Spoke Fatigue Study

Michael Wileman said:

Quoted message said:

I don't understand what is so complicated with this stress
picture. Residual stress is a fact when reaching or exceeding yield
in bending, some of it from manufacture, some from wheel building.
This is not new, only the challenge of that concept is new.

Quoted message said:

I don't understand why you are using the word residual. There are
elevated stresses at the stress concentrations as a result of
building the wheel. Stress "relieving" by applying an additional
load to cause local yielding reduces the stress concentration by
changing the shape of the contact. That much is classical
mechanical design.

Quoted message said:

Why are these "residual" stresses before the stress relief instead
of simply stresses resulting from the applied load of lacing the
spoke?

Residual in contrast to tension related stress. This stress is
present even when the spoke is not tensioned in a wheel and is
therefore, called residual.

Quoted message said:

Residual usually refers to stresses that remain after you remove the
load because the part of the component that was deformed elastically
cannot return to its undeformed shape because of resistance from the
part of the component that was deformed plastically. By that
definition, you have more residual stress after you plastically
deform the spoke by stress relieving, even though the total stress
has been reduced.

That is not the case. After the spoke has locally yielded where
stress concentrations remain, the rest stress is lower. That is the
purpose of stress relieving. It yields local high stresses.

Quoted message said:

This is not a criticism. I am just trying to figure out why you
claim that "residual" stresses cause the fatigue failure rather than
the simple tractions of the spoke/rim contact. Can you elucidate?
The material doesn't care where the stresses come from; It fails
based upon the complete tensor, not just the residual stresses.

Because the stress of tension in a wheel is the only one that has
received much attention, the combined stress of spoke tension and
residual stresses from manufacture and wheel building (that are
constants overlayed on tensile stress) and are the cause of spoke
failures, this is an important aspect. Computed tensile stress
derived from known spoke tension is about 1/3 yield stress, not enough
to cause a failure. Therefore, the "hidden" residual stress from
forming comes into play. It is this stress that can be reduced while
tensile stress cannot, since it is the essence of wheel strength (load
carrying ability).

Quoted message said:

I'm convinced that stress relieving prevents fatigue by reducing the
alternating stress at the stress concentrations (where the spoke
touches the rim), just not that it has anything to do with residual
stresses.

It has nothing to do with spoke contact with hub or spoke nipple, but
rather residual stresses from manufacture and wheel building as you
can see from the yielded spoke elbow in the item by Benjamin lewis:

http://www.cs.sfu.ca/~bclewis/personal/bike/spoke/spoke_after.jpeg

Quoted message said:

Are you saying that the spokes break in the straight section
as a result of residual stresses from the wire drawing?

No. I said that they break in the straight sections in tensile tests
where they are forcefully ruptured. Please review what has been
written in this thread. It is too much to repeat here.

Quoted message said:

Having never broken a spoke, I have to admit I don't know where the
fatigue failure usually occurs.

That has also been beaten to death here.

Ride bike and you'll see.

[email hidden]

Benjamin Lewis said:

jim beam said:

Benjamin Lewis said:

What difference does it make if the spokes come from the factory with
residual stress or not? As soon as you tension the spoke in the wheel,
the spoke elbow changes shape.

look at the elbow on a spoke where you have not already bent it, and
look at it again after you've built a wheel with it and the huib hole is
properly indented. they're pretty much identical. the only time
they're different is if the spoke line has been "corrected" in which
case, the angle is too acute.

Oh? Here are two brand new butted dt spokes, after lacing to a front wheel
(Mavic MA-3 rim, Shimano 105 hub), tensioning, and stress-relieving. Care
was taken not to bend the elbows during installation or removal, and no
spoke-line correction was done.

http://www.cs.sfu.ca/~bclewis/personal/bike/spoke/spoke_after.jpeg

The spoke on the left was laced inbound, the right spoke was laced
outbound.

Here is a photoshopped composite:

http://www.cs.sfu.ca/~bclewis/personal/bike/spoke/spoke_difference.jpeg

Red is inbound, green is outbound.

Unfortunately my "before" photos didn't turn out, but it's clear that these
spokes are not the same as each other afterwards.

i've emailed a pic to carl for hosting. it shows a selection of 6
spokes from an old wheel i disassembled a while back - shimano low
flange hub, trek matrix rim. of all 32 spokes, the 3 at the top are the
only ones to show any change in spoke angle. the other 29, of which the
bottom 3 are a selection, all show the factory ~95 degrees.

Quoted message said:

Michael Wileman said:

Quoted message said:

I don't understand what is so complicated with this stress
picture. Residual stress is a fact when reaching or exceeding yield
in bending, some of it from manufacture, some from wheel building.
This is not new, only the challenge of that concept is new.

Quoted message said:

I don't understand why you are using the word residual. There are
elevated stresses at the stress concentrations as a result of
building the wheel. Stress "relieving" by applying an additional
load to cause local yielding reduces the stress concentration by
changing the shape of the contact. That much is classical
mechanical design.

Quoted message said:

Why are these "residual" stresses before the stress relief instead
of simply stresses resulting from the applied load of lacing the
spoke?

Residual in contrast to tension related stress. This stress is
present even when the spoke is not tensioned in a wheel and is
therefore, called residual.

says the guy that thinks springback is evidence of residual stress.

Quoted message said:

Quoted message said:

Residual usually refers to stresses that remain after you remove the
load because the part of the component that was deformed elastically
cannot return to its undeformed shape because of resistance from the
part of the component that was deformed plastically. By that
definition, you have more residual stress after you plastically
deform the spoke by stress relieving, even though the total stress
has been reduced.

That is not the case. After the spoke has locally yielded where
stress concentrations remain, the rest stress is lower. That is the
purpose of stress relieving. It yields local high stresses.

but jobst, metallurgical "stress relief" requires it be applied in a
very short time, i.e. minutes, after the original cold work in order for
it to be properly effective. and it has to be a minimal amount of work
- only 1-2% strain. additional cold work exceeding this degree of
strain or being applied too long after original working just increases
the degree of cold work in the material and can increase the degree of
residual stress.

Quoted message said:

Quoted message said:

This is not a criticism. I am just trying to figure out why you
claim that "residual" stresses cause the fatigue failure rather than
the simple tractions of the spoke/rim contact. Can you elucidate?
The material doesn't care where the stresses come from; It fails
based upon the complete tensor, not just the residual stresses.

Because the stress of tension in a wheel is the only one that has
received much attention, the combined stress of spoke tension and
residual stresses from manufacture and wheel building (that are
constants overlayed on tensile stress) and are the cause of spoke
failures, this is an important aspect.

ok, which side of the spoke elbow is tensile & which compressive?
[let's not get confused about the location of the neutral plane shall
we?] and what is the resolved bending stress on the elbow both inside
and outside?

Quoted message said:

Computed tensile stress
derived from known spoke tension is about 1/3 yield stress, not enough
to cause a failure.

you're ignoring local loading at the elbow...

Quoted message said:

Therefore, the "hidden" residual stress from
forming comes into play. It is this stress that can be reduced while
tensile stress cannot, since it is the essence of wheel strength (load
carrying ability).

that's an obfuscation.

Quoted message said:

Quoted message said:

I'm convinced that stress relieving prevents fatigue by reducing the
alternating stress at the stress concentrations (where the spoke
touches the rim), just not that it has anything to do with residual
stresses.

It has nothing to do with spoke contact with hub or spoke nipple, but
rather residual stresses from manufacture and wheel building as you
can see from the yielded spoke elbow in the item by Benjamin lewis:

http://www.cs.sfu.ca/~bclewis/personal/bike/spoke/spoke_after.jpeg

but the spokes' contact with either the hub or the nipple are the only
way in which any load can be applied!

Quoted message said:

Quoted message said:

Are you saying that the spokes break in the straight section
as a result of residual stresses from the wire drawing?

No. I said that they break in the straight sections in tensile tests
where they are forcefully ruptured.

why not? if as you say, spokes leave the factory with a high degree of
residual stress, particularly the butted section of a spoke, the
straight section of the spoke would be highly susceptible to any
additional load cycling. particularly where the spokes cross and there
is an additional bending component at that juncture.

btw, to show how much you know about residual stress, and whether it's
tensile or compressive in critical areas, why don't you tell the class
the nature of residual stress in straight drawn wire, i.e., you have
circumferential and axial residual stress, a skin & a core. which is
compressive, which tensile?

Quoted message said:

Please review what has been
written in this thread. It is too much to repeat here.

Quoted message said:

Having never broken a spoke, I have to admit I don't know where the
fatigue failure usually occurs.

That has also been beaten to death here.

Ride bike and you'll see.

ah, the anecdotal fallback. sure.

Quoted message said:

[email hidden]

Benjamin Lewis said:

I don't know, but I imagine whatever forms the heads comes in two pieces
like a mold, and these ridges correspond to the gaps between the two
pieces. These ridges are not in consistent places on my spokes (but are
always 180 degrees apart from each other), suggesting that the elbow is
formed after the head had been made.

As someone who has made tooling for 4-slide and multislide machines,
it's *sortof* a mold. Just so you guys get an idea, a four-slide
machine is a cam-operated "press" that deals with thin flatstock or
wirestock, in this case, 304 or any other austenitic stainless steel
wire to make spokes. A four-slide is simply named that way because
there are 4 independent slides.

Here is a picture:

http://www.rdtoolengineering.com/largeviewlwf6a.html

In a 4-slide or multislide machine, you clamp the end of the wire around
its diameter and a "cold heading tool" - a hammer-like swage that has
the profile of the head (in a negative shape) that you want to form.
It's almost a coining operation, where the metal flows to fill the shape
of the cold-heading tool.

The two lines, on each side of the head behind the head correspond to
the seam between each side of the clamp where the metal has flowed into
the rounded edges of the clamp. A clamp that has dead-sharp edges will
show little seam. A clamp that has seen better days will show worn
edges behind the head which will enhance the "molded" effect. The more
wear, the bigger the seam.

A straight-gauge spoke is pretty damn simple. All it is, is a
cold-heading operation, a bending operation, and a cutoff, to be
threaded in another machine later, or transferred in-line to a threading
station, where the threads are rolled (formed) on the opposite end of
the spoke.

Each operation can be done in separate machines (little 1-ton benchtop
presses) and your observation that the bend is independent of the
orientation the clamp marks means that this might really be the case.
--
BMO

Boyle M. Owl said:

Benjamin Lewis said:

I don't know, but I imagine whatever forms the heads comes in two pieces
like a mold, and these ridges correspond to the gaps between the two
pieces. These ridges are not in consistent places on my spokes (but are
always 180 degrees apart from each other), suggesting that the elbow is
formed after the head had been made.

As someone who has made tooling for 4-slide and multislide machines,
it's *sortof* a mold. Just so you guys get an idea, a four-slide
machine is a cam-operated "press" that deals with thin flatstock or
wirestock, in this case, 304 or any other austenitic stainless steel
wire to make spokes. A four-slide is simply named that way because
there are 4 independent slides.

Here is a picture:

http://www.rdtoolengineering.com/largeviewlwf6a.html

In a 4-slide or multislide machine, you clamp the end of the wire around
its diameter and a "cold heading tool" - a hammer-like swage that has
the profile of the head (in a negative shape) that you want to form.
It's almost a coining operation, where the metal flows to fill the shape
of the cold-heading tool.

The two lines, on each side of the head behind the head correspond to
the seam between each side of the clamp where the metal has flowed into
the rounded edges of the clamp. A clamp that has dead-sharp edges will
show little seam. A clamp that has seen better days will show worn
edges behind the head which will enhance the "molded" effect. The more
wear, the bigger the seam.

A straight-gauge spoke is pretty damn simple. All it is, is a
cold-heading operation, a bending operation, and a cutoff, to be
threaded in another machine later, or transferred in-line to a threading
station, where the threads are rolled (formed) on the opposite end of
the spoke.

Each operation can be done in separate machines (little 1-ton benchtop
presses) and your observation that the bend is independent of the
orientation the clamp marks means that this might really be the case.

Dear Boyle and Benjamin,

What you two are saying about the ridges makes sense--whomp,
the head is formed, then the straight piece is taken to the
bending machine without any fuss about the alignment of the
two ridges.

I just got a batch in which the ridges were always a little
off 90 degrees to the shaft and wondered if there was some
subtle engineering trick involved.

Any notion whether all spokes are bent against some sort of
mandrel that leaves a flattened stripe along the inside of
the spoke elbow?

Thanks,

Carl Fogel

Quoted message said:

Dear Boyle and Benjamin,

What you two are saying about the ridges makes sense--whomp,
the head is formed, then the straight piece is taken to the
bending machine without any fuss about the alignment of the
two ridges.

I just got a batch in which the ridges were always a little
off 90 degrees to the shaft and wondered if there was some
subtle engineering trick involved.

Any notion whether all spokes are bent against some sort of
mandrel that leaves a flattened stripe along the inside of
the spoke elbow?

That depends on your definition of mandrel, and the use of that word
means that you're probably envisioning bending the spoke 'round some
sort of pin.

Nah. Not rigid enough if it's automated. Try a die block, with a
radius ground in the corner, and a punch that knocks the head over,
spaced 1 wire's width plus tolerance away from the block at full close.
I have never seen any overbend in a spoke, so I guess it's just an
air-bend. There are ways to overbend, to compensate for springback, but
a spoke is not a super-precision item.

I would grind a groove in the punch and radius it on the leading edge to
be more gentle and keep the elbow from wandering.

This can be done with a DeStaCo hand bender, if you can find one, which
actually *would* bend the radius around a mandrel. They're not made
anymore (automation, you know), but you never know what you can find in
a yard sale.

Here's approximately what it would look like:

http://www.wintonmachine.com/bmate.htm

It's for bending small diameter tubing, but that's about the same as
bending 15 ga. 303 stainless wire. A DeStaCo bender is very similar in
form and function.

There was a guy having a yard sale a few houses down from me. I noticed
he had a couple of antique kick presses in his garage...They weren't for
sale, dammit. He actually used them for hobby stuff and small-run
boutique items. If I had one, I would whip up some tooling to make my
own spokes.

--
BMO

jim beam said:

Benjamin Lewis said:

jim beam said:

Benjamin Lewis wrote:

> What difference does it make if the spokes come from the factory with
> residual stress or not? As soon as you tension the spoke in the wheel,
> the spoke elbow changes shape.
>

look at the elbow on a spoke where you have not already bent it, and
look at it again after you've built a wheel with it and the huib hole
is properly indented. they're pretty much identical. the only time
they're different is if the spoke line has been "corrected" in which
case, the angle is too acute.

Oh? Here are two brand new butted dt spokes, after lacing to a front
wheel
(Mavic MA-3 rim, Shimano 105 hub), tensioning, and stress-relieving.
Care
was taken not to bend the elbows during installation or removal, and no
spoke-line correction was done.

http://www.cs.sfu.ca/~bclewis/personal/bike/spoke/spoke_after.jpeg

The spoke on the left was laced inbound, the right spoke was laced
outbound.

Here is a photoshopped composite:

http://www.cs.sfu.ca/~bclewis/personal/bike/spoke/spoke_difference.jpeg

Red is inbound, green is outbound.

Unfortunately my "before" photos didn't turn out, but it's clear that
these
spokes are not the same as each other afterwards.

i've emailed a pic to carl for hosting. it shows a selection of 6
spokes from an old wheel i disassembled a while back - shimano low
flange hub, trek matrix rim. of all 32 spokes, the 3 at the top are the
only ones to show any change in spoke angle. the other 29, of which the
bottom 3 are a selection, all show the factory ~95 degrees.

http://home.comcast.net/~carlfogel/download/Img_0267.jpg

jim beam said:

jim beam said:

Benjamin Lewis said:

jim beam wrote:
> Benjamin Lewis wrote:
>
>> What difference does it make if the spokes come from the factory with
>> residual stress or not? As soon as you tension the spoke in the
>> wheel, the spoke elbow changes shape.
>
> look at the elbow on a spoke where you have not already bent it, and
> look at it again after you've built a wheel with it and the huib hole
> is properly indented. they're pretty much identical. the only time
> they're different is if the spoke line has been "corrected" in which
> case, the angle is too acute.

Oh? Here are two brand new butted dt spokes, after lacing to a front
wheel (Mavic MA-3 rim, Shimano 105 hub), tensioning, and
stress-relieving. Care was taken not to bend the elbows during
installation or removal, and no spoke-line correction was done.

http://www.cs.sfu.ca/~bclewis/personal/bike/spoke/spoke_after.jpeg

The spoke on the left was laced inbound, the right spoke was laced
outbound.

Here is a photoshopped composite:

http://www.cs.sfu.ca/~bclewis/personal/bike/spoke/spoke_difference.jpeg

Red is inbound, green is outbound.

Unfortunately my "before" photos didn't turn out, but it's clear that
these spokes are not the same as each other afterwards.

i've emailed a pic to carl for hosting. it shows a selection of 6
spokes from an old wheel i disassembled a while back - shimano low
flange hub, trek matrix rim. of all 32 spokes, the 3 at the top are the
only ones to show any change in spoke angle. the other 29, of which the
bottom 3 are a selection, all show the factory ~95 degrees.

http://home.comcast.net/~carlfogel/download/Img_0267.jpg

I'm not sure what you think this photograph proves.

--
Benjamin Lewis

Seeing is deceiving. It's eating that's believing.
-- James Thurber

On 21 Jun 2005 15:38:37 -0700, "41" <[email hidden]>

Quoted message said:

Jim Smith said:

"41" <[email hidden]> writes:

[snip]

Quoted message said:

Quoted message said:

An additional
complication is the major shift in spoke material during this same
time period.

No, not for the people who reported here. Usenet has not been active
that long and JB's book was published in the early 1980s.

[snip]

Dear 41,

Jobst himself commented on a major improvement in spoke
"durability" within a decade of the 1st edition of his book:

"It appears that the better spokes now available would have
made the discovery of many of the concepts of this book more
difficult for lack of failure data. I am grateful in
retrospect for the poor durability of earlier spokes. They
operated so near their limits that durability was
significantly altered by the techniques that I have
outlined."

--Jobst Brandt, "The Bicycle Wheel," 3rd Edition, 1993,
p.124

Carl Fogel

Quoted message said:

Jobst himself commented on a major improv ement in spoke
"durability" within a decade of the 1st edition of his book:

He commented within a decade but as he noted, this was in retrospect.
Improved spokes were current from the 1980s on, at least: since the
introduction of DT stainless steel, and others. In any case, I
referenced people who noted the improvement with the same components
before and after.a

Benjamin Lewis said:

jim beam said:

jim beam said:

Benjamin Lewis wrote:

>jim beam wrote:
>
>>Benjamin Lewis wrote:
>>
>>
>>>What difference does it make if the spokes come from the factory with
>>>residual stress or not? As soon as you tension the spoke in the
>>>wheel, the spoke elbow changes shape.
>>
>>look at the elbow on a spoke where you have not already bent it, and
>>look at it again after you've built a wheel with it and the huib hole
>>is properly indented. they're pretty much identical. the only time
>>they're different is if the spoke line has been "corrected" in which
>>case, the angle is too acute.
>
>Oh? Here are two brand new butted dt spokes, after lacing to a front
>wheel (Mavic MA-3 rim, Shimano 105 hub), tensioning, and
>stress-relieving. Care was taken not to bend the elbows during
>installation or removal, and no spoke-line correction was done.
>
>http://www.cs.sfu.ca/~bclewis/personal/bike/spoke/spoke_after.jpeg
>
>The spoke on the left was laced inbound, the right spoke was laced
>outbound.
>
>Here is a photoshopped composite:
>
>http://www.cs.sfu.ca/~bclewis/personal/bike/spoke/spoke_difference.jpeg
>
>Red is inbound, green is outbound.
>
>Unfortunately my "before" photos didn't turn out, but it's clear that
>these spokes are not the same as each other afterwards.
>

i've emailed a pic to carl for hosting. it shows a selection of 6
spokes from an old wheel i disassembled a while back - shimano low
flange hub, trek matrix rim. of all 32 spokes, the 3 at the top are the
only ones to show any change in spoke angle. the other 29, of which the
bottom 3 are a selection, all show the factory ~95 degrees.

http://home.comcast.net/~carlfogel/download/Img_0267.jpg

I'm not sure what you think this photograph proves.

i guess it proves that i can't be bothered to photograph 32 individual
spokes when 29 are the same.

jim beam wrote:

[not much]

Hey, since you can't salvage anything now, here's a better idea: slink
off for about six months, and then come back as "jack daniels". If
you're lucky, no one will remember you!

Quoted message said:

Benjamin Lewis said:

http://www.cs.sfu.ca/~bclewis/personal/bike/spoke/spoke_after.jpeg

Quoted message said:

Well! "In the presence of pictorial evidence of residual stress, I
think I hear my mother calling to dinner. I gotta go..."

I tried to post this question a few days ago, but it must have ended up
in a bit bucket somewhere...

I don't understand how these photos prove residual stress, and I don't
see how they prove that the spoke squeezing process relieves that
stress. Can you fill in the gaps that I'm missing? Thanks.

--
Dave
dvt at psu dot edu

Dave who? writes:

http://www.cs.sfu.ca/~bclewis/personal/bike/spoke/spoke_after.jpeg

Quoted message said:

Quoted message said:

Well! "In the presence of pictorial evidence of residual stress, I
think I hear my mother calling to dinner. I gotta go..."

Quoted message said:

I tried to post this question a few days ago, but it must have ended
up in a bit bucket somewhere...

Quoted message said:

I don't understand how these photos prove residual stress, and I
don't see how they prove that the spoke squeezing process relieves
that stress. Can you fill in the gaps that I'm missing? Thanks.

What the picture shows is that the elbow shape has been permanently
changed by what is called yielding. Yield stress is an unacceptably
high for spokes that are to operate under millions of stress cycles
and it will cause failure. That this stress exists and that it can be
relaxed mechanically is the basis of this thread. The stress is a
residual stress from building in contrast to tension stress that is
readily determined by tensiometer measurement.

If the spoke elbow plastically deforms, as the picture shows, then it
reached yield stress. Subsequent tensioning of the spoke guarantees
that the outside of the elbow, that changed from an obtuse angle to an
acute angle, remains at yield when wheel building is complete. Only
if the spoke is stress relieved by substantial over tensioning can
yield stress in this area be lowered. By increasing tension, the area
at yield will yield while the parts of that cross section that are not
will respond elastically. When the overload is relaxed, the entire
cross section reverts to a lower stress than it was under the
overload; the yielded portion coming down as much as the remainder of
the cross section.

Of course we have been here many times in the last 20 years, mostly
under attack by dissenters from GB. This is not new and is explained
in detail in "the Bicycle Wheel".

[email hidden]

Quoted message said:

If the spoke elbow plastically deforms, as the picture shows, then it
reached yield stress. Subsequent tensioning of the spoke guarantees
that the outside of the elbow, that changed from an obtuse angle to an
acute angle, remains at yield when wheel building is complete. Only
if the spoke is stress relieved by substantial over tensioning can
yield stress in this area be lowered. By increasing tension, the area
at yield will yield while the parts of that cross section that are not
will respond elastically. When the overload is relaxed, the entire
cross section reverts to a lower stress than it was under the
overload; the yielded portion coming down as much as the remainder of
the cross section.

I'm with you up to here (I have a very basic understanding of yield,
fatigue, Soderberg diagrams, etc). But here's what I don't understand.
You've yielded the spoke by overloading (squeezing the spokes). Now that
the spoke has yielded, it must have lower tension than before. If you
want balanced tension in the wheel, you must bring the tension back up
by turning the nipple. How do you know that this new tension doesn't
bring that overstressed region back near the yield stress?

--
Dave
dvt at psu dot edu

Dave vt? said:

Quoted message said:

If the spoke elbow plastically deforms, as the picture shows, then
it reached yield stress. Subsequent tensioning of the spoke
guarantees that the outside of the elbow, that changed from an
obtuse angle to an acute angle, remains at yield when wheel
building is complete. Only if the spoke is stress relieved by
substantial over tensioning can yield stress in this area be
lowered. By increasing tension, the area at yield will yield while
the parts of that cross section that are not will respond
elastically. When the overload is relaxed, the entire cross
section reverts to a lower stress than it was under the overload;
the yielded portion coming down as much as the remainder of the
cross section.

Quoted message said:

I'm with you up to here (I have a very basic understanding of yield,
fatigue, Soderberg diagrams, etc). But here's what I don't
understand. You've yielded the spoke by overloading (squeezing the
spokes). Now that the spoke has yielded, it must have lower tension
than before. If you want balanced tension in the wheel, you must
bring the tension back up by turning the nipple. How do you know
that this new tension doesn't bring that overstressed region back
near the yield stress?

Where is the residual stress going to come from now that it has been
relieved? As I mentioned in an earlier response, the elongation
necessary to relieve residual stress is in the micro-inches and has no
significance to spoke tension. For practical purposes, tension is not
affected whether a house fly sits on the spoke or not. That's about the
size of it.

[email hidden]

In article said:

Now that the spoke has yielded, it must have lower tension than
before. If you want balanced tension in the wheel, you must bring the
tension back up by turning the nipple. How do you know that this new
tension doesn't bring that overstressed region back near the yield stress?

The elbow will stretch ever so slightly with the yield, but
relative to the elasticity present through 300mm worth of spoke, it's
not enough to make a real effect on the net tension.

It's an easy matter to imagine what's going on using a graph of
the stress vs. position.

................... Tensile yield

Tension^
| /\ /
| / \ /
---/----\----/-- neutral stress
| / \ /
Copression v / \/

Outside Inside of Bend
of bend

As shown, this would be for a spoke out of a box (not to
scale, peaks may not be in proportion, and other disclaimers apply )
Applying tension to the spoke will shift the graph upwards, with the
total area under the graph (with appropriate weighting for sectional
geometry ) equalling the applied tension. Adding enough tension to
stress relieve the spoke will mash the peaks of this into the tensile
yield ceiling, and after the tension is relaxed, those peaks are
flattened out.

Yes, flattening those peaks does reduce the net tension IF
the stretch of the elbow is returned to where the unyielded material
is at the same strain as it was before stress relieving. The spoke
elbow isn't in a constant strain environment though, the other 300mm of
spoke having elasticity to make it look more like a constant
tension.

To return the elbow to its original strain (for the unyielded
material) you'd need to actually back off on the spoke nipple first so
that the tension on the rest of the spoke matches this loss in the
elbow. You can imagine adding an incremental turn of the nipple to add
tension back, but that really amounts to just undoing this backing-off
(plus a tiny tiny bit more, small enough to be immaterial).

With adding that tension back, what happens is that the entire
graph - not just the peaks - rises. Again, the total area under the
graph is the tension of the spoke.

The amount by which the graph as a whole rises is less
than the amount by which the maximum amount the peaks were reduced. If
you like, you can imagine the final graph as chopping the peaks off
the original graph, then taking the area contained in those peaks, and
sprinkling it uniformly over the entire graph.

What stress relieving amounts to, is taking tension from the
most stressed regions of the spoke, and spreading it across the entire
section of the spoke. The less stressed material takes on more tension
so that the peaks can deal with less.

Steal from the rich, give to the poor.

-Luns

Quoted message said:

Dave who? writes:

http://www.cs.sfu.ca/~bclewis/personal/bike/spoke/spoke_after.jpeg

Quoted message said:

Quoted message said:

Well! "In the presence of pictorial evidence of residual stress, I
think I hear my mother calling to dinner. I gotta go..."

Quoted message said:

I tried to post this question a few days ago, but it must have ended
up in a bit bucket somewhere...

Quoted message said:

I don't understand how these photos prove residual stress, and I
don't see how they prove that the spoke squeezing process relieves
that stress. Can you fill in the gaps that I'm missing? Thanks.

What the picture shows is that the elbow shape has been permanently
changed by what is called yielding.

true

Quoted message said:

Yield stress is an unacceptably
high for spokes that are to operate under millions of stress cycles
and it will cause failure.

logical disconnect. that it is bent does /not/ mean that bending is a
result of service tension. it's an assumption that is convenient to
your theory, but the fact remains that it's hard to build a wheel
without bending spokes on assembly, but this has nothing to do with
their stress in service.

Quoted message said:

That this stress exists and that it can be
relaxed mechanically is the basis of this thread.

mechanical stress is not residual stress. another logical disconnect.

Quoted message said:

The stress is a
residual stress from building in contrast to tension stress that is
readily determined by tensiometer measurement.

rubbish garnished with a grain of truth. mechanical tension can be
measured with a tensiometer, but it's mere supposition that residual
stress is the result of building - unless you have "corrected the spoke
line" in building of course, something that deforms spoke elbows and is
contrary to spoke manufacturer's build instructions.

Quoted message said:

If the spoke elbow plastically deforms, as the picture shows, then it
reached yield stress.

/when/ did it yield jobst?

Quoted message said:

Subsequent tensioning of the spoke guarantees
that the outside of the elbow, that changed from an obtuse angle to an
acute angle, remains at yield when wheel building is complete.

spoke tension alone does not change the angle. see
http://home.comcast.net/~carlfogel/download/Img_0267.jpg

out of 32 spokes, only 3 are bent - that can /only/ be a function of
build, not tension.

Quoted message said:

Only
if the spoke is stress relieved by substantial over tensioning can
yield stress in this area be lowered.

no proof for that contention. small deformations that do not
immediately follow initial forming can lead to higher residual stress,
not lower.

Quoted message said:

By increasing tension, the area
at yield will yield while the parts of that cross section that are not
will respond elastically.

eh?

Quoted message said:

When the overload is relaxed, the entire
cross section reverts to a lower stress than it was under the
overload; the yielded portion coming down as much as the remainder of
the cross section.

assuming stress relief is effective, but for the reasons cited above,
you can't make that assumption.

Quoted message said:

Of course we have been here many times in the last 20 years, mostly
under attack by dissenters from GB. This is not new and is explained
in detail in "the Bicycle Wheel".

translation: "i've been able to stick to this story for 20 years because
i've never paid attention to fact nor done any technical research in
case it would cause me to contradict myself."

Quoted message said:

[email hidden]

Quoted message said:

Dave vt? said:

Quoted message said:

If the spoke elbow plastically deforms, as the picture shows, then
it reached yield stress. Subsequent tensioning of the spoke
guarantees that the outside of the elbow, that changed from an
obtuse angle to an acute angle, remains at yield when wheel
building is complete. Only if the spoke is stress relieved by
substantial over tensioning can yield stress in this area be
lowered. By increasing tension, the area at yield will yield while
the parts of that cross section that are not will respond
elastically. When the overload is relaxed, the entire cross
section reverts to a lower stress than it was under the overload;
the yielded portion coming down as much as the remainder of the
cross section.

Quoted message said:

I'm with you up to here (I have a very basic understanding of yield,
fatigue, Soderberg diagrams, etc). But here's what I don't
understand. You've yielded the spoke by overloading (squeezing the
spokes). Now that the spoke has yielded, it must have lower tension
than before. If you want balanced tension in the wheel, you must
bring the tension back up by turning the nipple. How do you know
that this new tension doesn't bring that overstressed region back
near the yield stress?

Where is the residual stress going to come from now that it has been
relieved? As I mentioned in an earlier response, the elongation
necessary to relieve residual stress is in the micro-inches and has no
significance to spoke tension. For practical purposes, tension is not
affected whether a house fly sits on the spoke or not. That's about the
size of it.

translation: i can't prove it [sic] - you'll just have to take my word
for it. funny how gross yielding sufficient to visibly bend a spoke is
now micro-inches.

Quoted message said:

[email hidden]

jim beam said:

Quoted message said:

Yield stress is an unacceptably
high for spokes that are to operate under millions of stress cycles
and it will cause failure.

logical disconnect. that it is bent does /not/ mean that bending is a
result of service tension. it's an assumption that is convenient to
your theory, but the fact remains that it's hard to build a wheel
without bending spokes on assembly, but this has nothing to do with
their stress in service.

So now I'm lying, am I? I mentioned in the original post that I took
special care *not* to bend the elbows during lacing. In fact, this is very
easy to accomplish. It may be difficult not to bend the rest of the spoke,
but that is irrelevant.

Quoted message said:

Quoted message said:

If the spoke elbow plastically deforms, as the picture shows, then it
reached yield stress.

/when/ did it yield jobst?

It yielded either during tensioning or during stress-relief. Having done
it myself, I am absolutely 100% certain of that.

Quoted message said:

Quoted message said:

Subsequent tensioning of the spoke guarantees
that the outside of the elbow, that changed from an obtuse angle to an
acute angle, remains at yield when wheel building is complete.

spoke tension alone does not change the angle. see
http://home.comcast.net/~carlfogel/download/Img_0267.jpg

out of 32 spokes, only 3 are bent - that can /only/ be a function of
build, not tension.

Absolutely and completely false, and I invite anyone else who does not
believe me to try the same experiment for themselves.

I can think of two explanations offhand for your spoke elbows not bending
-- the hub was already perfectly flush with them (seems improbable to me),
or you did not apply enough tension to bring them to yield.

--
Benjamin Lewis

"Love is a snowmobile racing across the tundra and then suddenly it flips
over, pinning you underneath. At night, the ice weasels come."
--Matt Groening