spoke fatigue troll

Ben C

On 2008-05-01, Peter Cole <peter_cole@verizon.net> wrote:
> Ben C wrote:
>
>> OK, well I think I understand what you're saying now. But as the hole
>> deforms, the spoke is still touching the sides of the hole all the way
>> to the edge, so it's not clear that it's bending about any point except
>> where it leaves the hole.
>
> I was thinking of the case where stress relief deformed the hole
> additionally. If (and I don't know) the spoke bent slightly (in the
> hole) during stress relief, then it might spring back enough to increase
> the unsupported length.

Yes indeed, it might do that. I hadn't thought of that.

There's a sort of an equilibrium-- if there's a lot of moment then the
spoke bends easily as you overload it, and probably doesn't crush the
hub much. As it bends though and gets closer to the flange, the moment
reduces and it starts to cut into the hub more. By the time it's really
cutting into the hub, the spoke might actually be _unbending_ as you
continue to pull it. Its springback from that would then actually press
it against its support instead of pushing it away from it.

I don't know what actually happens.

> On the other hand, the crushing could deform the bore down into the
> countersink, decreasing the unsupported length. It seems like
> splitting hairs, but according to Peter White, when DT changed the
> elbow length by that kind of length (0.6mm) there were lots of spoke
> failures.

Yes, that's why I think moment is so important.

>> A herring I raised in a thread long ago was about how well supported the
>> spoke was in the hub hole. If the hub hole doesn't deform much at all
>> (steel hub for example) then the edge of it is probably digging into the
>> spoke somewhere and that's where all the force is concentrated. If
>> they're deformed together nicely then the situation is a little bit more
>> like a cable wrapped around a cylinder. You might get some force going
>> all the way to the head, but in any case, the force is spread over a
>> greater area, so less stress.
>>
>> I was speculating that this might be a benefit of putting oil on the
>> elbows as Gene does.
>
> I think we're pretty safe in assuming that the aluminum hub doesn't cut
> the steel spoke.

I was thinking more of pulling a stiff wire around a mandrel. Since
you're pulling it quite tight there could be stiction meaning it bends
in slightly the wrong places at times and you don't end up with the wire
in such a smooth curve, and therefore not so well supported.

I think Beam's suggestion is more plausible though: that the oil
prevents corrosion especially as it's quite salty in FL.

> That said, while thinking about this I realized that
> the shear stress is the same magnitude as the tensile stress. Things get
> complicated since a spoke has such a tight bend (radius of curvature
><< less than 10x diameter, the figure of merit for simple beam
> deflection). From what I've read, stainless (304), unthreaded fasteners
> are supposed to have a maximum safe shear stress of 185MPa, while 100kgf
> spokes are at 250MPa by my calculations. As for fatigue, you've got
> tensile, bending, shear and residuals all in a region that's supported
> in a kind of indeterminate way, the dimensions making it partly like a
> beam, partly not. The literature I've been able to find seems to
> conclude that fatigue in multiaxis loading is difficult to predict.
>
> It occurred to me that maybe the flattening you see on the insides of
> some spoke elbows might be a good thing.

Typically with steel hubs I think.

> If you "ovalized" the elbow
> bend even more, you'd lower the skin stress while not affecting the
> shear or tensile stress.
>
> Bottom line is that I think it's simplistic to say that because of spoke
> geometry all failures should happen at a certain spot, there's just too
> much going on.

Yes I agree with that.

> I think it's fair to say that stress relief will reduce static &
> residual tensile stresses and tend to improve things no matter what
> they really are. I can't see a way of supporting the "bedding in"
> argument (as reducing spoke failures, not spoke slacking).

I think Beam says it's mostly about spoke slacking since he tried not
doing it and rode around the block and his wheel went loose.

Theoretically that shouldn't happen since spoke tension isn't ever
supposed to increase in use, so riding around the block shouldn't bed
anything in.

I see it as basically spoke line correction, by bending the spoke and/or
deforming the hub. I'm not sure it's "microscopic" spoke line
correction, but probably reduces the perpendicular distance from the
spoke line to the support point by of the order of about 0.5mm or so.

For some hubs/spokes it's the only spoke line correction you need to do,
viz daveornee's and Beam's pictures, and it doesn't appear to bend the
spokes much because if you take them out again the elbows haven't
changed their angle much.

I found when I built with a Shimano hub that I didn't really need
to do any other spoke line correction-- the spokes kind of sat right
very easily.

But on the wheels I built with Campag hubs the spokes really looked
off-line so I did bend them in a bit with my thumb before full tension.
I'm sure if you took those wheels apart you'd be able to tell from the
elbow bends which were the inbound and which the outbound spokes.

[...]
>>> What Keith Bontrager said (over 9 years ago on rbt):
>>> http://tinyurl.com/6sx5nz
>> [...]
>>
>> Great link, thanks!
>
> It's amazing (and sad) to read the old archives sometimes. Civility and
> content has really tanked. Tragedy of the commons, I guess.

Well there's nothing to be done except try to bring standards back up
again, which you for one are certainly doing.

Peter Cole

Ben C wrote:
> On 2008-05-01, Peter Cole <peter_cole@verizon.net> wrote:

>> I think it's fair to say that stress relief will reduce static &
>> residual tensile stresses and tend to improve things no matter what
>> they really are. I can't see a way of supporting the "bedding in"
>> argument (as reducing spoke failures, not spoke slacking).
>
> I think Beam says it's mostly about spoke slacking since he tried not
> doing it and rode around the block and his wheel went loose.
>
> Theoretically that shouldn't happen since spoke tension isn't ever
> supposed to increase in use, so riding around the block shouldn't bed
> anything in.

I agree that stress relieving can cause spokes to lose tension. I think
that riding can do the same thing if the lateral loads (flopping the
bike) increase tension. Bowed spoke lines could also do it if they took
a little more permanent set. Since full spoke tension is 0.75mm, even a
tenth of hub deformation or a couple of degrees of spoke straightening
would make a significant change.

jobst.brandt@stanfordalumni.org

Stress relief.

I think deformation of flange holes in aluminum hubs is being
mis-characterized, leading to incorrect assumptions about effective
spoke hole position before and after stress relieving and that
trueness of the wheel suffers from the process.

Hole deformation is an asymptotic effect that with reasonable spoke
tension is already as deep as it will get. If that were not so,
spokes would gradually sink through the flange and pull out. Once
about a third of the spoke diameter bears fully on the flange hole it
is as deep as it will go while subsequent stress relief cannot "bed
them in" any deeper as the process is often depicted here. The terms
bedding in or stabilizing are a misnomers chosen by people who cannot
visualize mechanical stress relief or that spokes bed in naturally
from initial spoke tension.

I'm getting deja vu all over again, as Yogi Berra so aptly said, with
technical descriptions of wheel building. I sense that writers are
trying to return to pre 1980 wheel mysteries, when wheels could be
built only by people who spent their life doing so, for reasons that
remained secret. Beyond that, today it is all about materials that,
although claimed to be superior, perform more poorly.

Jobst Brandt

carlfogel@comcast.net

On 01 May 2008 22:38:24 GMT, jobst.brandt@stanfordalumni.org wrote:

[snip]

>Beyond that, today it is all about materials that,
>although claimed to be superior, perform more poorly.
>
>Jobst Brandt

Dear Jobst,

The last I heard, something changed between the 1st and 3rd editions
of "The Bicycle Wheel" and significantly altered the durability of
spokes:

"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

Cheers,

Carl Fogel

jobst.brandt@stanfordalumni.org

Carl Fogel wrote:

>> Beyond that, today it is all about materials that, although claimed
>> to be superior, perform more poorly.

> The last I heard, something changed between the 1st and 3rd editions
> of "The Bicycle Wheel" and significantly altered the durability of
> spokes:

You're trying too hard. The new materials are Kevlar and Carbon fiber
along with hard anodized rims and hubs. Because the spokes of old
required careful spoke alignment and stress relief, Today some spokes
survive in spite of less skilled attention, but that could also be
that the wheels don't get as much use as the ones of yore.

> "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."

That has more to do with assessing how effective stress relief is than
with how long spokes last before failure. As I have often mentioned,
spokes in my old wheels from the 1970's on which I developed the
methods, have more than 300,000 miles service at 10,000 miles per year
or more.

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

I'm glad you have a copy of the book. It still sells well today.

Jobst Brandt

clare at snyder dot ontario dot canada

On Wed, 30 Apr 2008 22:34:16 -0500, Tom Sherman
<sunsetss0003@REMOVETHISyahoo.com> wrote:

>Peter Cole wrote:
>> Ben C wrote:
>>> On 2008-04-30, Peter Cole <peter_cole@verizon.net> wrote:
>> [...]
>>>> One of my cars eats head gaskets, the other one doesn't.
>>
>>> I share the dim view. Next time get a Japanese car
>>
>> I've had those, and German, and Swedish, and Italian. US doesn't have a
>> monopoly on bad engineering, although they may have the lead.
>> [...]
>
>Nonsense. Buy a vintage British car, and you will gain understanding.


Yea, if it doesn't leak it doesn't contain oil. If it doesn't flicker
or spark it's not connected. If it isn't broken it was never
installed!!
My 1961 Mini 850 was the most dependable car I ever owned. I could
depend on it to make trouble just about any time - and REAL trouble at
the worst of times.

Austin A65 wasn't much better - I got that one for a friend (ouch).The
1972 Rover TC was a nice car- and relatively reliable(for a British
car) but was a REAL PIG to work on.(it was my brother's car) He also
had a Vauxhaul HA and a Victor Special. I had an HC (Firenza (or
Magnum))( Any F'renza yours ain't no f'renza mine). The Firenza was
actually a pretty decent car, considering I bought it for, IIRC, $250
when it was 4 years old!! (6 years later I got $700 for it)
** Posted from http://www.teranews.com **

Ben C

On 2008-05-01, jobst.brandt@stanfordalumni.org <jobst.brandt@stanfordalumni.org> wrote:
> Stress relief.
>
> I think deformation of flange holes in aluminum hubs is being
> mis-characterized, leading to incorrect assumptions about effective
> spoke hole position before and after stress relieving and that
> trueness of the wheel suffers from the process.
>
> Hole deformation is an asymptotic effect that with reasonable spoke
> tension is already as deep as it will get. If that were not so,
> spokes would gradually sink through the flange and pull out. Once
> about a third of the spoke diameter bears fully on the flange hole it
> is as deep as it will go while subsequent stress relief cannot "bed
> them in" any deeper as the process is often depicted here.

I am familiar with that theory which you and Peter Cole have explained
here before.

On that basis I thought stabilization quite likely worked predominantly
by deforming the spoke and not the hub.

Jim Beam had shown pictures earlier that a spoke removed from a finished
wheel doesn't necessarily show a big change in the elbow angle and that
the hub hole can get significantly deformed.

Here are his spokes:
http://www.flickr.com/photos/38636024@N00/331112190/
And this is the hub they came out of:
http://www.flickr.com/photos/38636024@N00/104463818/

But there's nothing there to suggest definitively that the hub hole
deformation didn't happen during the initial stages of tensioning rather
than during stabilization.

What has re-opened this discussion are the pictures daveornee posted.

http://www.flickr.com/photos/17085834@N08/2443679334/
http://www.flickr.com/photos/17085834@N08/2443679236/

In particular:

http://www.flickr.com/photos/17085834@N08/2442851273/
http://www.flickr.com/photos/17085834@N08/2443679416/

Seems to show more hub hole deformation and less spoke deformation when
compared with:

http://www.flickr.com/photos/17085834@N08/2442851411/
http://www.flickr.com/photos/17085834@N08/2442851349/

This implies to me strongly that stabilization can improve spoke line by
deforming the hub and not the spoke.

Otherwise why is the hub deformed more on the side on which spoke line
was not corrected before stabilization? During the initial stages of
tensioning (before spoke line correction on the side that was), both
sides of the hub were treated exactly the same. If all hub deformation
occurs during that phase, we would expect to see the same amount of
deformation on both sides.

A Muzi

>>>> Peter Cole <peter_cole@verizon.net> wrote:
>>>>> One of my cars eats head gaskets, the other one doesn't.

>>> Ben C wrote:
>>>> I share the dim view. Next time get a Japanese car

>> Peter Cole wrote:
>>> I've had those, and German, and Swedish, and Italian. US doesn't have a
>>> monopoly on bad engineering, although they may have the lead.
>>> [...]

> Tom Sherman <sunsetss0003@REMOVETHISyahoo.com> wrote:
>> Nonsense. Buy a vintage British car, and you will gain understanding.

clare at snyder dot ontario dot canada wrote:
> Yea, if it doesn't leak it doesn't contain oil. If it doesn't flicker
> or spark it's not connected. If it isn't broken it was never
> installed!!
> My 1961 Mini 850 was the most dependable car I ever owned. I could
> depend on it to make trouble just about any time - and REAL trouble at
> the worst of times.
>
> Austin A65 wasn't much better - I got that one for a friend (ouch).The
> 1972 Rover TC was a nice car- and relatively reliable(for a British
> car) but was a REAL PIG to work on.(it was my brother's car) He also
> had a Vauxhaul HA and a Victor Special. I had an HC (Firenza (or
> Magnum))( Any F'renza yours ain't no f'renza mine). The Firenza was
> actually a pretty decent car, considering I bought it for, IIRC, $250
> when it was 4 years old!! (6 years later I got $700 for it)

Wow. Sorta like second marriages, multiple British cars must be the
triumph of hope over experience. One was plenty for me.

(didn't keep me from subjecting myself to a Fiat later though)
--
Andrew Muzi
<www.yellowjersey.org/>
Open every day since 1 April, 1971
** Posted from http://www.teranews.com **

Clive George

"A Muzi" <am@yellowjersey.org> wrote in message
news:c33fe$481b382e$4677@news.teranews.com...

> Wow. Sorta like second marriages, multiple British cars must be the
> triumph of hope over experience. One was plenty for me.

As was my 3 weeks of experience of a USian car. That thing was pretty
dreadful - and a 2008 model at that.

Though I have no desire to spend time mucking around with _old_ British
cars, I'd be happy with a lot of currently-built stuff if I didn't have a
peculiar desire for a car which goes up and down.

cheers,
clive

carlfogel@comcast.net

On 02 May 2008 00:04:51 GMT, jobst.brandt@stanfordalumni.org wrote:

>Carl Fogel wrote:
>
>>> Beyond that, today it is all about materials that, although claimed
>>> to be superior, perform more poorly.
>
>> The last I heard, something changed between the 1st and 3rd editions
>> of "The Bicycle Wheel" and significantly altered the durability of
>> spokes:
>
>You're trying too hard. The new materials are Kevlar and Carbon fiber
>along with hard anodized rims and hubs. Because the spokes of old
>required careful spoke alignment and stress relief, Today some spokes
>survive in spite of less skilled attention, but that could also be
>that the wheels don't get as much use as the ones of yore.
>
>> "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."
>
>That has more to do with assessing how effective stress relief is than
>with how long spokes last before failure. As I have often mentioned,
>spokes in my old wheels from the 1970's on which I developed the
>methods, have more than 300,000 miles service at 10,000 miles per year
>or more.
>
>> --Jobst Brandt, "The Bicycle Wheel," 3rd Edition, 1993, p.124
>
>I'm glad you have a copy of the book. It still sells well today.
>
>Jobst Brandt

Dear Jobst,

I sense that you're not trying hard enough.

Someone or other tested spokes in 1981 and again 1988. He published
stress-strain graphs, showing the results.

In 1981, he tested carbon steel and stainless steel spokes. They all
pulled apart, failing after stretching less than 0.15" (3.8 mm).

By 1988, only seven years later, he didn't bother testing carbon steel
spokes because most riders had stopped using the old material by 1988
and embraced the new stainless steel material for some reason or
other.

Interestingly, some stainless steel spokes from the same company (DT)
stretched so far in 1988 that the tester gave up trying to make them
fail in his second test. The stress-strain graph just goes level.

The 1981 stainless steel spokes all failed below 4 mm of stretch. The
1988 spokes stretch 25% to 50% more, failing at 5 mm to 6 mm, or even
showing no sign of failure.

It's as if the spoke material changed significantly.

That might explain the tester's comment five years later in 1993 that
spokes had become significantly more durable.

He never shared whatever "data" he had in mind about durability, but
his claim for an impressive improvement in spoke durability is
supported by the fact that the practice of carrying spare spokes
practically vanished around that time.

Cheers,

Carl Fogel

Helmut Springer

A Muzi <am@yellowjersey.org> wrote:
> Wow. Sorta like second marriages, multiple British cars must be
> the triumph of hope over experience. One was plenty for me.

Going for a Mini nowadays brings you BMW technology...


--
MfG/Best regards
helmut springer panta rhei

Peter Cole

jobst.brandt@stanfordalumni.org wrote:
> Stress relief.
>
> I think deformation of flange holes in aluminum hubs is being
> mis-characterized, leading to incorrect assumptions about effective
> spoke hole position before and after stress relieving and that
> trueness of the wheel suffers from the process.
>
> Hole deformation is an asymptotic effect that with reasonable spoke
> tension is already as deep as it will get. If that were not so,
> spokes would gradually sink through the flange and pull out. Once
> about a third of the spoke diameter bears fully on the flange hole it
> is as deep as it will go while subsequent stress relief cannot "bed
> them in" any deeper as the process is often depicted here. The terms
> bedding in or stabilizing are a misnomers chosen by people who cannot
> visualize mechanical stress relief or that spokes bed in naturally
> from initial spoke tension.

The Rockwell hardness test (B scale, used for materials like aluminum)
uses a 1/16" (1.6mm) steel ball and measures the depth indented with
100kgf. Typical hardness numbers for aluminum would indicate a depth of
around 0.14mm (into a flat surface of reasonable thickness). While the
scenario is a bit different, the dimensions and forces are similar.
Given that the spoke and hole diameters are close and the angle is
small, it seems that the absolute "bedding in" is small and happens
early (with tension). From there it seems safe to assume that stress
relief doesn't cause any significant additional "bedding in".

Perhaps this issue is finally put to bed.

Ben C

On 2008-05-02, Peter Cole <peter_cole@verizon.net> wrote:
> jobst.brandt@stanfordalumni.org wrote:
>> Stress relief.
>>
>> I think deformation of flange holes in aluminum hubs is being
>> mis-characterized, leading to incorrect assumptions about effective
>> spoke hole position before and after stress relieving and that
>> trueness of the wheel suffers from the process.
>>
>> Hole deformation is an asymptotic effect that with reasonable spoke
>> tension is already as deep as it will get. If that were not so,
>> spokes would gradually sink through the flange and pull out. Once
>> about a third of the spoke diameter bears fully on the flange hole it
>> is as deep as it will go while subsequent stress relief cannot "bed
>> them in" any deeper as the process is often depicted here. The terms
>> bedding in or stabilizing are a misnomers chosen by people who cannot
>> visualize mechanical stress relief or that spokes bed in naturally
>> from initial spoke tension.
>
> The Rockwell hardness test (B scale, used for materials like aluminum)
> uses a 1/16" (1.6mm) steel ball and measures the depth indented with
> 100kgf. Typical hardness numbers for aluminum would indicate a depth of
> around 0.14mm (into a flat surface of reasonable thickness). While the
> scenario is a bit different, the dimensions and forces are similar.
> Given that the spoke and hole diameters are close and the angle is
> small, it seems that the absolute "bedding in" is small and happens
> early (with tension). From there it seems safe to assume that stress
> relief doesn't cause any significant additional "bedding in".
>
> Perhaps this issue is finally put to bed.

I'm not quite ready to put it to bed yet.

The most likely explanation in my mind is that during stabilization the
spoke cuts into a new part of the hub flange (the outside edge of the
hole probably) that it wasn't quite bearing on before. The parts it's
already been pulled against by tensioning don't deform any further for
the reasons you state.

jobst.brandt@stanfordalumni.org

Peter Cole wrote:

>> Stress relief.

>> I think deformation of flange holes in aluminum hubs is being
>> mis-characterized, leading to incorrect assumptions about effective
>> spoke hole position before and after stress relieving and that
>> trueness of the wheel suffers from the process.

>> Hole deformation is an asymptotic effect that with reasonable spoke
>> tension is already as deep as it will get. If that were not so,
>> spokes would gradually sink through the flange and pull out. Once
>> about a third of the spoke diameter bears fully on the flange hole
>> it is as deep as it will go while subsequent stress relief cannot
>> "bed them in" any deeper as the process is often depicted here.
>> The terms bedding in or stabilizing are a misnomers chosen by
>> people who cannot visualize mechanical stress relief or that spokes
>> bed in naturally from initial spoke tension.

> The Rockwell hardness test (B scale, used for materials like
> aluminum) uses a 1/16" (1.6mm) steel ball and measures the depth
> indented with 100kgf. Typical hardness numbers for aluminum would
> indicate a depth of around 0.14mm (into a flat surface of reasonable
> thickness). While the scenario is a bit different, the dimensions
> and forces are similar. Given that the spoke and hole diameters are
> close and the angle is small, it seems that the absolute "bedding
> in" is small and happens early (with tension). From there it seems
> safe to assume that stress relief doesn't cause any significant
> additional "bedding in".

> Perhaps this issue is finally put to bed.

Don't you believe that! Not with the argumentative anti technocrats
of this forum. My distinct impression is that the situation is only
getting worse. The "In Your Face" style popular these days makes
logical and civil discourse difficult. Just the same, thanks for
researching the numbers. I arrived on this from years of observation
that led to writing about it.

Jobst Brandt