Re: Spoke tension meter-John Allen responds

[Luns Tee]

In article <[email protected]>,
jim beam <[email protected]> wrote:
>luns, to be honset, i'm not going to invest time in a guy that's letting
>personal hostility get in the way of their cognative abilities. go to
>the library. look up bending mechanics. check into spring manufacture.

Translation: you're pulling stuff out of thin air, and now that
you can't support it any more, you pretend it's common knowledge and try
to make me do your homework for you. You're the one who needs to look up
bending mechanics. What you've purported is implausible and I've
explained - now past the point of repeating myself - exactly why, and
it's exactly why what you've been claiming will not be found in a
library. You're welcome to prove me wrong with a citation, but to tell
me to go the library to prove _your_ argument - one which I've already
explained to be fallacious and would not be found there - is as absurd
as demanding a certain regime surrender its non-existant weapons.

Spring manufacture: you claim that the existance of close-wound
springs is evidence that wire winds with its neutral axis on the
inner surface. This is a specious claim that does not follow at all. The
only argument you've offered to connect the two is that the
spring-back that occurs if the wire is wound with the neutral axis in
the material - springback that happens even if the neutral axis is at
the surface anyway - would cause coil turns to extend from each other.

Extension springs can be wound with coils touching each
other - indeed, pressing egainst each other - by winding them to one
helix and then elastically inverting the spiral to a reverse helix.
This inversion is easily demonstrated with a coiled telephone handset
cord: an old stretched cord can be inverted and turns of the coil then
tightly press against each other. A fully formed metal spring is
harder to invert in the same manner, but this inversion happens on the
winding arbor, at the time of spring winding. The just-formed
to-be-inverted helix only exists as roughly a quarter turn on the
arbor, and the just-completed spring gets its first extension exercise
stretching to meet it. It has absolutely nothing to do with applied
tension during winding.

Don't believe me? Go look it up in the library. I'll even save
you the trouble of getting out of your chair - what you find there
will agree with:

http://home.earthlink.net/~bazillion/extension.html

the key to the process is the gap highlighted in the figure for
step 4.

-Luns

 

[[email protected]]

On Tue, 21 Jun 2005 17:02:34 +0000 (UTC),
[email protected] (Luns Tee) wrote:

>In article <[email protected]>,
>jim beam <[email protected]> wrote:
>>luns, to be honset, i'm not going to invest time in a guy that's letting
>>personal hostility get in the way of their cognative abilities. go to
>>the library. look up bending mechanics. check into spring manufacture.
>
> Translation: you're pulling stuff out of thin air, and now that
>you can't support it any more, you pretend it's common knowledge and try
>to make me do your homework for you. You're the one who needs to look up
>bending mechanics. What you've purported is implausible and I've
>explained - now past the point of repeating myself - exactly why, and
>it's exactly why what you've been claiming will not be found in a
>library. You're welcome to prove me wrong with a citation, but to tell
>me to go the library to prove _your_ argument - one which I've already
>explained to be fallacious and would not be found there - is as absurd
>as demanding a certain regime surrender its non-existant weapons.
>
> Spring manufacture: you claim that the existance of close-wound
>springs is evidence that wire winds with its neutral axis on the
>inner surface. This is a specious claim that does not follow at all. The
>only argument you've offered to connect the two is that the
>spring-back that occurs if the wire is wound with the neutral axis in
>the material - springback that happens even if the neutral axis is at
>the surface anyway - would cause coil turns to extend from each other.
>
> Extension springs can be wound with coils touching each
>other - indeed, pressing egainst each other - by winding them to one
>helix and then elastically inverting the spiral to a reverse helix.
>This inversion is easily demonstrated with a coiled telephone handset
>cord: an old stretched cord can be inverted and turns of the coil then
>tightly press against each other. A fully formed metal spring is
>harder to invert in the same manner, but this inversion happens on the
>winding arbor, at the time of spring winding. The just-formed
>to-be-inverted helix only exists as roughly a quarter turn on the
>arbor, and the just-completed spring gets its first extension exercise
>stretching to meet it. It has absolutely nothing to do with applied
>tension during winding.
>
> Don't believe me? Go look it up in the library. I'll even save
>you the trouble of getting out of your chair - what you find there
>will agree with:
>
>http://home.earthlink.net/~bazillion/extension.html
>
> the key to the process is the gap highlighted in the figure for
>step 4.
>
>-Luns

Dear Luns,

Off-topic, but browsing the spring site led me to two odd
things:

"Put your spring into the oven and relieve the stress.
Remember, springs made of stainless steel wire will open up
a little in the heat, while springs made of music wire will
contract."

http://home.earthlink.net/~bazillion/extension.html

Elsewhere on the site, the glossary describes stainless as s
302/17-7 steel, which may not behave like the 316/18-8 used
in spokes:

http://home.earthlink.net/~bazillion/glossary.html

But it's odd that carbon and stainless steel behave so
differently when bent into springs and heated, since they're
the two main kinds of spoke material.

Does anyone know why the springs made from carbon steel and
stainless steel behave in opposite ways after being stress
relieved in an oven?

The other odd thing was the passivation section here:

http://home.earthlink.net/~bazillion/finish.html

"Working with stainless steel creates "free iron"
contamination on the surface that can later rust and create
problems with the part and its working environment. The
process of removing this free iron is called passivation."

Boyle M. Owl has mentioned this elsewhere in this thread.

If "working" means bending the stainless steel, does this
mean that the spoke makers are (or should be) dunking the
spokes in muriatic or nitric acid for an hour after bending
the elbows? Failure to get rid of this 'free iron' might
explain a recent thread that puzzled people about apparently
rusty stainless steel spokes:

http://groups-beta.google.com/group...43df?q=rust+stainless&rnum=2#d53b8baa32d143df

Could these rusty and (possibly) stainless steel spokes have
been poorly finished (no passivating) and therefore showing
real rust spots, which in turn would be a bad thing for
spoke fatigue?

Carl Fogel

 

[41]

41 wrote:

> Quick, "Carl Fogel", come save your bourbon-soaked a lter-ego "jim beam"
> from yet another public and complete intellectual humilation! Hurry,
> change the subject, or start a lengthy digression! The thread hasn't
> reached 500 messages yet, don't let it die with him!
>
> Really, somebody do something, because "jim beam" has been exposed yet
> again, he's walking around naked, and it ain't a pretty sight.

A scant 2hrs22min later by the Google clock,
[email protected] saved the day, with at least three separate
messages:

> Off-topic, but

Whew! Bravo Carl, you really outdid yourself this time! And thank
you- it was getting really ugly! My eyes!!!! My eyes!!!!!

 

[Luns Tee]

In article <[email protected]>,
<[email protected]> wrote:
>"Put your spring into the oven and relieve the stress.
>Remember, springs made of stainless steel wire will open up
>a little in the heat, while springs made of music wire will
>contract."
>
>http://home.earthlink.net/~bazillion/extension.html

>But it's odd that carbon and stainless steel behave so
>differently when bent into springs and heated, since they're
>the two main kinds of spoke material.
>
>Does anyone know why the springs made from carbon steel and
>stainless steel behave in opposite ways after being stress
>relieved in an oven?

It doesn't surprise me that one opens and the other contracts,
but I don't know the exact details of the material properties to cause
this difference. I would expect the difference to only apply in some
range of spring radii relative to wire thickness, but perhaps the range
is broad enough to encompass most/all of what one would typically wind.

What it amounts to though, is whether the residual stress
that's relieved by annealing is primarily the outermost layers,
or the inner ones. Relaxing the stress in the outermost layers would
open up the bending, while relaxing internal stress would tighten it
more.

There are four peaks in the residual stress (three if the wire
is tensioned enough during winding that the inside of the bend doesn't
yield much or at all - the unyielded surface counting as an internal
peak), and it's a question of which of these is the most affected by
annealing temperatures. The yield points under tension and compression
aren't necessarily symmetric, and the relative proportions of each,
and how the annealing affects them, is beyond my knowledge.

-Luns

 

[[email protected]]

On Wed, 22 Jun 2005 00:54:34 +0000 (UTC),
[email protected] (Luns Tee) wrote:

>In article <[email protected]>,
> <[email protected]> wrote:
>>"Put your spring into the oven and relieve the stress.
>>Remember, springs made of stainless steel wire will open up
>>a little in the heat, while springs made of music wire will
>>contract."
>>
>>http://home.earthlink.net/~bazillion/extension.html
>
>>But it's odd that carbon and stainless steel behave so
>>differently when bent into springs and heated, since they're
>>the two main kinds of spoke material.
>>
>>Does anyone know why the springs made from carbon steel and
>>stainless steel behave in opposite ways after being stress
>>relieved in an oven?
>
> It doesn't surprise me that one opens and the other contracts,
>but I don't know the exact details of the material properties to cause
>this difference. I would expect the difference to only apply in some
>range of spring radii relative to wire thickness, but perhaps the range
>is broad enough to encompass most/all of what one would typically wind.
>
> What it amounts to though, is whether the residual stress
>that's relieved by annealing is primarily the outermost layers,
>or the inner ones. Relaxing the stress in the outermost layers would
>open up the bending, while relaxing internal stress would tighten it
>more.
>
> There are four peaks in the residual stress (three if the wire
>is tensioned enough during winding that the inside of the bend doesn't
>yield much or at all - the unyielded surface counting as an internal
>peak), and it's a question of which of these is the most affected by
>annealing temperatures. The yield points under tension and compression
>aren't necessarily symmetric, and the relative proportions of each,
>and how the annealing affects them, is beyond my knowledge.
>
>-Luns

Dear Luns,

Hmmm . . . I think that I follow you.

Maybe two identically wound springs would decide to relieve
stresses differently according to materials.

Now I'm wondering if instead there's something that the
steel-folk would explain about body and face centered
this-and-that at different temperatures.

In the finishing section on that site, different
temperatures and times in the oven are listed:

http://home.earthlink.net/~bazillion/finish.html

The music wire closes after an hour at 500 F, and oil
tempered steel of some kind takes only half an hour at 500F
to "close slightly."

The stainless 302 opens after an hour at 600 F, and the
stainless 17-4 or 17-7 takes an hour at 900 F.

Some other materials open or close (titanium needs ten hours
(!) to open at 800 F) and a fair number don't do anything,
like "basic" spring wire, whatever that's made of.

Carl Fogel

 

[jim beam]

Luns Tee wrote:
> In article <[email protected]>,
> jim beam <[email protected]> wrote:
>
>>luns, to be honset, i'm not going to invest time in a guy that's letting
>>personal hostility get in the way of their cognative abilities. go to
>>the library. look up bending mechanics. check into spring manufacture.
>
>
> Translation: you're pulling stuff out of thin air, and now that
> you can't support it any more, you pretend it's common knowledge and try
> to make me do your homework for you. You're the one who needs to look up
> bending mechanics. What you've purported is implausible and I've
> explained - now past the point of repeating myself - exactly why, and
> it's exactly why what you've been claiming will not be found in a
> library. You're welcome to prove me wrong with a citation, but to tell
> me to go the library to prove _your_ argument - one which I've already
> explained to be fallacious and would not be found there - is as absurd
> as demanding a certain regime surrender its non-existant weapons.
>
> Spring manufacture: you claim that the existance of close-wound
> springs is evidence that wire winds with its neutral axis on the
> inner surface. This is a specious claim that does not follow at all. The
> only argument you've offered to connect the two is that the
> spring-back that occurs if the wire is wound with the neutral axis in
> the material - springback that happens even if the neutral axis is at
> the surface anyway - would cause coil turns to extend from each other.
>
> Extension springs can be wound with coils touching each
> other - indeed, pressing egainst each other - by winding them to one
> helix and then elastically inverting the spiral to a reverse helix.
> This inversion is easily demonstrated with a coiled telephone handset
> cord: an old stretched cord can be inverted and turns of the coil then
> tightly press against each other. A fully formed metal spring is
> harder to invert in the same manner, but this inversion happens on the
> winding arbor, at the time of spring winding. The just-formed
> to-be-inverted helix only exists as roughly a quarter turn on the
> arbor, and the just-completed spring gets its first extension exercise
> stretching to meet it. It has absolutely nothing to do with applied
> tension during winding.
>
> Don't believe me? Go look it up in the library. I'll even save
> you the trouble of getting out of your chair - what you find there
> will agree with:
>
> http://home.earthlink.net/~bazillion/extension.html
>
> the key to the process is the gap highlighted in the figure for
> step 4.
>
> -Luns

on one hand, i should thank you for bothering to look something up, even
if it [conveniently?] covers no bending theory, just practice. on the
other, i gotta ask, did you bother to read it? see in paragraph 4,
where it mentions the requirement for tension while winding? i said
that to you 3 days ago, but it didn't register then, and sadly, it
doesn't seem to register now. you cannot wind a close bound spring
without sufficient tension to ensure tensile yielding through the whole
cross section of the wire, not just the outer part. and by definition,
if you yield the whole wire in tension, there is no neutral plane!!!

 

[[email protected]]

jim beam wrote:
>
>
> on one hand, i should thank you for bothering to look something up, even
> if it [conveniently?] covers no bending theory, just practice. on the
> other, i gotta ask, did you bother to read it? see in paragraph 4,
> where it mentions the requirement for tension while winding? i said
> that to you 3 days ago, but it didn't register then, and sadly, it
> doesn't seem to register now.

Seems to me you're misinterpreting that web page.

The author of the web page appears to be telling how to wind an
extension spring so that it contains preload, which he calls "initial
tension."

IOW: The typical extension spring's force-vs-deflection curve is not a
straight line passing through the zero-zero origin. Instead, the
typical curve is a vertical line running up from the origin to a
certain force value, that required to begin separating the coils. From
that force value (the author's "initial tension") the F-x curve becomes
a straight line with positive slope. In mathematical terms, F =/= k*x
as it does with a typical compression spring; instead, F = k*x + Fo
where Fo is the preload, or "initial tension."

This has nothing to do with the tension you are claiming must be
applied, AFAIKT. I find no mention of applying tension to the wire
during the winding process, other than the (unmentioned) tension due to
the friction of the wire sliding over the wire guide.

> you cannot wind a close bound spring
> without sufficient tension to ensure tensile yielding through the whole
> cross section of the wire, not just the outer part.

I don't see anything in that reference to verify this. Do you have
another source for corroboration?

- Frank Krygowski

 

[jim beam]

[email protected] wrote:
>
> jim beam wrote:
>
>>
>>on one hand, i should thank you for bothering to look something up, even
>>if it [conveniently?] covers no bending theory, just practice. on the
>>other, i gotta ask, did you bother to read it? see in paragraph 4,
>>where it mentions the requirement for tension while winding? i said
>>that to you 3 days ago, but it didn't register then, and sadly, it
>>doesn't seem to register now.
>
>
> Seems to me you're misinterpreting that web page.
>
> The author of the web page appears to be telling how to wind an
> extension spring so that it contains preload, which he calls "initial
> tension."
>
> IOW: The typical extension spring's force-vs-deflection curve is not a
> straight line passing through the zero-zero origin. Instead, the
> typical curve is a vertical line running up from the origin to a
> certain force value, that required to begin separating the coils. From
> that force value (the author's "initial tension") the F-x curve becomes
> a straight line with positive slope. In mathematical terms, F =/= k*x
> as it does with a typical compression spring; instead, F = k*x + Fo
> where Fo is the preload, or "initial tension."
>
> This has nothing to do with the tension you are claiming must be
> applied, AFAIKT. I find no mention of applying tension to the wire
> during the winding process, other than the (unmentioned) tension due to
> the friction of the wire sliding over the wire guide.

well, you could get some copper wire & a nail & test for yourelf how
easy is it to wind your own close bound spring. you need to apply
yielding tension to achieve it.

>
>
>>you cannot wind a close bound spring
>>without sufficient tension to ensure tensile yielding through the whole
>>cross section of the wire, not just the outer part.
>
>
> I don't see anything in that reference to verify this. Do you have
> another source for corroboration?
>
> - Frank Krygowski
>

hard to search isn't it... seems "spring" is a common word.

i'll look more in the deformation mechanics stuff & see what i can
reference for you.

 

[Luns Tee]

In article <[email protected]>,
jim beam <[email protected]> wrote:
>> Extension springs can be wound with coils touching each
>> other - indeed, pressing egainst each other - by winding them to one
>> helix and then elastically inverting the spiral to a reverse helix.
>> This inversion is easily demonstrated with a coiled telephone handset
>> cord: an old stretched cord can be inverted and turns of the coil then
>> tightly press against each other. A fully formed metal spring is
>> harder to invert in the same manner, but this inversion happens on the
>> winding arbor, at the time of spring winding. The just-formed
>> to-be-inverted helix only exists as roughly a quarter turn on the
>> arbor, and the just-completed spring gets its first extension exercise
>> stretching to meet it. It has absolutely nothing to do with applied
>> tension during winding.
>>
>> Don't believe me? Go look it up in the library. I'll even save
>> you the trouble of getting out of your chair - what you find there
>> will agree with:
>>
>> http://home.earthlink.net/~bazillion/extension.html
>>
>> the key to the process is the gap highlighted in the figure for
>> step 4.
>>
>> -Luns
>
>on one hand, i should thank you for bothering to look something up, even
>if it [conveniently?] covers no bending theory, just practice. on the
>other, i gotta ask, did you bother to read it? see in paragraph 4,
>where it mentions the requirement for tension while winding?

Before you ask, try reading it yourself. The only mention of
tension in the article is of "initial tension" as the author calls it
which is the preload of the completed spring, the meaning of which is
described right there immediately after the term is first used. There
is no mention whatsoever of tensioning the wire as it's fed.

The existance of coils with turns touching each other has
absolutely nothing to do with yielding the entire wire in tension.
Contact between turns is supported by residual _torsional_ shear that's
accomplished by bending the wire on line that lies on a different (tighter
or even reverse) helix from what it ends up after spring-back. I've
already explained how this happens, and the article I gave - one which
you obviously didn't read and understand aside from skimming for the word
'tension' and finding the word 'initial' before it.

>i said
>that to you 3 days ago, but it didn't register then, and sadly, it
>doesn't seem to register now. you cannot wind a close bound spring
>without sufficient tension to ensure tensile yielding through the whole
>cross section of the wire, not just the outer part. and by definition,
>if you yield the whole wire in tension, there is no neutral plane!!!

I knew this would be your next dodge. First you were arguing
that there is a neutral plane at the surface of the spoke and only
material on the outside of the bend yields. Now you're arguing the
entire spoke is in yield and there is no neutral plane. You keep
changing your story every time you get cornered without admitting what
you'd been claiming before is wrong, instead saying your new story is
what you meant all along.

Your theory of the entire section being in tensile yield doesn't
hold water either. The gouge on the outside of the spoke elbow is not an
adequate grip on the spoke to support against the entire spoke's section
yielding. Any grip on the end of the spoke would have to be supported
around the full cone of the head, and the spoke then pulled around the
mandrel you've been claiming is used. If the spoke is being pulled,
and under yield, there is no need to press on the spoke end where the
hammer mark is found.

Besides which, if the spoke is pulled to complete yield, then
there will be significant necking down of the spoke around the elbow,
the outside of the bend being at a radius roughly 3x the inside of the
bend, hence material there having to extend to 3x the length, and thin to
~1/3 of its original area. While the spoke as a whole would thin to an
overall section not as drastic as 1/3, it'd still be significantly
thinner than the straight sections of the spoke, with roughly half the
area of the wire it was stretched from. Measuring the metal in the
elbow of a DT spoke I have here with me, the elbow wire is identical
in width to the straight wire, and only 5% thinner in thickness
(inside of bend to outside of bend). This is not the significant
reduction in cross section that a complete tensile yield would dictate.

-Luns

 

[[email protected]]

On Wed, 22 Jun 2005 13:55:06 +0000 (UTC),
[email protected] (Luns Tee) wrote:

[snip]

> Besides which, if the spoke is pulled to complete yield, then
>there will be significant necking down of the spoke around the elbow,
>the outside of the bend being at a radius roughly 3x the inside of the
>bend, hence material there having to extend to 3x the length, and thin to
>~1/3 of its original area. While the spoke as a whole would thin to an
>overall section not as drastic as 1/3, it'd still be significantly
>thinner than the straight sections of the spoke, with roughly half the
>area of the wire it was stretched from. Measuring the metal in the
>elbow of a DT spoke I have here with me, the elbow wire is identical
>in width to the straight wire, and only 5% thinner in thickness
>(inside of bend to outside of bend). This is not the significant
>reduction in cross section that a complete tensile yield would dictate.
>
>-Luns

Dear Luns,

I'm not able to follow the tension debate that I snipped
very well, but it's worth trying just for little nuggets
like your comment about the thickness.

I took a straight 14 gauge Sapim spoke and measured it with
my dial calipers and came up with the same results.

The spoke shaft is 0.078" thick--the needle doesn't waver
when I slide and twirl the main spoke through the caliper
jaws.

Then the jaws hit the little flattened stripe coming out of
the elbow and there's a faint but definite change--the
needle bumps up to 0.079", still on the shaft and just short
of the elbow. I assume that flattening the stripe squashed
the metal out sidways just a tiny bit.

In the elbow itself, I get two readings, just like yours.
From side to side, the width stays about the same, 0.078",
but from inner to outer curve, it drops to 0.074" or maybe
even a smidgen more, with the 4/78 the same as your 5%
figure.

The necking may not be significant in terms of the tensile
yield question that you and Jim are discussing, but I'm
fascinated to learn that the bending process leaves the
damned spoke apparently butted (or possibly swaged) right in
the worst possible place!

For comparison, I measured a 2.0-1.8-2.0 spoke and found
that the same 0.078" drops to 0.070" and then figured out
the cross-section, assuming an ellipse with a major and
minor radius of 0.039" and 0.037":

0.078 x 0.078 round 2.0mm shaft 0.004778 "^2 100.0%
0.078 x 0.074 elliptical elbow 0.004288 "^2 89.7%
0.070 x 0.070 round 1.8mm shaft 0.003848 "^2 80.5%

I may not follow the rest of what you two are saying, but I
love finding out things like this. With luck, I'll hear more
about this necking and whether it matters.

Carl Fogel.

 

[[email protected]]

Luns Tee writes:

About preloaded tension springs that do not start from zero tension
when at rest. A coil spring is a torsion bar in effect and it can be
wound so that it has a tension preload when its coils are "in block"
stacked against one another. This is done by winding the coil spring
around a mandrel with spring wire that is under torsion in the
appropriate direction. Subsequently the spring is not heat treated
allowing its work hardened wire serve in the as wound condition.

I don't see what bearing this has on spokes in a bicycle wheel
however. I think you are being thrown a decoy the the alcoholic one.

[email protected]

 

[jim beam]

Luns Tee wrote:
> In article <[email protected]>,
> jim beam <[email protected]> wrote:
>
>>> Extension springs can be wound with coils touching each
>>>other - indeed, pressing egainst each other - by winding them to one
>>>helix and then elastically inverting the spiral to a reverse helix.
>>>This inversion is easily demonstrated with a coiled telephone handset
>>>cord: an old stretched cord can be inverted and turns of the coil then
>>>tightly press against each other. A fully formed metal spring is
>>>harder to invert in the same manner, but this inversion happens on the
>>>winding arbor, at the time of spring winding. The just-formed
>>>to-be-inverted helix only exists as roughly a quarter turn on the
>>>arbor, and the just-completed spring gets its first extension exercise
>>>stretching to meet it. It has absolutely nothing to do with applied
>>>tension during winding.
>>>
>>> Don't believe me? Go look it up in the library. I'll even save
>>>you the trouble of getting out of your chair - what you find there
>>>will agree with:
>>>
>>>http://home.earthlink.net/~bazillion/extension.html
>>>
>>> the key to the process is the gap highlighted in the figure for
>>>step 4.
>>>
>>>-Luns
>>
>>on one hand, i should thank you for bothering to look something up, even
>>if it [conveniently?] covers no bending theory, just practice. on the
>>other, i gotta ask, did you bother to read it? see in paragraph 4,
>>where it mentions the requirement for tension while winding?
>
>
> Before you ask, try reading it yourself. The only mention of
> tension in the article is of "initial tension" as the author calls it
> which is the preload of the completed spring, the meaning of which is
> described right there immediately after the term is first used. There
> is no mention whatsoever of tensioning the wire as it's fed.
>
> The existance of coils with turns touching each other has
> absolutely nothing to do with yielding the entire wire in tension.
> Contact between turns is supported by residual _torsional_ shear that's
> accomplished by bending the wire on line that lies on a different (tighter
> or even reverse) helix from what it ends up after spring-back. I've
> already explained how this happens, and the article I gave - one which
> you obviously didn't read and understand aside from skimming for the word
> 'tension' and finding the word 'initial' before it.

with respect luns, i did read the article thanks. it's a very handy
guide for the hobbyist. but it does not cover deformation theory, the
part i've been vainly trying to explain. in the absence of having carls
patience for ascii art, yet again, i invite you to actually try winding
a close bound spring and tell me how far you get without using tension.
the tension causes tensile yielding across the whole section of the
wire, more on the outside obviously, less on the inside, but tensile
yielding nonetheless. springback from that yielding holds the spring
coils close. "ordinary" bending, with tensile on the outside,
compressive on the outside causes springback that has the coils apart.
think about it ust a little more.

>
>
>>i said
>>that to you 3 days ago, but it didn't register then, and sadly, it
>>doesn't seem to register now. you cannot wind a close bound spring
>>without sufficient tension to ensure tensile yielding through the whole
>>cross section of the wire, not just the outer part. and by definition,
>>if you yield the whole wire in tension, there is no neutral plane!!!
>
>
> I knew this would be your next dodge. First you were arguing
> that there is a neutral plane at the surface of the spoke and only
> material on the outside of the bend yields. Now you're arguing the
> entire spoke is in yield and there is no neutral plane. You keep
> changing your story every time you get cornered without admitting what
> you'd been claiming before is wrong, instead saying your new story is
> what you meant all along.
>
> Your theory of the entire section being in tensile yield doesn't
> hold water either. The gouge on the outside of the spoke elbow is not an
> adequate grip on the spoke to support against the entire spoke's section
> yielding. Any grip on the end of the spoke would have to be supported
> around the full cone of the head, and the spoke then pulled around the
> mandrel you've been claiming is used. If the spoke is being pulled,
> and under yield, there is no need to press on the spoke end where the
> hammer mark is found.
>
> Besides which, if the spoke is pulled to complete yield

sorry luns, but what exactly is "complete yield"? you mean a single
atomic dislocation movement [which is yield], 0.2% plastic strain [which
is yield], 10% plastic strain? 50% plastic strain? or do you mean rupture?

>, then
> there will be significant necking down of the spoke around the elbow,

no, you get yielding long before you get necking.

> the outside of the bend being at a radius roughly 3x the inside of the
> bend, hence material there having to extend to 3x the length, and thin to
> ~1/3 of its original area. While the spoke as a whole would thin to an
> overall section not as drastic as 1/3, it'd still be significantly
> thinner than the straight sections of the spoke, with roughly half the
> area of the wire it was stretched from. Measuring the metal in the
> elbow of a DT spoke I have here with me, the elbow wire is identical
> in width to the straight wire, and only 5% thinner in thickness
> (inside of bend to outside of bend). This is not the significant
> reduction in cross section that a complete tensile yield would dictate.

not significant??? even if the above arguments made sense, considerably
less than "5% thinner" can be the difference between a 747 engine bolt
holding or failing. i personally consider that "significant" when my
plane is at 160kts, maximum takeoff weight, maximum fuel, 8x10^5N thrust
and we're 2/3'rds down the runway. please, that's an incredibly naive
argument.

>
> -Luns

 

[jim beam]

[email protected] wrote:
> Luns Tee writes:
>
> About preloaded tension springs that do not start from zero tension
> when at rest. A coil spring is a torsion bar in effect and it can be
> wound so that it has a tension preload when its coils are "in block"
> stacked against one another. This is done by winding the coil spring
> around a mandrel with spring wire that is under torsion in the
> appropriate direction.

and torsion around the mandrel is tension in the wire.

> Subsequently the spring is not heat treated
> allowing its work hardened wire serve in the as wound condition.
>
> I don't see what bearing this has on spokes in a bicycle wheel
> however. I think you are being thrown a decoy the the alcoholic one.
>
> [email protected]

 

[Luns Tee]

In article <[email protected]>,
jim beam <[email protected]> wrote:
>with respect luns, i did read the article thanks. it's a very handy
>guide for the hobbyist. but it does not cover deformation theory, the
>part i've been vainly trying to explain. in the absence of having carls
>patience for ascii art, yet again, i invite you to actually try winding
>a close bound spring and tell me how far you get without using tension.
> the tension causes tensile yielding across the whole section of the
>wire, more on the outside obviously, less on the inside, but tensile
>yielding nonetheless. springback from that yielding holds the spring
>coils close. "ordinary" bending, with tensile on the outside,
>compressive on the outside causes springback that has the coils apart.
>think about it ust a little more.

Just because _you_ can't wind such a spring doesn't prove it's
impossible: it proves that your understanding of how material deforms
is deficient. You obviously didn't follow the article: it describes
the process involving only incidental tension on the wire.

Here's a picture of some 28 gauge galvanized wire that I
wrapped, just rolling the wire onto a pen.

http://www.ocf.berkeley.edu/~tee/rbt/Dsc00251.jpg

The key is to maintain the gap shown in the figure for step 4 of

http://home.earthlink.net/~bazillion/extension.html

For this, rather than rely on wire stiffness to make the gap I
just stuck my fingernail in between the turns of the wire while
rolling the pen and coil onto the wire tail along my thumb. The wire
is pressed onto the pen: no tension is applied.

>> around the full cone of the head, and the spoke then pulled around the
>> mandrel you've been claiming is used. If the spoke is being pulled,
>> and under yield, there is no need to press on the spoke end where the
>> hammer mark is found.
>>
>> Besides which, if the spoke is pulled to complete yield
>
>sorry luns, but what exactly is "complete yield"? you mean a single
>atomic dislocation movement [which is yield], 0.2% plastic strain [which
>is yield], 10% plastic strain? 50% plastic strain? or do you mean rupture?

Yield across the whole section of the spoke as you claim
occurs.

>>, then
>> there will be significant necking down of the spoke around the elbow,
>
>no, you get yielding long before you get necking.
>
>> the outside of the bend being at a radius roughly 3x the inside of the
>> bend, hence material there having to extend to 3x the length, and thin to
>> ~1/3 of its original area. While the spoke as a whole would thin to an
>> overall section not as drastic as 1/3, it'd still be significantly
>> thinner than the straight sections of the spoke, with roughly half the
>> area of the wire it was stretched from. Measuring the metal in the
>> elbow of a DT spoke I have here with me, the elbow wire is identical
>> in width to the straight wire, and only 5% thinner in thickness
>> (inside of bend to outside of bend). This is not the significant
>> reduction in cross section that a complete tensile yield would dictate.
>
> not significant??? even if the above arguments made sense, considerably
>less than "5% thinner" can be the difference between a 747 engine bolt
>holding or failing.

Either you're deliberately misreading what's been written - it
wouldn't surprise me the least bit - or your reading comprehension
fails you. I never said that 5% is insigificant, only that it's far
less than the much more significant reduction that would occur with
yielding as you've described it.
The wire is bent with an outer radius of 3x its inner radius,
with the material being nominally at 2x the inner radius. If all the
material is in tensile yield as you claim, then the material inside
the bend is at least as long as the length of straight wire the elbow
is formed from, and the material being at nominally twice that radius
stretches on average to 2x its starting length. This would cause a
reduction in area of 50%. What's observed is on the order of 5%, if
even that. This shows quite clearly that material does not undergo
anywhere near as much tensile yield as would occur if your claim were
true.

>i personally consider that "significant" when my
>plane is at 160kts, maximum takeoff weight, maximum fuel, 8x10^5N thrust
>and we're 2/3'rds down the runway. please, that's an incredibly naive
>argument.

I see. You can't support your arguments so now you thump your
chest trying to associate yourself to big machines to allege some
authority associated with them. I suppose you drive a big macho truck,
too.

-Luns

 

[Luns Tee]

In article <[email protected]>,
<[email protected]> wrote:
>On Wed, 22 Jun 2005 13:55:06 +0000 (UTC),
>[email protected] (Luns Tee) wrote:
>The necking may not be significant in terms of the tensile
>yield question that you and Jim are discussing, but I'm
>fascinated to learn that the bending process leaves the
>damned spoke apparently butted (or possibly swaged) right in
>the worst possible place!

If not for residual stress from bending, the elbow is no worse a
place for thinning than anywhere else on the spoke.

For perspective, try estimating the root diameter of the
threads. Assume the thread crest rises above the wire diameter by as
much as the thread root sinks into it.

>For comparison, I measured a 2.0-1.8-2.0 spoke and found
>that the same 0.078" drops to 0.070" and then figured out
>the cross-section, assuming an ellipse with a major and
>minor radius of 0.039" and 0.037":
>
>0.078 x 0.078 round 2.0mm shaft 0.004778 "^2 100.0%
>0.078 x 0.074 elliptical elbow 0.004288 "^2 89.7%
>0.070 x 0.070 round 1.8mm shaft 0.003848 "^2 80.5%

A 5% change on one axis causes a 5% change in area - the numbers
on your second line are for a 0.074 diameter circle, not an ellipse.

-Luns

 

[Luns Tee]

In article <[email protected]>,
jim beam <[email protected]> wrote:
>[email protected] wrote:
>> Luns Tee writes:
>>
>> About preloaded tension springs that do not start from zero tension
>> when at rest. A coil spring is a torsion bar in effect and it can be
>> wound so that it has a tension preload when its coils are "in block"
>> stacked against one another. This is done by winding the coil spring
>> around a mandrel with spring wire that is under torsion in the
>> appropriate direction.
>
>and torsion around the mandrel is tension in the wire.

The appropriate direction is on the axis of the wire, not around
the mandrel.

-Luns

 

[jim beam]

Luns Tee wrote:
> In article <[email protected]>,
> jim beam <[email protected]> wrote:
>
>>[email protected] wrote:
>>
>>>Luns Tee writes:
>>>
>>>About preloaded tension springs that do not start from zero tension
>>>when at rest. A coil spring is a torsion bar in effect and it can be
>>>wound so that it has a tension preload when its coils are "in block"
>>>stacked against one another. This is done by winding the coil spring
>>>around a mandrel with spring wire that is under torsion in the
>>>appropriate direction.
>>
>>and torsion around the mandrel is tension in the wire.
>
>
> The appropriate direction is on the axis of the wire, not around
> the mandrel.
>
> -Luns

how many more times do i have to repeat that? yes, *tension* as in
*axial tension in the wire*!

 

[[email protected]]

Wild Turkey writes:

>> About preloaded tension springs that do not start from zero tension
>> when at rest. A coil spring is a torsion bar in effect and it can be
>> wound so that it has a tension preload when its coils are "in block"
>> stacked against one another. This is done by winding the coil spring
>> around a mandrel with spring wire that is under torsion in the
>> appropriate direction.

> and torsion around the mandrel is tension in the wire.

Not torsion about the mandrel but torsion about the axis of the wire
as in torsion bar causing compression preload betweenits coils.

>> Subsequently the spring is not heat treated allowing its work
>> hardened wire serve in the as wound condition.

>> I don't see what bearing this has on spokes in a bicycle wheel
>> however. I think you are being thrown a decoy the the alcoholic
>> one.

[email protected]

 

[[email protected]]

On Thu, 23 Jun 2005 09:29:29 +0000 (UTC),
[email protected] (Luns Tee) wrote:

>In article <[email protected]>,
> <[email protected]> wrote:
>>On Wed, 22 Jun 2005 13:55:06 +0000 (UTC),
>>[email protected] (Luns Tee) wrote:
>>The necking may not be significant in terms of the tensile
>>yield question that you and Jim are discussing, but I'm
>>fascinated to learn that the bending process leaves the
>>damned spoke apparently butted (or possibly swaged) right in
>>the worst possible place!
>
> If not for residual stress from bending, the elbow is no worse a
>place for thinning than anywhere else on the spoke.
>
> For perspective, try estimating the root diameter of the
>threads. Assume the thread crest rises above the wire diameter by as
>much as the thread root sinks into it.
>
>>For comparison, I measured a 2.0-1.8-2.0 spoke and found
>>that the same 0.078" drops to 0.070" and then figured out
>>the cross-section, assuming an ellipse with a major and
>>minor radius of 0.039" and 0.037":
>>
>>0.078 x 0.078 round 2.0mm shaft 0.004778 "^2 100.0%
>>0.078 x 0.074 elliptical elbow 0.004288 "^2 89.7%
>>0.070 x 0.070 round 1.8mm shaft 0.003848 "^2 80.5%
>
> A 5% change on one axis causes a 5% change in area - the numbers
>on your second line are for a 0.074 diameter circle, not an ellipse.
>
>-Luns

Dear Luns,

Aaargh!

Yes, I got careless cutting diameters in half on my
spreadsheet.

0.078 x 0.078 round 2.0mm shaft 0.004778 "^2 100.0%
0.078 x 0.074 elliptical elbow 0.004533 "^2 94.9%
0.070 x 0.070 round 1.8mm shaft 0.003848 "^2 80.5%

I already looked into the threads rolled onto the end of the
spoke after posting yesterday. You can assume that the
rolled thread descends down as high as it rises, or you can
use a computer ruler on any of the nicely enlarged DTSwiss
spoke pictures (which gives the same result).

The crests of the rolled threads on a 0.078" 14-gauge 2.0 mm
shaft widen to about 0.087", or 2.21 mm (major diameter).

The roots of the rolled threads on the same spoke narrow to
about 0.071", or 1.81 mm (minor diameter).

See http://www.parktool.com/repair_help/thread.shtml for
nomenclature.

See
http://www.dtswiss.com/index.asp?fuseaction=spokes.bikedetail&id=7
for enlarged pictures of 2.0 mm DTSwiss Competition spokes.

See http://www.markus-bader.de/MB-Ruler/ for a nice onscreen
ruler and protractor, with other features.

Anyway, it's interesting that the two common places where
spokes break are necked down, one by the bending process
(which I'd never heard mentioned), the other by the
threading process (invariably discounted).

The manufacturing process reduces thickness by 5% at the
elbow, where anything but a perfect spoke-hub fit results in
faint bending as tension is lost and regained when the spoke
rolls under the hub.

The threaded end is reduced even more, down to about 0.071"
or so, or 1.8 mm. The bending at this end is much less, but
again, anything except a perfect alignment leads to bending
as tension is lost and regained.

Both processes presumably also lead to residual stresses,
but I don't recall the thinning of the material at the two
places where spokes break being given much comment. The
butting (or swaging) of spokes in the middle span is well
out on the straight shaft, where bending isn't going to
occur.

Do you know of any earlier mention on rec.bicycles.tech of
the 5% reduction in cross-section at the elbow? I may well
have missed someone else mentioning it, but I was quite
startled when I read your comment--and quite pleased to
learn something new, so thanks again, whether the reduction
matters or not.

Carl Fogel

 

[[email protected]]

[snip]

> Here's a picture of some 28 gauge galvanized wire that I
>wrapped, just rolling the wire onto a pen.
>
>http://www.ocf.berkeley.edu/~tee/rbt/Dsc00251.jpg
>
> The key is to maintain the gap shown in the figure for step 4 of
>
>http://home.earthlink.net/~bazillion/extension.html
>
> For this, rather than rely on wire stiffness to make the gap I
>just stuck my fingernail in between the turns of the wire while
>rolling the pen and coil onto the wire tail along my thumb. The wire
>is pressed onto the pen: no tension is applied.

[snip]

Dear Luns,

I'm not following how we can coil a wire onto a pen without
tension.

Your idea, I think, that the wire is being pushed against
the round surface of the pen at a constant right angle with
no tension on the wire:

pen
wire about to enter________ O
|
force

But I end up with this result:

pen
wire enters w/no tension____O______wire exits unbent
|
force

A tension-free wire passing between the roller of the pen
and some frictionless force would never coil around the pen,
would it? We need tension like this:

pen tension with upward
wire enters with tension____O/ and forward components
|
force

Unless the wire is being pulled upward (tension), it seems
to me that the wire wouldn't wrap around the pen.

Sorry if I'm missing some obvious physics (or the whole
point) here, but I think that you're providing tension by
turning the pen.

Carl Fogel