Odd spoke behavior
carl...@comcast.net
I bent them at their midspans, stuck them in a vise with their free
ends pointing straight up, and applied a propane torch to the bends.
None of the spokes stood up to torture.
Old, used, 2 mm carbon spokes curled up when heated.
This bend _| became _\ when an old used spoke was heated.
But a new, unused, 1.8 mm stainless steel spoke uncurled when heated.
This bend _| became _/ when an unused ss spoke was heated.
Is torture unreliable where spokes are concerned? Why does the same
90-degree bend respond differently to heat in the different spokes?
Someone on RBT ought to know what's going on.
Is it carbon plated versus stainless steel? Different thicknesses?
Used versus unused? Something else that I'm too ignorant to think of?
Here are some before and after diagrams and pictures.
On the left is the used, plated, carbon 2 mm spoke.
On the right is the unused stainless steel 1.8 mm spoke.
bent heated
c s c s
| | \ /
carbon | | carbon \ /
________| | ________\ /
vise| | vise| /
____|_____| ____|_____/
stainless stainless
The bent spokes stick up || like this:
http://server5.theimagehosting.com/image.php?img=256acarbonstainlessbent.jpg
After heating, they splay apart \/ like this:
http://server5.theimagehosting.com/image.php?img=257acarbonstainlessheated.jpg
It's not an optical illusion. The pegboard and ruler are about a foot
past the yellow hammer handle, and the yellow hammer handle is about a
foot past the spokes in the vise:
camera spokes hammer ruler-on-pegboard
Cheers,
Carl Fogel
Bob Bongloaded
<carl...@comcast.net> wrote in message
news:0tsti2pjamtldd12i...@4ax.com...
(snip)
This is a classical example of bongification, a term I introduced to rbt
many years ago. At the time, no one other wreck.bikers had even heard the
word, to say nothing of actually comprehending.
BB
jim beam
> Never mind why, but I recently tortured some spokes.
>
> I bent them at their midspans, stuck them in a vise with their free
> ends pointing straight up, and applied a propane torch to the bends.
>
> None of the spokes stood up to torture.
>
> Old, used, 2 mm carbon spokes curled up when heated.
>
> This bend _| became _\ when an old used spoke was heated.
>
> But a new, unused, 1.8 mm stainless steel spoke uncurled when heated.
>
> This bend _| became _/ when an unused ss spoke was heated.
>
> Is torture unreliable where spokes are concerned? Why does the same
> 90-degree bend respond differently to heat in the different spokes?
> Someone on RBT ought to know what's going on.
>
> Is it carbon plated versus stainless steel?
yes.
> Different thicknesses?
to some degree.
> Used versus unused?
no.
> Something else that I'm too ignorant to think of?
>
> Here are some before and after diagrams and pictures.
>
> On the left is the used, plated, carbon 2 mm spoke.
>
> On the right is the unused stainless steel 1.8 mm spoke.
>
> bent heated
>
> c s c s
> | | \ /
> carbon | | carbon \ /
> ________| | ________\ /
> vise| | vise| /
> ____|_____| ____|_____/
> stainless stainless
>
> The bent spokes stick up || like this:
>
> http://server5.theimagehosting.com/image.php?img=256acarbonstainlessbent.jpg
>
> After heating, they splay apart \/ like this:
>
> http://server5.theimagehosting.com/image.php?img=257acarbonstainlessheated.jpg
>
> It's not an optical illusion. The pegboard and ruler are about a foot
> past the yellow hammer handle, and the yellow hammer handle is about a
> foot past the spokes in the vise:
>
> camera spokes hammer ruler-on-pegboard
>
> Cheers,
>
> Carl Fogel
i don't know the exact mechanism, but here are some thoughts:
1. residual stress. if there's a residual stress profile, as the
material is heated towards yield [yield is temperature dependent as any
blacksmith knows], then the portion that remains stressed can prevail in
a tug-o-war over the portion that is not.
2. this is relatively simple situation with the stainless spoke as it
remains in a largely austenitic phase both before and after heating.
with a carbon steel however, heating will cause a phase transformation
from ferritic to austenitic, two entirely different crystal structures,
and i believe that will throw the cat among the pigeons regarding
residual stress behavior - i don't know exactly how residual stress
survives the phase change, if at all, and i don't know whether residual
stress has any influence on exact temperature of transition, but i think
you can be sure there is a thesis or two out there that will explain.
what's really interesting carl is whether the pre-formed spoke elbows
behave like this when heated. during manufacture, residual stress can
be mechanically relieved by a small scale secondary operation that
quickly follows the primary forming operation. if spokes leave the
factory with residual stress, then your spoke elbows should evidence the
same kinds of behavior when heated. if they leave the factory in a
relieved condition, then they won't.
carl...@comcast.net
Dear Jim,
Both spokes start to bend almost as soon as the propane flame hits the
bend, and keep bending in the same direction with more heat (curling
up for the carbon, uncurling for the stainless steel).
I've heated the straight section beyond the bend to a cheerful orange
glow and then moved the glowing orange section slowly through the bend
and into the straight section beyond.
Checking spoke elbows was indeed my hope, but that would take very
careful measurements and an understanding of why the hell obvious
bends in the midspans curl or uncurl in different directions depending
on whether they're used carbon 2 mm or unused 1.8 mm stainless. You
can see in the pictures how little the angle of the 90-degree bend
changes---the end of a 6-inch "needle" moves, but not dramatically.
But there's no point in heating tiny spoke elbows when I haven't the
faintest idea what monster 90 degree midspan bends are doing when I
heat them.
I expected that the bends might change to a different degree, due to
different materials or different diameters, but I didn't expect the
spokes to thumb their noses at me and curl and uncurl in completely
different directions when I toasted their bends.
It's a ridiculously simple test, so maybe other posters will try it
and figure out something using different spokes. Just bend a spoke
sideways, stick the spoke in a vise, and heat the bend with the flame
from a soldering torch. The bending motion is visible, but it seems to
go one way for carbon and the other way for stainless.
If the spoke is heated enough, it loses so much of its strength that
it becomes quite flexible, so sticking the free end straight up or
down will reassure you that the heat is bending things, not gravity.
Cheers,
Carl Fogel
jim beam
of course - you're imparting energy that allows dislocations and atoms
to move about in their lattice. during the drawing process, wire stock
is heated multiple times to soften and de-stress.
> and keep bending in the same direction with more heat (curling
> up for the carbon, uncurling for the stainless steel).
to a degree. there's only so much stored energy.
>
> I've heated the straight section beyond the bend to a cheerful orange
> glow and then moved the glowing orange section slowly through the bend
> and into the straight section beyond.
>
> Checking spoke elbows was indeed my hope, but that would take very
> careful measurements and an understanding of why the hell obvious
> bends in the midspans curl or uncurl in different directions depending
> on whether they're used carbon 2 mm or unused 1.8 mm stainless. You
> can see in the pictures how little the angle of the 90-degree bend
> changes---the end of a 6-inch "needle" moves, but not dramatically.
but you should be able to get a visual on the "before and after".
>
> But there's no point in heating tiny spoke elbows when I haven't the
> faintest idea what monster 90 degree midspan bends are doing when I
> heat them.
why not? it's exactly the same principle.
>
> I expected that the bends might change to a different degree, due to
> different materials or different diameters, but I didn't expect the
> spokes to thumb their noses at me and curl and uncurl in completely
> different directions when I toasted their bends.
>
> It's a ridiculously simple test, so maybe other posters will try it
> and figure out something using different spokes. Just bend a spoke
> sideways, stick the spoke in a vise, and heat the bend with the flame
> from a soldering torch. The bending motion is visible, but it seems to
> go one way for carbon and the other way for stainless.
>
> If the spoke is heated enough, it loses so much of its strength that
> it becomes quite flexible,
that's a function of both temperature and time.
ngri...@aol.com
> http://server5.theimagehosting.com/image.php?img=256acarbonstainlessb...
>
> After heating, they splay apart \/ like this:
>
> http://server5.theimagehosting.com/image.php?img=257acarbonstainlessh...
>
> It's not an optical illusion. The pegboard and ruler are about a foot
> past the yellow hammer handle, and the yellow hammer handle is about a
> foot past the spokes in the vise:
>
> camera spokes hammer ruler-on-pegboard
>
> Cheers,
>
> Carl Fogel
Dear Carl,
Too many variables. To possibly eliminate some of these you should
test (if you have them):
New carbon spokes, 2.0mm and 1.8mm
Used stainless spokes, 2.0mm and 1.8mm
Not sure I understand why you wouldn't similarly test the bends at the
spoke head.
No offence, but do you have a day job?
Best wishes,
Nigel Grinter
www.wellspokenwheels.com
carl...@comcast.net
Dear Nigel,
I was hoping that someone in our pack of well-educated engineers would
know why used carbon 2 mm spokes curl up when their bends are heated,
but unused 1.8 mm stainless steel spokes uncurl.
Testing more spokes of different dimensions, brands, and history might
show some other patterns, but it wouldn't tell me why the patterns are
different.
The reason that I tested giant 90 degree bends that I made myself with
half a foot of spoke sticking out is that it's easy to see that much
spoke actually moving right in front of your eyes and there's no
question that the spoke has just been bent and not treated in any
other way.
Think about how disconcerting it would have been if two elbows curled
and uncurled only as much as the angles in that picture. One "needle"
is about 150 mm long. The other is about 3 mm long. Would you believe
your measurements on a tiny spoke elbow? Or would you wonder if it was
just too hard to measure?
Your browser seems to truncate the picture links, so here they are
again:
http://server5.theimagehosting.com/image.php?img=256acarbonstainlessbent.jpg
or http://tinyurl.com/y8lodt
http://server5.theimagehosting.com/image.php?img=257acarbonstainlessheated.jpg
or http://tinyurl.com/tnrhw
A quick check with a protractor program indicates that the carbon
spoke curls up less than 2 degrees.
In other words, I was checking first to see if anything at all
happened on a large scale before I wasted any time trying to see if I
could measure extremely small changes in tiny spoke elbows.
And I'm not sure if the bending is caused by stresses being relieved
or by some weird expansion that overwhelms any stress relief.
Yes, I've got a day (and night) job. Luckily, it allows me enough time
to bend two spokes with my forepaws, clamp them in a vise, wave a
propane torch at them for less than ten seconds, and wonder what the
hell is going on.
No offence, but I take it that you haven't a clue, either. Maybe
someone else on this technical newsgroup does. If not, I'll enjoy the
vision of puzzled engineers trying to figure it out by toasting
spokes.
Cheers,
Carl Fogel
carl...@comcast.net
After scribbling the reply above, I tested a 2.34 mm soft carbon
spoke. Most of the time was spent digging the propane torch out and
then untwisting the coat hang--
Er, preparing the 2.34 mm soft carbon spoke.
A 90-degree bend was made in a straight section and the spoke was
placed in a vise, sticking straight up. The rest of the spoke was
straightened to provide about two feet of "needle" to show which way
it moved.
Toasting produced the expected result, namely the carbon spoke curled
up a little more. I toasted a few more pieces, including the massive
factory bends in the coat hang--
Er, in the spoke. They curled up very slightly, too.
That took about 15 minutes, most of it wasted on setting up to check
the change in small sections of the coat hang--
Er, in small sections of the spoke.
Cheers,
Carl Fogel
Joe Riel
> The reason that I tested giant 90 degree bends that I made myself with
> half a foot of spoke sticking out is that it's easy to see that much
> spoke actually moving right in front of your eyes and there's no
> question that the spoke has just been bent and not treated in any
> other way.
What if you clamp the (swaged) head of the spoke in the vise, then
heat the elbow. That way the long part of the spoke will serve as a
suitable pointer. Any resulting movement isn't necessarily in the
elbow, it could be in the head or head/vise interface, however, if
there is no consistent movement than you should be able to conclude
that the elbow doesn't move.
> Think about how disconcerting it would have been if two elbows curled
> and uncurled only as much as the angles in that picture. One "needle"
> is about 150 mm long. The other is about 3 mm long. Would you believe
> your measurements on a tiny spoke elbow? Or would you wonder if it was
> just too hard to measure?
--
Joe Riel
carl...@comcast.net
Dear Joe,
The problem with clamping the hub end of the spoke is that it's so
tiny and such a weird shape.
Would any movement be due to the heating of the elbow bend only 3 mm
away causing the swaged end to soften or expand and shift in the
precarious grip of the vise?
For that matter, the vise would be squashing the end of the spoke
before any heat was applied. The button-elbow section is such a tiny
area that clamping it in a vise might introduce new stresses.
So my plan is to clamp the long part of the spoke in the vise, well
away from the elbow, and take pictures against a cross-hatched
background, before and after heating the end.
If the change is only a degree or two, pictures are much more
reassuring than just eyeballing things.
But before I fuss around trying to take pictures of 1 degree bend
changes in 3 mm sections, I want to check whether I'm looking at
thermal stress relief or at some weird heat distortion that overwhelms
any stress relief changes.
Why does the used carbon plated 2 mm spoke curl up, but the unused
stainless 1.8 mm spoke uncurls? There must be an explanation, but it's
going to have to be a good one if the changes are due to stress
relief.
Of course, someone might spend five minutes duplicating my test and
figure out that all spokes curl the same way if the test is done
right. At this point, I'm almost hoping that someone will point out a
really dumb mistake that I'm making.
Now I'm going to toast another pair of bent spokes laid flat on a
concrete floor--no vise or gravity involved to do anything weird.
Cheers,
Carl Fogel
carl...@comcast.net
Luckily, I remembered that heating concrete with a torch is a good way
to find out if there's enough water present to make the concrete
explode, so I toasted the spokes on a pair of 2x4's.
An unused stainless steel 1.8 spoke is on top. A used carbon plated 2
mm with a nipple is on bottom.
The unused stainless spoke has some spots from where it was taped in a
bundle with others. You can see that it's the thinner spoke by
comparing the two at the curve.
Spokes bent at 90 degrees:
http://server5.theimagehosting.com/image.php?img=260acarbonlower.jpg
or http://tinyurl.com/ygl3nw
Spokes after heating with propane torch:
http://server5.theimagehosting.com/image.php?img=261acarbonlower.jpg
or http://tinyurl.com/ykd3vs
The carbon spoke still curls up, and the stainless spoke still
uncurls. Neither a vise nor gravity is involved.
Carl Fogel
Joe Riel
> Dear Joe,
>
> The problem with clamping the hub end of the spoke is that it's so
> tiny and such a weird shape.
Agreed. I hesitated to suggest that because of those difficulties.
I think, as you suspect, that it will be hard to measure the
change without some type of "amplification". Is there any practical
way to bond a pointer to the head of spoke in a manner that won't
be affected by the temperature?
Joe
carl...@comcast.net
Dear Joe,
Asking me about practical may not be the most practical approach.
Someone like Chalo might know how.
Me, I'm going to take the easy way out and let a digital camera and a
computer protractor show an unused spoke elbow bend, the same elbow
after tensioning to 100 kgf in a hub,
once more after the spoke has been squeezed in place, and then after
it's been toasted.
But that's only after someone figures out if I'm looking at thermal
stress relief, thermal distortion, something affected by carbon versus
stainless, or an effect that varies with how thick the spoke is or
where exactly you start heating it.
If heating is showing something besides stress relief, then it's
interesting, but useless for examining spoke elbows. My naive and
ignorant reaction is that heating isn't showing stress relief because
two spokes bent the same way, heated by the same torch, curl in
opposite directions. But maybe someone will figure it out and explain
that curling in opposite directions really is stress relief.
Until what's going on in plain sight on a gross scale is explained, it
just doesn't seem worth the trouble to enlarge hard-to-focus pictures
of tiny elbows and measure single-degree bends with computer protractor
programs.
Toasting spokes seems like the kind of thing that a high-school science
class would use as a quick and easy experiment to illustrate some basic
principle, but I'm damned if I know what I'm seeing, apart from the
willingness of simple bicycle parts to baffle me.
And bending by hand could turn out to be quite different from bending
around a mandrel, to bring up just one potential problem.
Cheers,
Carl Fogel
carl...@comcast.net
The camera was stuck on a mount and the spokes and 2x4's didn't move.
Using an onscreen protractor, I found apparent angles for the upper
stainless steel spoke and the lower carbon spoke with the nipple.
stainless
39.13 degrees /\ 126.93 degrees
87.80
carbon
38.70 degrees /\ 128.37 degrees
89.67
stainless
38.71 degrees /\ 133.52 degrees
94.81, bend widened 7.01 degrees
carbon
40.26 degrees /\ 125.66 degrees
85.40, bend closed 4.27 degrees
These are only apparent degrees from a roughly 45 degree viewing
angle, not straight down, but they show the same pattern of the
stainless bend opening and the carbon bend closing.
CF