge.
carl...@comcast.net
Dear Dan,
You're right.
I just didn't read your post with the attention that it
deserved--sorry.
I thought that you were just opening the calipers to 100 mm to run
their tips in and out and confirm that the digital readout was
working.
After my bike ride, I'll have to look into measuring things that way
and see what I get.
Cheers,
Carl Fogel
jobst....@stanfordalumni.org
>>> My dial calipers show that a MTB tire spoke next to the valve is
>>> 0.078" at 60 psi, with the calipers at an easily repeatable spot and
>>> angle.
>>> The spoke is still 0.078" after I let the air out of the tire.
>>> And the spoke is still 0.078" after I unscrew it completely.
>>> I suspect that your 0.03 mm / 0.001" change was an artifact of the
>>> testing procedure and the limits of the caliper's accuracy.
>>> You can check whether this is the case by measuring a spoke in a
>>> wheel and then measuring it again after you unscrew the spoke,
>>> releasing all tension. Do your calipers show a much larger change
>>> in spoke width than 0.03 mm when the tension changes 200 pounds?
>> I think you may have misinterpreted my post. I was not measuring
>> the change in spoke diameter (Poisson's effect) but the change in
>> length. I opened the caliper to 100mm (~4") and attached the tips
>> to a spoke with two small clamps thus creating an axial strain
>> gauge.
> I just didn't read your post with the attention that it deserved
> -sorry.
> I thought that you were just opening the calipers to 100 mm to run
> their tips in and out and confirm that the digital readout was
> working.
> After my bike ride, I'll have to look into measuring things that way
> and see what I get.
Hold it! A tensiometer measures far smaller length changes than can
be measured by lineal measurements by indirect means. By deflecting
spokes bu 0.015-0.020" in the middle of a 4 inch span to measure
reasonable tension, an angular change of arctan(0.020 / 2.00) = 0.57°
or 0.0001" length change. You are looking to measure a small part of
that, something you cannot measure directly. For such measurements
strain gauges are used and that is why making a tensiometer is not a
trivial design problem.
Jobst Brandt
Dan
jobst....@stanfordalumni.org wrote:
>
> Hold it! A tensiometer measures far smaller length changes than can
> be measured by lineal measurements by indirect means. By deflecting
> spokes bu 0.015-0.020" in the middle of a 4 inch span to measure
> reasonable tension, an angular change of arctan(0.020 / 2.00) = 0.57°
> or 0.0001" length change. You are looking to measure a small part of
> that, something you cannot measure directly. For such measurements
> strain gauges are used and that is why making a tensiometer is not a
> trivial design problem.
>
> Jobst Brandt
Yes, my machinist's caliper method is not very sensitive. I can only
detect .01mm in 100mm or about 100 micro-strains. This is very crude by
strain gauge standards.
Joe Riel
> Hold it! A tensiometer measures far smaller length changes than can
> be measured by lineal measurements by indirect means. By deflecting
> spokes bu 0.015-0.020" in the middle of a 4 inch span to measure
> reasonable tension, an angular change of arctan(0.020 / 2.00) = 0.57°
> or 0.0001" length change.
I'm not sure what you mean. A tensiometer doesn't measure a change in
length (of the spoke/wire). While the deflected path is longer than
the straight path, we don't know, nor do we need to know, the actual
elongation. Ideally the elongation is minimized so that the tension
change is minimal.
--
Joe Riel
carl...@comcast.net
Dear Dan,
For what it's worth, I laid an MTB wheel out on a workbench and fixed
some wire coils about 6 inches apart, leaving the ends of the coils
facing up to make a convenient stop for my dial calipers between x and
y in the diagram below:
______///___________________________///___________spoke
///x y///
A long quarter-inch strip of tape secured each coil on the 2 mm
straight spoke. I measured things, backed off a full turn, and
measured again.
turns dial Park
loosened caliper gauge
mark
20 68 kgf
0 6.060" 19.8
19 62 kgf
18 56 kgf
17 51 kgf
1 6.067" 16.0 ~45 kgf, off park chart
2 6.070" ~9.0 too far off Park chart to even guess
The .8 means only that the needle kept going just below the 20.0 mark,
and eyeballed to that approximation. 16.0 is the very bottom of the
scale's useful reading, and the ~9.0 indicates absurdly low tension.
A much more expensive dial-indicator DT-Swiss-style tension gauge
would probably be even more accurate
Apart from being immensely more convenient, the Park gauge looks to be
much more accurate than a reasonable dial caliper that reads in
thousandths of an inch of spoke stretch over a 6-inch span, but the
dial caliper can indeed show full-nipple-turn changes.
Cheers,
Carl Fogel
jobst....@stanfordalumni.org
>> Hold it! A tensiometer measures far smaller length changes than
>> can be measured by lineal measurements by indirect means. By
>> deflecting spokes bu 0.015-0.020" in the middle of a 4 inch span to
>> measure reasonable tension, an angular change of arctan(0.020 /
>> 2.00) = 0.57° or 0.0001" length change.
> I'm not sure what you mean.
What I mean is that the strain in spokes is on the order of 1/1000"
and that trying to assess changes with linear manual measurement tools
is not going to reveal any repeatable or significant changes.
> A tensiometer doesn't measure a change in length (of the
> spoke/wire). While the deflected path is longer than the straight
> path, we don't know, nor do we need to know, the actual elongation.
> Ideally the elongation is minimized so that the tension change is
> minimal.
The tensiometer makes apparent what realm these dimensions reside in.
I have mentioned that people who can feel the smoother ride of 1.6mm
diameter spokes over 2.0mm would have to be able to tell whether they
rode over a sheet of copier paper lying on the road to sense such
changes. Elasticity in wheels is not in the realm of what you can
feel. These can only be detected through secondary means such as
lateral deflections.
Jobst Brandt
Mark Hickey
>Carl-
>
>I think you may have misinterpreted my post. I was not measuring the
>change in spoke diameter (Poisson's effect) but the change in length.
I'm just drooling even THINKING about a pair of calipers with the
resolution to measure changes in the diameter of a spoke under load.
Maybe with one of those, I could start measuring frame vertical
compliance... ;-)
Mark Hickey
Habanero Cycles
http://www.habcycles.com
Home of the $795 ti frame
Dan
jobst....@stanfordalumni.org wrote:
>
> Hold it! A tensiometer measures far smaller length changes than can
> be measured by lineal measurements by indirect means. By deflecting
> spokes bu 0.015-0.020" in the middle of a 4 inch span to measure
> reasonable tension, an angular change of arctan(0.020 / 2.00) = 0.57°
> or 0.0001" length change. You are looking to measure a small part of
> that, something you cannot measure directly. For such measurements
> strain gauges are used and that is why making a tensiometer is not a
> trivial design problem.
>
> Jobst Brandt
Went back to the garage and tried to hone my technique but ended
in clutsy failure.
http://www.geocities.com/banquo_lives/Spokes/index.html
carl...@comcast.net
Dear Dan,
Half the fun is trying.
And people like me enjoy looking at your pictures. I like the clamping
of the calipers to the spokes--much quicker and more reliable than my
approach of winding wire and taping and fiddling.
Cheers,
Carl Fogel
jobst....@stanfordalumni.org
>> Hold it! A tensiometer measures far smaller length changes than can
>> be measured by lineal measurements by indirect means. By deflecting
>> spokes bu 0.015-0.020" in the middle of a 4 inch span to measure
>> reasonable tension, an angular change of arctan(0.020 / 2.00) = 0.57°
>> or 0.0001" length change. You are looking to measure a small part of
>> that, something you cannot measure directly. For such measurements
>> strain gauges are used and that is why making a tensiometer is not a
>> trivial design problem.
> Went back to the garage and tried to hone my technique but ended
> in clutsy failure.
http://www.geocities.com/banquo_lives/Spokes/index.html
Nice work. That looks like you will get some reasonable results if
you can convert the readings to inches or inches per inch.
Jobst Brandt