Spoke bends for cross-3

Carl Fogel said:

I didn't know how much spokes bend around each other where they
cross, so I cut two straight spoke sections and taped them to the
rim side of of some outer spokes on the drive and non-drive side of
an 8-speed hub on a Performance 700c 32-spoke rim:

Quoted message said:

http://i18.tinypic.com/3ym7bz4.jpg

Quoted message said:

Click on the lower right in Explorer to see things full-size.

Quoted message said:

The ends of the cut lengths of reference spoke are at the bottom of
the white paper and are marked with yellow tape. The upper ends of
the reference spokes are taped out of sight above the top of the
picture.

I don't understand why clipped spokes, that seem to have been heated
to glowing temperatures, are taped to a wheel. Spokes that I have
removed from a well used wheel had no bends at the crossing points;
points that are easily identified by fretting wear marks.

I don't understand what effect the picture you made is showing.

Jobst Brandt

carlfogel said:

I didn't know how much spokes bend around each other where they cross,
so I cut two straight spoke sections and taped them to the rim side of
of some outer spokes on the drive and non-drive side of an 8-speed hub
on a Performance 700c 32-spoke rim:

http://i18.tinypic.com/3ym7bz4.jpg

Click on the lower right in Explorer to see things full-size.

The ends of the cut lengths of reference spoke are at the bottom of
the white paper and are marked with yellow tape. The upper ends of the
reference spokes are taped out of sight above the top of the picture.

You can get an exaggerated idea of the bend where spokes cross by
looking just above and to the right of the "263.71° caption, where a
blurry inner spoke is coming from the hub toward the viewer and then
bending away to the right.

The real spokes are bending around the spokes that they cross. Since
the forest of spokes may be confusing, here's a crude ASCI diagram:

upper rim

\\ //
\\ //
_\\______________________//__________top of picture
\\ //
\\ //
\\ // <--where real spokes bend around
/ \ / \ other real spokes
/ \ / \
/ \ / \
real ref ref real
nds spoke spoke ds
spoke yellow yellow spoke
flange------hub----------flange sprockets

bend about 4.6° bend about 3.7°

An onscreen protractor measured the red lines drawn down the left
sides of the four spokes, but of course any camera angle and parallax
will confuse things slightly at such tiny angles. (I think that the
tendency is to minimize the angles.)

I expect that someone could easily improve on my snapshot.

Posters who bend elbows and nipples might offer some pictures of bent
and unbent spoke ends. It would be interesting to see what angles and
changes are actually involved. I'd be glad to run raw pictures through
a paint program, add lines, and measure angles.

Here's a free Windows screen protractor, the little brother of a
version with six more measuring tools:

http://www.markus-bader.de/MB-Ruler

Cheers,

Carl Fogel

Carl,
I had some trouble picturing what was going on, so tell me if my
restatement is incorrect. You're using the yellow-tagged pieces of spoke
as straight-edges. You've lined them up parallel with spokes in a wheel
(along the section of spoke that runs from the rim to the spoke
crossing). By following the straight-edge past the point where it deviates
from the spoke at the spoke crossing, you get a measure of the angle of
deflection caused by the spoke crossing.

One preliminary question: are you sure the spoke pieces are straight
enough to get the measurements you want?

My bigger question is what you hope to achieve with this test. It seems
that you're interested in what effect correcting the spoke line at the
elbow and nipple will have on the angle of deflection at the spoke
crossing. (Is that why you're only measuring outbound spokes?)

First of all, I'm doubtful that the effect of correcting the spoke line
will show up using your measuring apparatus. If you look at page 97
of the Bicycle Wheel, you'll see that the corrections to the spoke line
are quite subtle, probably too slight to make much of a visible difference
in the spoke's angle along even half of its length.

Furthermore, I don't see what you hope to gain from gathering photos like
this from lots of different people. The spoke angles are determined by
lots of things in addition to whether the spoke line is corrected or not
-- the size of and the distance between the flanges, the rim's ERD, the
number of spokes, the length of the nipples, whether the rim's spoke
holes are offset or angled, etc.

To get the results you seem to want, wouldn't you have to compare two
identical wheels, one with spoke line corrected, the other one without?
(Or one wheel with symmetrical flanges, one flange corrected and the other
not?)

My biggest question, though, is what you would prove if you did manage to
get good before-and-after measurements?

On Fri, 22 Dec 2006 14:41:16 -0600, Gary Young <[email hidden]>

Quoted message said:

carlfogel said:

I didn't know how much spokes bend around each other where they cross,
so I cut two straight spoke sections and taped them to the rim side of
of some outer spokes on the drive and non-drive side of an 8-speed hub
on a Performance 700c 32-spoke rim:

http://i18.tinypic.com/3ym7bz4.jpg

Click on the lower right in Explorer to see things full-size.

The ends of the cut lengths of reference spoke are at the bottom of
the white paper and are marked with yellow tape. The upper ends of the
reference spokes are taped out of sight above the top of the picture.

You can get an exaggerated idea of the bend where spokes cross by
looking just above and to the right of the "263.71° caption, where a
blurry inner spoke is coming from the hub toward the viewer and then
bending away to the right.

The real spokes are bending around the spokes that they cross. Since
the forest of spokes may be confusing, here's a crude ASCI diagram:

upper rim

\\ //
\\ //
_\\______________________//__________top of picture
\\ //
\\ //
\\ // <--where real spokes bend around
/ \ / \ other real spokes
/ \ / \
/ \ / \
real ref ref real
nds spoke spoke ds
spoke yellow yellow spoke
flange------hub----------flange sprockets

bend about 4.6° bend about 3.7°

An onscreen protractor measured the red lines drawn down the left
sides of the four spokes, but of course any camera angle and parallax
will confuse things slightly at such tiny angles. (I think that the
tendency is to minimize the angles.)

I expect that someone could easily improve on my snapshot.

Posters who bend elbows and nipples might offer some pictures of bent
and unbent spoke ends. It would be interesting to see what angles and
changes are actually involved. I'd be glad to run raw pictures through
a paint program, add lines, and measure angles.

Here's a free Windows screen protractor, the little brother of a
version with six more measuring tools:

http://www.markus-bader.de/MB-Ruler

Cheers,

Carl Fogel

Carl,
I had some trouble picturing what was going on, so tell me if my
restatement is incorrect. You're using the yellow-tagged pieces of spoke
as straight-edges. You've lined them up parallel with spokes in a wheel
(along the section of spoke that runs from the rim to the spoke
crossing). By following the straight-edge past the point where it deviates
from the spoke at the spoke crossing, you get a measure of the angle of
deflection caused by the spoke crossing.

One preliminary question: are you sure the spoke pieces are straight
enough to get the measurements you want?

[snip]

Dear Gary,

Yes, you've got the idea. Here's an attempt to diagram a single spoke
with a reference spoke taped to it:

rim ---------------------------- straight spoke taped to real spoke
--------------------- .
' . real spoke bends at crossing

Yes, I'm sure that the cut-off spoke sections are straight. All you
have to do is roll them back and forth on a flat surface.

An email asked me why the spokes were cut, so other posters may be
equally puzzled. If you don't cut the spokes, they won't fit. They run
right into spokes on the other side of the flange.

Cheers,

Carl Fogel

Quoted message said:

Carl Fogel said:

I didn't know how much spokes bend around each other where they
cross, so I cut two straight spoke sections and taped them to the
rim side of of some outer spokes on the drive and non-drive side of
an 8-speed hub on a Performance 700c 32-spoke rim:

Quoted message said:

http://i18.tinypic.com/3ym7bz4.jpg

Quoted message said:

Click on the lower right in Explorer to see things full-size.

Quoted message said:

The ends of the cut lengths of reference spoke are at the bottom of
the white paper and are marked with yellow tape. The upper ends of
the reference spokes are taped out of sight above the top of the
picture.

I don't understand why clipped spokes, that seem to have been heated
to glowing temperatures, are taped to a wheel.

[snip]

Dear Jobst,

Possibly because they weren't heated? 🙂

See Gary's post and my reply. He managed to figure things out, despite
my labored explanation, and there's a diagram.

Tape a straight section of spoke to an outer spoke in a cross-3 wheel,
taping it between the rim and the crossing.

The straight section of spoke is just a reference point to see how
much the real spoke bends at the crossing.

No heat, no fretting marks, nothing up my sleeve, just curious about
what the bend angle at the crossing is.

------------------------------------ straight section of spoke
-------------------------- .
' . real spoke bending at crossing

It would be interesting if posters who bend spokes to change the spoke
line could take some before and after pictures of some sample spokes
that have been bent, or pictures of some unused spokes next to some
spokes that they've bent.

I don't think that anyone has ever posted pictures of bent elbows and
nipple ends, but digital cameras have made it fairly easy to show such
small things. In fact, it's probably easier to take a simple picture
of an elbow or nipple end that has been bent to change the spoke line
next to its unbent twin brother than to take a picture of a crossing
bend with all the other spokes confusing the image.

Cheers,

Carl Fogel

On 2006-12-23, [email hidden] <[email hidden]> wrote:
[snip]

Quoted message said:

I don't think that anyone has ever posted pictures of bent elbows and
nipple ends,

I did post a picture of some bent elbows the other day

http://www.tidraso.co.uk/misc/elbows.jpg

next to the canted hub they spent their life in.

You know it occurs to me that the spoke interleaving may be what causes
the bulge and then spoke line correction.

When I laced a wheel recently I put the inbound spokes in first, then
the outbound ones. Each outbound spoke has to be curved under an inbound
one, which I think is what makes it bulge out near the hub. It bulges
much less if you try, for the sake of argument, inserting it into the
spoke hole without interleaving it at the third crossing. I squeezed the
bulges in a bit after some tensioning, which very likely actually bent
the spokes at the elbows into tighter angles.

I'll post some more pictures as I have all the spokes from the old
wheel, and the other old wheel with the spokes still in it, and the two
new wheels.

The old spokes (from I guess machine-built wheels but don't know) have
plastic deformations in them not just at the elbows: some seem to be
deformed at the crossings, and some seem to curve at the ends towards
the threads. I was easily able to sort the spokes from the drive side of
the wheel into two groups of 8 (it was a 32 wheel) based on the elbow
angles. I haven't tried the other side yet.

Ben C said:

On 2006-12-23, [email hidden] <[email hidden]> wrote:
[snip]

Quoted message said:

I don't think that anyone has ever posted pictures of bent elbows and
nipple ends,

I did post a picture of some bent elbows the other day

http://www.tidraso.co.uk/misc/elbows.jpg

next to the canted hub they spent their life in.

You know it occurs to me that the spoke interleaving may be what causes
the bulge and then spoke line correction.

When I laced a wheel recently I put the inbound spokes in first, then
the outbound ones. Each outbound spoke has to be curved under an inbound
one, which I think is what makes it bulge out near the hub. It bulges
much less if you try, for the sake of argument, inserting it into the
spoke hole without interleaving it at the third crossing. I squeezed the
bulges in a bit after some tensioning, which very likely actually bent
the spokes at the elbows into tighter angles.

I'll post some more pictures as I have all the spokes from the old
wheel, and the other old wheel with the spokes still in it, and the two
new wheels.

The old spokes (from I guess machine-built wheels but don't know) have
plastic deformations in them not just at the elbows: some seem to be
deformed at the crossings, and some seem to curve at the ends towards
the threads. I was easily able to sort the spokes from the drive side of
the wheel into two groups of 8 (it was a 32 wheel) based on the elbow
angles. I haven't tried the other side yet.

Dear Ben,

Sorry, I didn't notice the spokes in that picture.

But what I meant is a before and after picture large enough to see
what the bending looks like compared to a factory-fresh spoke.

That is, set a spoke aside for reference.

Then build a wheel and bend the spokes to change them however to
they're supposed to be bent.

When the wheel is finished, take a drive-side and non-drive-side spoke
out and take some close-up pictures of their elbows and nipple ends
next to the unused spoke.

I think that the pictures will have to be close ups in order to show
the small bends in short section. Anyone who normally tries to change
the spoke line and has a digital camera should be able to do show us
his results without much trouble. As I said, I'll be glad to take any
raw pictures and add lines and measure angles.

Cheers,

Carl Fogel

On 2006-12-23, [email hidden] <[email hidden]> wrote:
[snip]

Quoted message said:

But what I meant is a before and after picture large enough to see
what the bending looks like compared to a factory-fresh spoke.

That is, set a spoke aside for reference.

Then build a wheel and bend the spokes to change them however to
they're supposed to be bent.

When the wheel is finished, take a drive-side and non-drive-side spoke
out and take some close-up pictures of their elbows and nipple ends
next to the unused spoke.

I think that the pictures will have to be close ups in order to show
the small bends in short section. Anyone who normally tries to change
the spoke line and has a digital camera should be able to do show us
his results without much trouble. As I said, I'll be glad to take any
raw pictures and add lines and measure angles.

Here are the 8 outbound spokes from the non-drive side of a rear wheel:

http://www.tidraso.co.uk/misc/elbows1.jpg

I didn't build the wheel, it came new with the bike, and was ridden
around for a few years and thousands of km before the rim cracked.

They all have elbows that are bent into tighter angles than those of the
inbound spokes (http://www.tidraso.co.uk/misc/elbows.jpg compares an
inbound with an outbound spoke from the same wheel). They were on a
Campag Veloce hub which has canted flanges and holes normal to the
flanges (not parallel to the hub axis).

Interestingly the 8 outbound spokes that have tighter angles divide into
two groups of 4, 4 with even tighter elbow bends but less deformation at
the crossings. The lower 4 spokes in the picture are the ones with the
tighter elbows.

Here is a closeup of the elbows showing the difference in angle (unless
I'm imagining it). The upper 4 spokes had to be taped down since they
were deformed quite a bit at the crossings:

http://www.tidraso.co.uk/misc/elbows2.jpg

There are 11 spokes in this picture, the 8 from the previous picture,
and above them 2 new spokes for comparison (different brand though), and
one inbound spoke from the non-drive side of the wheel showing a wider
elbow angle even than the new spokes.

It looks like four of the spokes had the spoke line corrected more than
the other four, resulting in a graceful curve instead of a bend at the
crossing, but a bigger change of elbow angle.

I also looked at some of the spokes under a microscope. At 80x
magnification, it was possible to see pitting on some of them inside the
elbows. In most cases the pits were lined with what definitely looked
like rust. I cleaned the spokes in paraffin/kerosene, and it didn't wash
off. I'm pretty sure the spokes are stainless steel-- they're obviously
not galvanized, and not rusty to the naked eye.

I didn't find any rusty pitting on the outsides of any elbows, but I did
find it on both inbound and outbound spokes.

The new spokes (butted DT ones) didn't have any rust, and looked very
neatly formed at the elbows.

I had a look on Wikipedia, and I wonder if this was pitting corrosion
(http://en.wikipedia.org/wiki/Stainless_steel#Pitting_corrosion).

In article
<[email hidden]>,

Ben C said:

I also looked at some of the spokes under a microscope. At 80x
magnification, it was possible to see pitting on some of them inside the
elbows. In most cases the pits were lined with what definitely looked
like rust. I cleaned the spokes in paraffin/kerosene, and it didn't wash
off. I'm pretty sure the spokes are stainless steel-- they're obviously
not galvanized, and not rusty to the naked eye.

I didn't find any rusty pitting on the outsides of any elbows, but I did
find it on both inbound and outbound spokes.

The residual stress in the cold formed bend is tension
on the inside of the bend, and compression on the
outside of the bend. We expect surface cracking to
initiate mostly on the surfaces under tension.

--
Michael Press

Michael Press said:

In article
<[email hidden]>,

Ben C said:

I also looked at some of the spokes under a microscope. At 80x
magnification, it was possible to see pitting on some of them inside the
elbows. In most cases the pits were lined with what definitely looked
like rust. I cleaned the spokes in paraffin/kerosene, and it didn't wash
off. I'm pretty sure the spokes are stainless steel-- they're obviously
not galvanized, and not rusty to the naked eye.

I didn't find any rusty pitting on the outsides of any elbows, but I did
find it on both inbound and outbound spokes.

The residual stress in the cold formed bend is tension
on the inside of the bend, and compression on the
outside of the bend. We expect surface cracking to
initiate mostly on the surfaces under tension.

So you mean this might be stress corrosion? In my simplistic
understanding of these things that would require chloride. I've yet to
try that experiment.

It was the "orange peel" that jim beam mentioned that I started out
looking for, but I think these were quite high quality spokes. The new
DT spokes were beautifully smooth at 80x.

I thought perhaps the pressure where the spoke sits against the hub
might be depriving the stainless steel of the oxygen required to form
its "passivating layer" leading to these tiny amounts of corrosion.

Whether it has any role in nucleating fatigue etc. I don't know, but the
slightly crushed looking irregular craters it was found in probably do.

Michael Press said:

Quoted message said:

I also looked at some of the spokes under a microscope. At 80x
magnification, it was possible to see pitting on some of them
inside the elbows. In most cases the pits were lined with what
definitely looked like rust. I cleaned the spokes in
paraffin/kerosene, and it didn't wash off. I'm pretty sure the
spokes are stainless steel-- they're obviously not galvanized, and
not rusty to the naked eye.

Quoted message said:

Quoted message said:

I didn't find any rusty pitting on the outsides of any elbows, but
I did find it on both inbound and outbound spokes.

Quoted message said:

Residual stress in the cold formed bend is tension on the inside of
the bend, and compression on the outside of the bend. We expect
surface cracking to initiate mostly on the surfaces under tension.

Well that depends on the residual stress introduced by lacing.
Inbound spokes tent to fail on the inside of the elbow as you say, and
outbound spokes on the outside, because the change from obtuse elbow
bend to acute. That puts the outside of the elbow in tension before
considering spoke tension as an additional load, the kicker to
residual stress.

Jobst Brandt

Quoted message said:

Michael Press said:

Quoted message said:

I also looked at some of the spokes under a microscope. At 80x
magnification, it was possible to see pitting on some of them
inside the elbows. In most cases the pits were lined with what
definitely looked like rust. I cleaned the spokes in
paraffin/kerosene, and it didn't wash off. I'm pretty sure the
spokes are stainless steel-- they're obviously not galvanized, and
not rusty to the naked eye.

Quoted message said:

Quoted message said:

I didn't find any rusty pitting on the outsides of any elbows, but
I did find it on both inbound and outbound spokes.

Quoted message said:

Residual stress in the cold formed bend is tension on the inside of
the bend, and compression on the outside of the bend. We expect
surface cracking to initiate mostly on the surfaces under tension.

Well that depends on the residual stress introduced by lacing.
Inbound spokes tent to fail on the inside of the elbow as you say, and
outbound spokes on the outside, because the change from obtuse elbow
bend to acute. That puts the outside of the elbow in tension before
considering spoke tension as an additional load, the kicker to
residual stress.

Jobst Brandt

utter fudge. you say it's residual stress, but that it works both ways
to cause fatigue? two words: first word: "bull".

have you done your residual stress testing yet jobst?

In article
<[email hidden]>,

Ben C said:

Michael Press said:

In article
<[email hidden]>,

Ben C said:

I also looked at some of the spokes under a microscope. At 80x
magnification, it was possible to see pitting on some of them inside the
elbows. In most cases the pits were lined with what definitely looked
like rust. I cleaned the spokes in paraffin/kerosene, and it didn't wash
off. I'm pretty sure the spokes are stainless steel-- they're obviously
not galvanized, and not rusty to the naked eye.

I didn't find any rusty pitting on the outsides of any elbows, but I did
find it on both inbound and outbound spokes.

The residual stress in the cold formed bend is tension
on the inside of the bend, and compression on the
outside of the bend. We expect surface cracking to
initiate mostly on the surfaces under tension.

So you mean this might be stress corrosion? In my simplistic
understanding of these things that would require chloride. I've yet to
try that experiment.

Yes. Does not require chloride, though there is enough
of it around.

Since I learned about the geometry of residual stress
in spoke elbows, I have been looking for evidence that
indicates where surface cracks initiate.

Quoted message said:

It was the "orange peel" that jim beam mentioned that I started out
looking for, but I think these were quite high quality spokes. The new
DT spokes were beautifully smooth at 80x.

I thought perhaps the pressure where the spoke sits against the hub
might be depriving the stainless steel of the oxygen required to form
its "passivating layer" leading to these tiny amounts of corrosion.

Unlikely that the pressure against the Al has much to
do with the mechanical fatigue, since the hub flange
yields immediately forming a nice soft pillow for the
spoke elbow.

Never heard of a passivating layer on stainless steel.
(?)

Quoted message said:

Whether it has any role in nucleating fatigue etc. I don't know, but the
slightly crushed looking irregular craters it was found in probably do.

It gets my vote.

--
Michael Press

In article <[email hidden]>,

Quoted message said:

Michael Press said:

Quoted message said:

I also looked at some of the spokes under a microscope. At 80x
magnification, it was possible to see pitting on some of them
inside the elbows. In most cases the pits were lined with what
definitely looked like rust. I cleaned the spokes in
paraffin/kerosene, and it didn't wash off. I'm pretty sure the
spokes are stainless steel-- they're obviously not galvanized, and
not rusty to the naked eye.

Quoted message said:

Quoted message said:

I didn't find any rusty pitting on the outsides of any elbows, but
I did find it on both inbound and outbound spokes.

Quoted message said:

Residual stress in the cold formed bend is tension on the inside of
the bend, and compression on the outside of the bend. We expect
surface cracking to initiate mostly on the surfaces under tension.

Well that depends on the residual stress introduced by lacing.
Inbound spokes tent to fail on the inside of the elbow as you say, and
outbound spokes on the outside, because the change from obtuse elbow
bend to acute. That puts the outside of the elbow in tension before
considering spoke tension as an additional load, the kicker to
residual stress.

Understood.

--
Michael Press

Michael Press said:

In article
<[email hidden]>,

Ben C said:

Michael Press said:

In article
<[email hidden]>,
Ben C <[email hidden]> wrote:

> I also looked at some of the spokes under a microscope. At 80x
> magnification, it was possible to see pitting on some of them inside the
> elbows. In most cases the pits were lined with what definitely looked
> like rust. I cleaned the spokes in paraffin/kerosene, and it didn't wash
> off. I'm pretty sure the spokes are stainless steel-- they're obviously
> not galvanized, and not rusty to the naked eye.
>
> I didn't find any rusty pitting on the outsides of any elbows, but I did
> find it on both inbound and outbound spokes.
The residual stress in the cold formed bend is tension
on the inside of the bend, and compression on the
outside of the bend. We expect surface cracking to
initiate mostly on the surfaces under tension.

So you mean this might be stress corrosion? In my simplistic
understanding of these things that would require chloride. I've yet to
try that experiment.

Yes. Does not require chloride, though there is enough
of it around.

chloride is the #1 stress corrosion test for stainless steels, so it's
the obvious choice. and since the objective is to /prove/ the
supposition...

Quoted message said:

Since I learned about the geometry of residual stress
in spoke elbows, I have been looking for evidence that
indicates where surface cracks initiate.

that's easy - get a broken spoke and a magnifier.

Quoted message said:

Quoted message said:

It was the "orange peel" that jim beam mentioned that I started out
looking for, but I think these were quite high quality spokes. The new
DT spokes were beautifully smooth at 80x.

I thought perhaps the pressure where the spoke sits against the hub
might be depriving the stainless steel of the oxygen required to form
its "passivating layer" leading to these tiny amounts of corrosion.

it's possible to have galvanic corrosion, or even corrosion fatigue in
this situation, but if that /is/ the mechanism, it's not apparent.

Quoted message said:

Unlikely that the pressure against the Al has much to
do with the mechanical fatigue, since the hub flange
yields immediately forming a nice soft pillow for the
spoke elbow.

the fatigue is due to the bending of the elbow as the spokes goes
through its stress cycle. flange deformation does not change this
fundamental spoke configuration.

Quoted message said:

Never heard of a passivating layer on stainless steel.
(?)

indeed. it's present from manufacture. spokes might even acid
passivated to ensure full corrosion resistance.

Quoted message said:

Quoted message said:

Whether it has any role in nucleating fatigue etc. I don't know, but the
slightly crushed looking irregular craters it was found in probably do.

It gets my vote.