Need Jobst (and anyone else's) help: Twisted-spoke damping factor

>citing gerd schraner from his book "the art of wheelbuilding" --

>the twisting of spokes, instead of classic crossing work is a waste
of time for >road use. tests have shown that it results in a very
unstable wheel. the >lateral stability is about the same as a
normally spoked wheel, but radially the >wheel reacts so sluggishly
that encounters with momentary radial overloads the >highly praised
damping effect is delayed. ... twisted spoking patterns are an
>additional mechanical disadvantage resulting from the extreme angle
of the spoke >directly at the nipple. (schraner, p. 60)

So, there is a claim that wheels with twisted spokes have a
significantly different "damping factor" than normally laced wheels.
The above quote given as evidence.

It is also claimed that this "damping factor" inherent in
twisted-spoke wheels are good for trials riding. This conclusion
deduced somehow.

Sounds like a lot of pseudoscience to me. What's the scoop on this?

>> citing Gerd Schraner from his book "the art of wheelbuilding" --

# the twisting of spokes, instead of classic crossing work is a waste
# of time for road use. Tests have shown that it results in a very
# unstable wheel. The lateral stability is about the same as a
# normally spoked wheel, but radially the wheel reacts so sluggishly
# that encounters with momentary radial overloads the highly praised
# damping effect is delayed. ...twisted spoking patterns are an
# additional mechanical disadvantage resulting from the extreme angle
# of the spoke directly at the nipple. (Schraner, pp. 60)

What he means by "twisting of spokes" is unclear without looking it up
in his book that I can't find at the moment.

If you read the book, you'll have noticed that it is not a technical
account nor does it have any analysis of loads and deflections. It is
what the title says, "The Art of Wheelbuilding". Don't take what is
stated too analytically.

Jobst Brandt
jobst....@stanfordalumni.org

Dear PC,

I can think of only two ways to twist spokes.

One way is to barber-pole the individual spoke while
tightening the nipple, something more easily seen (and done)
on bladed spokes. This is a bad thing because the spoke will
untwist and unscrew itself slightly from the nipple, which
means that the rim will go out of true.

Here's the other way to twist spokes, which is probably what
they're talking about:

http://www.ne.jp/asahi/julesandjames/home/disk_and_quick_release/stratos.jpg

This chain-link twisted-spoke pattern might provide some
spring action, with the spokes making 30-degree bends
against each other, but any spring-like action in tensioned
spokes is likely to be dwarfed by the springy nature of the
inflated rubber tire, the flexing of the fork, and the
movement of the grip end of the handlebar when the rider
leans on things.)

(If they meant damping, just replace "spring action" with
"damping action." The same objection remains, whether for
touring or trials, that the rubber tires spring and damp
much more than steel spokes.)

The chain-link pattern looks like just another reminder that
almost any spoke pattern can be pedalled down the road.

Carl Fogel

in this case, you do get a pronounced spring action. the spokes no
longer axially load. imagine pulling on a piece of wire - not much
spring. now coil that wire, then pull. substantially different results.

Dear Jim,

You may be right, but I'm having trouble seeing how things
move more with pairs of spokes twisted around each other.

Assuming that the twisted spokes are tensioned, where's the
spring? That is, doesn't a curved spoke actually have to
elongate more than a straight spoke to produce the same
result in terms of the distance from the nipple to the hub?

And unless the rim deflects substantially, what would
move--the axle in relation to the rest of the wheel? Is the
rim supposed to flatten out significantly more? Or what?

Puzzled,

Carl Fogel

the coiled spring is an illustrative example. _any_ deviation from
straight will increase flexibility experienced. yes, it's supported by
the other spoke, and the effect is mitigated by the [in]flexibility of
the rim, but the principle still applies.

I don't see how the example of a coil spring is germane. Coil springs
work by converting linear motion to torsion. How does this apply to
snowflake wheels?

Dear Jim,

Well, the spokes do twist around each other, with each spoke
making a sort of half-hearted single coil.

But unlike a free coil spring, they're linked like a
chain-link fence. And unlike the chain-link fence strands,
each spoke is presumably under considerable tension.

It might make things clearer if someone could give some
numbers for comparison, such as how much tension is used and
how much the rim (and how long a span of it) would deflect
for both straight cross-3 spokes and for snowflake pattern
twisted spokes.

I still suspect that the rubber inflated tires, the front
fork, the handlebars, and the seat plus seat post would both
give and damp far more than stainless steel spokes,
regardless of spoke pattern.

Carl Fogel

This seems clear to me; I hope my effort to explain it is not too murky.

They can't "sharpen their curvature" if it's already been sharpened to the
maximum, which might be the case on a properly tensioned wheel; and if not,
would lead to spokes failing more quickly at these crossings, I believe.

--
Benjamin Lewis

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

> ... I believe it would take less force to stretch the woven spokes
> than straight spokes...

Unless you build your wheels such that the hubs hang from the upper
spokes, I don't think anything should be stretching much.

it's not a straight line between the two anchor points. anything that
does not take this straight line in not in pure axial load. non-axial =
greater deflection. the coil spring is a a polar extreme. _any_ shape,
whether coil, random, zigzag, clothes hanger, _any_ shape that is not
the pure axial load will bend at any point not directly on the load
axis. bracing adds complexity to the math, but the same principles
still apply.

Twisting two spokes together just couples them in the sense of coupling two
springs together. The effect of this is to make the spoke (effectively)
more compliant, exactly like reducing the diameter. It will not affect
damping, since that would require the introduction of friction.

I see no advantage to this not obtained more easily with butted spokes, and
there are disadvantages: the effect of spoke breakage is doubled (actually
worse, since they are adjacent), and the angle at the nipple is bad.

This can be simplified and at the same time take out any inelegant
bends that might change elasticity, by pulling the to-be-twisted spoke
pairs together by tying a collar of some type around them (and giving
them a kink) so they lie straight in their free runs. It boils down
to the previous discussion where all this does is make two spokes
"fail" when one of them fails.

That's my take on this as well. But what did Schraner mean by all
this and why did he bring it up in his book without any supporting
explanation or evidence?

Jobst Brandt
jobst....@stanfordalumni.org

>> to the maximum, (clip) ^^^^^^^^^^^^

If there are any flexing bends, then you can be sure spokes will fail
in fatigue in a short time. This has no advantage.

Jobst Brandt
jobst....@stanfordalumni.org

>From a recent post elsewhere:

Here's a better picture of the snowflake, chain-link,
twisted-spoke pattern, showing the whole wheel:

http://www.dtonline.org/structures/get.html?_Action=GetImage&_Key=MainImage&_Id=14&_Wizard=0&_DontCache=1062613691&_Where=AllValidated
or http://tinyurl.com/5jhy4

Carl Fogel

Viewed the picture, looks ugly. Like a "pre-ruined" wheel. Bet it costs
a lot of money to get it done too.

--
waxbytes

Buy long, and just keep twisting until you get to the right
length.

With all the interest generated by this thread, I fear that the
next K$yrium wheels will have aluminum twisted spokes, or maybe
the Spox will come out with a fully laced doily pattern. As Mike
puts it in the RaceLite thread, this would "offer yet another way
to differentiate or classify a bike" -- which I suppose is reason
enough to build something like this. -- Jay Beattie.

>From a recent post elsewhere:

And here's an even better pair of pictures of the snowflake,
chain-link, twisted-spoke wheel:

http://www.ozbizz.com/melton/kensbikes/snowflake.htm

Note that these Australian wonders have a double-twist,
producing God knows what kind of effect.

Technically, this has moved beyond chain-link to
chicken-wire.

Now I'm hoping to find a wheel with four spokes twisted.

Carl Fogel

>From a recent post elsewhere:

And yet another set of twisted spokes, this one a unicycle:

http://www.xs4all.nl/~klaasbil/107_0755.JPG

Note that these use the double twist that distinguishes mere
chain-link from chicken-wire.

I doubt that I'll find four spokes twisted together, but
maybe some daring soul has gone to a triple twist?

Carl Fogel

Wrong type of 'bike' is all.

+-------------------------------------------------------------------+
|Filename: wheels-Chromed60GoldSpokeTwisted.jpg |
|Download: http://www.cyclingforums.com/attachment.php?attachmentid=2495|
+-------------------------------------------------------------------+

--
waxbytes

Dear WB,

Those purely cosmetic spiral spokes make it hard to google
for the truly intertwined ones.

It would be interesting to learn if the process of creating
the cosmetic twisting has any ill effects on these gaudy
straight spokes.

Does the metal-working leave strains that lead to spokes
breaking? Do the artistic sharp edges and corners act as
stress risers?

And what happens where such spokes cross each other? Your
example is a little unusual in that it uses a cross-spoke
pattern. Most of the ones that I saw while searching for
snowflake wheels used a radial pattern to avoid chafing of
the irregular twisted surfaces.

Carl Fogel

IMO, this one is suitable for any claims that twisting the spokes is a
good idea:

http://www.farmvet.com/farmvet/shop/shopdisplaydetailfarm.asp?prodid=1002003&store=o

or

http://tinyurl.com/5rw68

--
Typoes are a feature, not a bug.
Some gardening required to reply via email.
Words processed in a facility that contains nuts.

>From a recent post elsewhere:

Dear PC,

Tucked away in all the theorizing below is the first actual
observation that I've found about a snowflake wheel in
action, from a unicyclist:

"My friend Jason has a MUni with twisted spokes. I did a
double take when I first saw it ans asked what was up with
the wheel. He said he bought the spokes too long and didn't
want to bother returning them!"

"As for strength, in theory the wheel could go either way.
When pulled tight, the twists in the spokes will shorten,
then lengthen. This could add some springiness to the wheel
(making it more robust in the long run), but the movement of
the spokes could cause the to weaken at the twist and
possibly break. In practice, the untensioning of the
bottom spokes is a much bigger factor than tighhtening of
the top ones. In any case, Jason's wheel creaks and groans
when he rides, so the spokes are indeed moving."

http://www.unicycle-forum.com/unicycling/Spoke_patternswhy_do_this_108028.html

The creaking and groaning from a unicycle wheel doesn't
sound (pun intended) reassuring.

Buying too-long spokes seems to be a common reason for this
strange spoke pattern:

http://groups-beta.google.com/group/rec.bicycles.tech/browse_frm/thread/93a58b0f5a475510/73d77100b9de61c1?q=dear+james+twisted&_done=%2Fgroup%2Frec.bicycles.tech%2Fsearch%3Fgroup%3Drec.bicycles.tech%26q%3Ddear+james+twisted%26qt_g%3D1%26&_doneTitle=Back+to+Search&&d#73d77100b9de61c1
or http://tinyurl.com/44snj

Carl Fogel

> take when I first saw it and asked what was up with the wheel. He

> in the long run), but the movement of the spokes could cause them to

> http://www.unicycle-forum.com/unicycling/Spoke_patternswhy_do_this_108028.html

The item above about the untensioning of bottom spokes is telling.
That has little to do with sprinyness since deflections are amazingly
small (in the few 0.001") so the upshot is that it shortens fatigue
life... but then how many miles does a unicycle get?

I'd like to hear that.

If you believe that baloney you'll believe most any dodge. Can you
imagine how a spoke that is too long can be made right by this method.
Consider that spoke lengths need to be within 2mm of the right length
to work reasonably. What are the chances that a spoke that was bought
for a specific wheel and was found to be a little too long would reach
the rim if wrapped around an adjacent spoke?

My take is that those who use this excuse are embarrassed that they
built such a wheel and want to pass it of as something that had to be
done that way rather than by choice.

Jobst Brandt
jobst....@stanfordalumni.org

Dear Jobst,

James Annan has explained that no spokes of the right length
were available when he ruined a wheel on a trip and that's
why he used the odd lacing.

Your familiar paranoid theory that people must be lying is
even less convincing than usual in this case. If they're
embarrassed as you claim, why are they taking the trouble to
tell everyone and post pictures of their shame?

Carl Fogel

[snip]

[snip]

>Jobst Brandt
>jobst....@stanfordalumni.org

Dear Jobst,

Actually, the odds look pretty good when I measure the three
sections of a double-twisted spoke, add them up, and compare
them to the straight-line distance with a graphics ruler
program: http://www.markus-bader.de/MB-Ruler/

For the unicycle picture on my screen, a straight spoke from
hub to nipple looks like about 142 units, while the three
sections of the double twisted spoke look like 64+35+47=146
units:

http://www.xs4all.nl/~klaasbil/107_0755.JPG

For the Australian snowflake wheel on my screen, a straight
spoke looks like about 112 units, while the three sections
of the double-twisted spoke look like 30+21+63=114 units.

http://www.ozbizz.com/melton/kensbikes/snowflake.htm

My measurements are hardly exact, but the weird twisting
pattern seems to require only slightly longer spokes than a
normal pattern--and there are plenty of posts on
rec.bicycles.tech from people who have spokes that are close
to the right size, but not close enough.

So if the spokes are just a few millimeters too long to work
in a normal pattern, then the strange twisted spoke pattern
might well let you use them. (Your 2 mm minimum strikes me
as a bit low.)

And if they're still just a little too long with a single
twist, then a double twist might adjust things. So far, I
haven't found anything about how to calculate the required
spoke length for these oddball lacing patterns.

Thanks for leading me to do some actual measuring--until I
thought about it, I assumed that the spokes would have to be
much, much longer to make the twisted pattern work, but it
looks as if just a little longer is enough.

Carl Fogel

[big snip]

Or you could just do the reasonable thing and increase the spoke
cross, but I guess that would be too easy. Note that the
snowflake wheel is only three cross (from what I can tell). Same
with the unicycle wheel. Going from three to four cross on a 700c
wheel would add almost 8mm to the spoke length, and going five
cross would add about 15mm. Only after about the 20mm mark would
the "oops, I bought the wrong length spoke" seem to justify a
twisted spoke pattern -- and that assumes that with a three cross
700c wheel, 20mm reasonably can be taken up with a twist or
two. -- Jay Beattie.

Five times twenty degrees is 100, so if you have a 36 hole hub cross
five is going to be troublesome.

Dear Jay,

I'm not sure that "reasonable" or "justify" are exactly the
kind of words running through the minds of people who build
snowflake wheels.

I assume that your 8mm increase for 3 to 4 cross normal
lacing is roughly correct, so for spokes 8mm too long, going
to 4 cross would seem reasonable.

But I think that the twisted spoke trick is actually for
shorter mismatches--that is, spokes too long for normal 3
cross, but too short for 4 cross.

When I measured the spokes in the pictures, the crossed
spokes seemed to be only slightly longer than a straight
run. And that was for small wheels, where the angles are
greater. For 700c wheels, the crossing takes place
relatively closer to the hub and requires an even smaller
increase in spoke length.

Or so I think. I haven't found a spoke-length calculator for
twisted spokes and their variants. I may experiment with a
spare wheel and some string to see what the actual length
differences are.

Carl Fogel

If you do this, try to find string that similar in thickness to a spoke,
since this will make a difference in the winding portion.

Dear Benjamin,

String, hell!

The trip to the garage revealed a long-forgotten 27 & 1/4
inch 36-spoke cross-3 front wheel, probably from a 1980's
Schwinn.

After undoing four spokes, I found that I could cross each
pair once (give a pair of crossing spokes a half-twist) and
still get them to thread into the nipples.

On a 36-spoke wheel, this means half-twisting spokes
separated by six other spokes on the hub, a pair that
already crosses.

It looks as if about 4mm more spoke was exposed, so this
weird pattern would have allowed using a spoke that was
about 4mm too long for normal spoking. If you had nothing
but such too-long spokes handy, you could shove a pair into
a wheel to replace a broken spoke and a good spoke and hope
that no one noticed the weird pair.

God knows how hard it would be to true such a wheel, since
tightening or loosening one spoke would tighten or loosen
the one that it was twisted around.

There must be a number of odd patterns for this sort of
weird twisted-spoke lacing. The next obvious step up on a
36-spoke hub (I think) might be to re-arrange the spoke
heads in and out to allow crossing pairs of parallel spokes,
the ones separated by eight other spokes. This would require
longer than original spokes and a lot of fuss.

But the Australian 36-spoke wheel seems to cross spokes
separated by only four other spokes and uses a double twist:

http://www.ozbizz.com/melton/kensbikes/snowflake.htm

So does the 36-spoke unicycle:

http://www.xs4all.nl/~klaasbil/107_0755.JPG

The 32-spoke example uses a different pattern, but I can't
make out the details because the hub is obscured by the disc
brake:

I still doubt that this crossing has any benefit (other than
allowing emergency use of too-long spokes), but it's amusing
to see how much fuss it requires in its more exotic forms.
Somewhere, I suspect that an enthusiast must have posted
some details about patterns and lengths.

Carl Fogel

Yes, twisted spoke wheels with 4 twists are out there:

"The rear wheel is totally unique as far as lacing is
concerned. No one in this region has copied it yet (phew).
Its 3X on the static side and 2X-Twist on the drive side.
And to make it even stranger, while most twisted/snowflake
pattern wheels use two wrappings of the spokes, I used 4
wrappings. It was a bitch to dish right (tensioning
snowflake pattern spokes is more involved anyways but...)."

http://home.ca.inter.net/~kroberge/9000.html

More and more this twisted-spoke stuff sounds like
decorative art, much like the extended forks of chopper
motorcycles, something that shows how a robust design can be
distorted to ridiculous extremes and still work.

Carl Fogel

> ([Jobst's] 2 mm minimum [for the correct length of a spoke] strikes me
>as a bit low.)

I would say that if anything, it's a trifle generous; plus or minus
1mm is the standard that I find acceptable, though I'll go with plus
two or three (and apply a little elective surgery if necessary) if I
don't have one that's within the preferred spec.

Dear Werehatrack,

I think that you're right in that I didn't make myself
clear. For "correct" or "acceptable" length, people on
rec.bicycles.tech often fuss about 1mm differences in spoke
lengths.

But for a spoke length that will work in the crudest sense
of being not quite too short to thread into the nipple and
not quite so long that the spoke sticks out alarmingly past
the nipple, the full range is probably around 6mm or so.

Many spoke suppliers reduce inventory and fuss by carrying
spokes in 2mm increments---292, 294, 296, and so on--without
any real problems.

The illusion of even more range is created by the different
methods used to measure spoke lengths, as suggested by
frequent posts and replies here about adding or subtracting
a millimeter on spoke calculators according to manufacturer.

Carl Fogel