SAPIM Spoke FAtigue Estimate



On Wed, 07 Feb 2007 18:38:10 +0000, Kinky Cowboy wrote:

> On 7 Feb 2007 06:07:53 -0800, "Johan Bornman" <[email protected]>
> wrote:
>
>>
>>Why would the company think that it's bladed spoke will last longer
>>than the round one?

>
> Because the two spokes have different metallurgy. The production
> process for the CX-Ray is much more involved than the basic "take a
> bit of wire and bang one end into a mushroom" of a cheap straight
> guage spoke.


The Sapim website says that all of their butted spokes are made out of
Inox 18/8 and are cold forged. Yet the CX-Ray supposedly has double the
fatigue life of a 2.0/1.5 double butted spoke. What sort of magic are they
using?

>
> Disregarding the Sapim "hype", the sheer number of boutique wheels
> (HED, Corima, Xtreme) which are built with CX-Rays confirms its
> position as the uber-spoke
>
> Kinky Cowboy*
>
> *Batteries not included
> May contain traces of nuts
> Your milage may vary
 
On Thu, 08 Feb 2007 11:50:03 -0600, A Muzi <[email protected]>
wrote:

>>>>>> Dear Johan,
>>>>>> Your question has come up before--it would be nice if Sapim would
>>>>>> include a description of its testing.
>>>>>> The Sapim spoke fatigue tests must use exaggerated stress cycles
>>>>>> because it takes literally years to break spokes in continuous testing
>>>>>> under ordinary low stress cycles--no spoke manufacturer sells spokes
>>>>>> that are expected to fail in only 2,000 km of normal riding.

>
>>>>> I have no doubt that it is an exaggerated test cycle. But don't they
>>>>> know in marketing hype one doesn't cite exaggerated examples, but best
>>>>> case scenarios. I have SAPIM-built wheels which have done closer to 40
>>>>> 000 kms. Why would a company misrepresent itself so much? I don't
>>>>> understand and I don't suppose there is a rational answer for this.

>
>jim beam wrote:
>>>> there /is/ a rational answer for this - almost all fatigue testing is
>>>> done on an accelerated basis. since that's standard practice, their
>>>> reporting is simply consistent with standard practice.-

>
>> Johan Bornman wrote:
>>> You miss the point, being scientifically rational doesn't make sense
>>> in this brochure.

>>

>jim beam wrote:
>> it makes more sense than being scientifically irrational or making
>> claims like "our spokes last forever if you don't use them", which is
>> basically where you end up if you exclude presentation of normal data.

>
>> Johan Bornman wrote:
>>> It is intended for consumers who can quickly do the
>>> sums and discover that SAPIM kind of suggests a useful life of 1 000
>>> kms for wheels built with its spokes.

>
>> Johan Bornman wrote:
>>> A good copywriter with the
>>> company's interest at heart would have phrased the exercise completely
>>> differently.

>
>jim beam wrote:
>> bottom line, while /you/ may not appreciate the format, the fact is,
>> sapim are the only ones that dare to present fatigue data. that should
>> tell you more than anything else.

>
>Maybe.
>Or were those numbers pulled out of the air? "2000k lifetime" is
>nonsensical to anyone with familiarity. But any ad writer would eagerly
>write "one million cycles" for any product - especially if he weren't
>burdened by knowledge of bikes or spokes.
>
>If they meant 1MM cycles at an extreme load, why no mention of that? If
>that was intended, it's simple in English:
>"Our product will bear up to 1MM cycles of extreme load"
>"Our product sustained 1MM cycles of industry standard fatigue testing"
> "Our competitor's product failed at merely 985,00 cycles"
>
>That is not what they wrote.
>
>I was also curious about the smaller-section aero spoke showing 2.3x
>longevity. That does not make any sense at all as _less_ material just
>couldn't bear _2.3x_ the load cycles. And the values are nice and round,
>too, 1,000,000 and 2,300,000. Convenient numbers, those, as they just
>happen to fit the price curve but do they relate to the actual product?
>
>It is not at all clear to me.


Dear Andrew,

Here's the data, culled from the Sapim site's individual pages,
ordered by cycles to failure:

per n/mm^2
64 x middle test cycles
model 260mm hub x middle x rim strength to failure
----- ---- --- --------- --- -------- ---------
cxray 278g 2.0 x (0.9 x 2.3) x 2.0 1600 3,500,000
strong 430g 2.3 x 2.0 x 2.0 1400 1,600,000
laser 279g 2.0 x 1.5 x 2.0 1500 1,250,000
cx 423g 2.0 x (1.3 x 2.8) x 2.0 1200 1,220,000
race 360g 2.0 x 1.8 x 2.0 1350 980,000
leader 424g 2.0 x 2.0 x 2.0 1080 870,000

Note that the cycles to failure aren't all that rounded.

Judging by weight, the cxray aero spokes (278 grams) are probably
flattened laser butted spokes (279 grams).

The cx aero spokes (423 grams) are probably flattened straight leader
spokes (424 grams).

Since the round spokes are drawn to thin their midsections, their
midsections work harden and strengthen, with the n/mm^2 strength
rising (1080, 1350, 1500) as they grow thinner (2.0, 1.8, 1.5).

The exception is the "strong," which has a 2 mm middle and rim end
(single-butted like a motorcycle spoke). The "strong" probably lasts
about twice as long as the straight 2 mm "leader" because it's got a
much thicker 2.3 mm elbow, which accounts for the extra 6 grams of
weight over the straight 2 mm spoke. It's probably a thinned 2.3 spoke
blank, which accounts for its higher 1400 n/mm^2 midsection strength

The aero spokes also show greater midsection strength than the round
spokes from which they were probably stamped, again due to work
hardening, but this doesn't affect the unworked elbows where they
probably fail.

Jim Beam's explanation for why the flattened aero spokes last longer
makes sense--their angle to the elbow lets them put less bending
strain on the elbow. That is, when an aero spoke bends at the elbow,
it probably does less bending at the elbow and more bending at the
thin section than a straight-section spoke, spreading the stress out.

The round spokes (leader 2.0, race 1.8, laser 1.5) probably last
longer in testing as their midsections grow thinner for the same
reason--as the thinner midsection bends a little more, it puts a
little less bending strain on the elbow, spreading the bending stress
over more spoke.

Cheers,

Carl Fogel
 
On Thu, 08 Feb 2007 12:50:20 -0700, carlfogel wrote:

> On Thu, 08 Feb 2007 11:50:03 -0600, A Muzi <[email protected]>
> wrote:
>
>>>>>>> Dear Johan,
>>>>>>> Your question has come up before--it would be nice if Sapim would
>>>>>>> include a description of its testing.
>>>>>>> The Sapim spoke fatigue tests must use exaggerated stress cycles
>>>>>>> because it takes literally years to break spokes in continuous testing
>>>>>>> under ordinary low stress cycles--no spoke manufacturer sells spokes
>>>>>>> that are expected to fail in only 2,000 km of normal riding.

>>
>>>>>> I have no doubt that it is an exaggerated test cycle. But don't they
>>>>>> know in marketing hype one doesn't cite exaggerated examples, but best
>>>>>> case scenarios. I have SAPIM-built wheels which have done closer to 40
>>>>>> 000 kms. Why would a company misrepresent itself so much? I don't
>>>>>> understand and I don't suppose there is a rational answer for this.

>>
>>jim beam wrote:
>>>>> there /is/ a rational answer for this - almost all fatigue testing is
>>>>> done on an accelerated basis. since that's standard practice, their
>>>>> reporting is simply consistent with standard practice.-

>>
>>> Johan Bornman wrote:
>>>> You miss the point, being scientifically rational doesn't make sense
>>>> in this brochure.
>>>

>>jim beam wrote:
>>> it makes more sense than being scientifically irrational or making
>>> claims like "our spokes last forever if you don't use them", which is
>>> basically where you end up if you exclude presentation of normal data.

>>
>>> Johan Bornman wrote:
>>>> It is intended for consumers who can quickly do the
>>>> sums and discover that SAPIM kind of suggests a useful life of 1 000
>>>> kms for wheels built with its spokes.

>>
>>> Johan Bornman wrote:
>>>> A good copywriter with the
>>>> company's interest at heart would have phrased the exercise completely
>>>> differently.

>>
>>jim beam wrote:
>>> bottom line, while /you/ may not appreciate the format, the fact is,
>>> sapim are the only ones that dare to present fatigue data. that should
>>> tell you more than anything else.

>>
>>Maybe.
>>Or were those numbers pulled out of the air? "2000k lifetime" is
>>nonsensical to anyone with familiarity. But any ad writer would eagerly
>>write "one million cycles" for any product - especially if he weren't
>>burdened by knowledge of bikes or spokes.
>>
>>If they meant 1MM cycles at an extreme load, why no mention of that? If
>>that was intended, it's simple in English:
>>"Our product will bear up to 1MM cycles of extreme load"
>>"Our product sustained 1MM cycles of industry standard fatigue testing"
>> "Our competitor's product failed at merely 985,00 cycles"
>>
>>That is not what they wrote.
>>
>>I was also curious about the smaller-section aero spoke showing 2.3x
>>longevity. That does not make any sense at all as _less_ material just
>>couldn't bear _2.3x_ the load cycles. And the values are nice and round,
>>too, 1,000,000 and 2,300,000. Convenient numbers, those, as they just
>>happen to fit the price curve but do they relate to the actual product?
>>
>>It is not at all clear to me.

>
> Dear Andrew,
>
> Here's the data, culled from the Sapim site's individual pages,
> ordered by cycles to failure:
>
> per n/mm^2
> 64 x middle test cycles
> model 260mm hub x middle x rim strength to failure
> ----- ---- --- --------- --- -------- ---------
> cxray 278g 2.0 x (0.9 x 2.3) x 2.0 1600 3,500,000
> strong 430g 2.3 x 2.0 x 2.0 1400 1,600,000
> laser 279g 2.0 x 1.5 x 2.0 1500 1,250,000
> cx 423g 2.0 x (1.3 x 2.8) x 2.0 1200 1,220,000
> race 360g 2.0 x 1.8 x 2.0 1350 980,000
> leader 424g 2.0 x 2.0 x 2.0 1080 870,000
>
> Note that the cycles to failure aren't all that rounded.
>
> Judging by weight, the cxray aero spokes (278 grams) are probably
> flattened laser butted spokes (279 grams).
>
> The cx aero spokes (423 grams) are probably flattened straight leader
> spokes (424 grams).
>
> Since the round spokes are drawn to thin their midsections, their
> midsections work harden and strengthen, with the n/mm^2 strength
> rising (1080, 1350, 1500) as they grow thinner (2.0, 1.8, 1.5).


What does the strength of the middle section have to do with fatigue?
>
> The exception is the "strong," which has a 2 mm middle and rim end
> (single-butted like a motorcycle spoke). The "strong" probably lasts
> about twice as long as the straight 2 mm "leader" because it's got a
> much thicker 2.3 mm elbow, which accounts for the extra 6 grams of
> weight over the straight 2 mm spoke. It's probably a thinned 2.3 spoke
> blank, which accounts for its higher 1400 n/mm^2 midsection strength
>


Maybe so, but to my mind that calls into question the bending moment
explanation put forth by "jim beam." The Strong should have a greater
bending moment than the Leader, right? The two factors are working
against each other.

I'm also curious how cold working the center section of the Strong from 2.3 to 2.0 gives it a middle section with a strength
about 60% greater than that of the Race, which is cold worked from 2.0 to
1.8mm. It seems like you'd have to do an awful lot of cold working to
overcome the Strong's thickness disadvantage. But again, it's not clear to
me what the strength of the middle section has to do with fatigue of the
elbows.

I find it interesting that "jim beam" is advancing his bending moment
thesis, since it seems to show the importance of correcting the spoke
line, something he's dismissed in the past. He's suggesting that the spoke
that bends most to have the right spoke line is the one that's going to
last the longest. I suppose one could say that the difference is that "jim
beam" thinks elastic bending improves fatigue life, whereas Jobst
recommends plastic bending, but I don't really see how that works. If the
spoke is bent elastically, won't it tend to flex more when the wheel goes
through cycles of loading and unloading?


> The aero spokes also show greater midsection strength than the round
> spokes from which they were probably stamped, again due to work
> hardening, but this doesn't affect the unworked elbows where they
> probably fail.
>
> Jim Beam's explanation for why the flattened aero spokes last longer
> makes sense--their angle to the elbow lets them put less bending strain
> on the elbow. That is, when an aero spoke bends at the elbow, it
> probably does less bending at the elbow and more bending at the thin
> section than a straight-section spoke, spreading the stress out.
>
> The round spokes (leader 2.0, race 1.8, laser 1.5) probably last longer
> in testing as their midsections grow thinner for the same reason--as the
> thinner midsection bends a little more, it puts a little less bending
> strain on the elbow, spreading the bending stress over more spoke.
>
> Cheers,
>
> Carl Fogel
 
On Thu, 08 Feb 2007 12:50:20 -0700, carlfogel wrote:

> On Thu, 08 Feb 2007 11:50:03 -0600, A Muzi <[email protected]>
> wrote:
>
>>>>>>> Dear Johan,
>>>>>>> Your question has come up before--it would be nice if Sapim would
>>>>>>> include a description of its testing.
>>>>>>> The Sapim spoke fatigue tests must use exaggerated stress cycles
>>>>>>> because it takes literally years to break spokes in continuous testing
>>>>>>> under ordinary low stress cycles--no spoke manufacturer sells spokes
>>>>>>> that are expected to fail in only 2,000 km of normal riding.

>>
>>>>>> I have no doubt that it is an exaggerated test cycle. But don't they
>>>>>> know in marketing hype one doesn't cite exaggerated examples, but best
>>>>>> case scenarios. I have SAPIM-built wheels which have done closer to 40
>>>>>> 000 kms. Why would a company misrepresent itself so much? I don't
>>>>>> understand and I don't suppose there is a rational answer for this.

>>
>>jim beam wrote:
>>>>> there /is/ a rational answer for this - almost all fatigue testing is
>>>>> done on an accelerated basis. since that's standard practice, their
>>>>> reporting is simply consistent with standard practice.-

>>
>>> Johan Bornman wrote:
>>>> You miss the point, being scientifically rational doesn't make sense
>>>> in this brochure.
>>>

>>jim beam wrote:
>>> it makes more sense than being scientifically irrational or making
>>> claims like "our spokes last forever if you don't use them", which is
>>> basically where you end up if you exclude presentation of normal data.

>>
>>> Johan Bornman wrote:
>>>> It is intended for consumers who can quickly do the
>>>> sums and discover that SAPIM kind of suggests a useful life of 1 000
>>>> kms for wheels built with its spokes.

>>
>>> Johan Bornman wrote:
>>>> A good copywriter with the
>>>> company's interest at heart would have phrased the exercise completely
>>>> differently.

>>
>>jim beam wrote:
>>> bottom line, while /you/ may not appreciate the format, the fact is,
>>> sapim are the only ones that dare to present fatigue data. that should
>>> tell you more than anything else.

>>
>>Maybe.
>>Or were those numbers pulled out of the air? "2000k lifetime" is
>>nonsensical to anyone with familiarity. But any ad writer would eagerly
>>write "one million cycles" for any product - especially if he weren't
>>burdened by knowledge of bikes or spokes.
>>
>>If they meant 1MM cycles at an extreme load, why no mention of that? If
>>that was intended, it's simple in English:
>>"Our product will bear up to 1MM cycles of extreme load"
>>"Our product sustained 1MM cycles of industry standard fatigue testing"
>> "Our competitor's product failed at merely 985,00 cycles"
>>
>>That is not what they wrote.
>>
>>I was also curious about the smaller-section aero spoke showing 2.3x
>>longevity. That does not make any sense at all as _less_ material just
>>couldn't bear _2.3x_ the load cycles. And the values are nice and round,
>>too, 1,000,000 and 2,300,000. Convenient numbers, those, as they just
>>happen to fit the price curve but do they relate to the actual product?
>>
>>It is not at all clear to me.

>
> Dear Andrew,
>
> Here's the data, culled from the Sapim site's individual pages,
> ordered by cycles to failure:
>
> per n/mm^2
> 64 x middle test cycles
> model 260mm hub x middle x rim strength to failure
> ----- ---- --- --------- --- -------- ---------
> cxray 278g 2.0 x (0.9 x 2.3) x 2.0 1600 3,500,000
> strong 430g 2.3 x 2.0 x 2.0 1400 1,600,000
> laser 279g 2.0 x 1.5 x 2.0 1500 1,250,000
> cx 423g 2.0 x (1.3 x 2.8) x 2.0 1200 1,220,000
> race 360g 2.0 x 1.8 x 2.0 1350 980,000
> leader 424g 2.0 x 2.0 x 2.0 1080 870,000
>
> Note that the cycles to failure aren't all that rounded.
>
> Judging by weight, the cxray aero spokes (278 grams) are probably
> flattened laser butted spokes (279 grams).
>
> The cx aero spokes (423 grams) are probably flattened straight leader
> spokes (424 grams).
>
> Since the round spokes are drawn to thin their midsections, their
> midsections work harden and strengthen, with the n/mm^2 strength
> rising (1080, 1350, 1500) as they grow thinner (2.0, 1.8, 1.5).
>
> The exception is the "strong," which has a 2 mm middle and rim end
> (single-butted like a motorcycle spoke). The "strong" probably lasts
> about twice as long as the straight 2 mm "leader" because it's got a
> much thicker 2.3 mm elbow, which accounts for the extra 6 grams of
> weight over the straight 2 mm spoke. It's probably a thinned 2.3 spoke
> blank, which accounts for its higher 1400 n/mm^2 midsection strength
>

According to this pdf brochure --

www.codagex.com/hardware/brochures/SAPIM.pdf

-- Sapim claim that the Strong is less likely to fatigue because the extra
material gives it a snugger fit with the spoke hole in the flange. If
true, that suggests that other double-butted spokes (and maybe
straight-gauge?) would have similarly higher fatigue
resistance if paired with hubs with smaller spoke holes.

That raises another question -- were their tests of the Race done on
2.0/1.8mm spokes or 1.8/1.6mm? Similarly, were the Leaders 2.0 or 1.8mm?
> The aero spokes also show greater midsection strength than the round
> spokes from which they were probably stamped, again due to work
> hardening, but this doesn't affect the unworked elbows where they
> probably fail.
>
> Jim Beam's explanation for why the flattened aero spokes last longer
> makes sense--their angle to the elbow lets them put less bending strain
> on the elbow. That is, when an aero spoke bends at the elbow, it
> probably does less bending at the elbow and more bending at the thin
> section than a straight-section spoke, spreading the stress out.
>
> The round spokes (leader 2.0, race 1.8, laser 1.5) probably last longer
> in testing as their midsections grow thinner for the same reason--as the
> thinner midsection bends a little more, it puts a little less bending
> strain on the elbow, spreading the bending stress over more spoke.
>
> Cheers,
>
> Carl Fogel
 
On 2007-02-08, Gary Young <[email protected]> wrote:
[...]
> I find it interesting that "jim beam" is advancing his bending moment
> thesis, since it seems to show the importance of correcting the spoke
> line, something he's dismissed in the past. He's suggesting that the spoke
> that bends most to have the right spoke line is the one that's going to
> last the longest.


I think the distinction is between bending in the midsection and bending
at the elbow. If the spoke flexes a bit more at the midsection it can
flex a bit less at the elbow. It's unlikely to fail at the midsection,
and if it is able to flex there, is less likely to fail at the elbow.

As for the "spoke line", it's better if the spoke leaves the flange at
something close to flush. This means less flexing at the elbow, and
references have been posted to premature failures caused by long-shank
spokes which DT made for a while.

http://www.peterwhitecycles.com/DTspokes.htm

I think we all want the spoke (close to) flush, the question about spoke
line correction is one of how you get it there: correct it manually, or
let it find its way there under normal tensioning.
 
On Thu, 08 Feb 2007 15:45:40 -0600, Ben C wrote:

> On 2007-02-08, Gary Young <[email protected]> wrote:
> [...]
>> I find it interesting that "jim beam" is advancing his bending moment
>> thesis, since it seems to show the importance of correcting the spoke
>> line, something he's dismissed in the past. He's suggesting that the spoke
>> that bends most to have the right spoke line is the one that's going to
>> last the longest.

>
> I think the distinction is between bending in the midsection and bending
> at the elbow. If the spoke flexes a bit more at the midsection it can
> flex a bit less at the elbow. It's unlikely to fail at the midsection,
> and if it is able to flex there, is less likely to fail at the elbow.


I don't see how that goes against my point. The mere fact that it's a good
thing that there's some bending going on means that the natural spoke line
inherent in the materials (elbow bend, etc.) is not optimized. We're
talking about flexing that occurs after the spokes have been tensioned and
bedding in has occured. I don't recall that figuring in "jim beam's"
spoke-line theories.

>
> As for the "spoke line", it's better if the spoke leaves the flange at
> something close to flush. This means less flexing at the elbow, and
> references have been posted to premature failures caused by long-shank
> spokes which DT made for a while.
>
> http://www.peterwhitecycles.com/DTspokes.htm
>
> I think we all want the spoke (close to) flush, the question about spoke
> line correction is one of how you get it there: correct it manually, or
> let it find its way there under normal tensioning.
 
On 2007-02-08, Gary Young <[email protected]> wrote:
> On Thu, 08 Feb 2007 15:45:40 -0600, Ben C wrote:
>
>> On 2007-02-08, Gary Young <[email protected]> wrote:
>> [...]
>>> I find it interesting that "jim beam" is advancing his bending moment
>>> thesis, since it seems to show the importance of correcting the spoke
>>> line, something he's dismissed in the past. He's suggesting that the spoke
>>> that bends most to have the right spoke line is the one that's going to
>>> last the longest.

>>
>> I think the distinction is between bending in the midsection and bending
>> at the elbow. If the spoke flexes a bit more at the midsection it can
>> flex a bit less at the elbow. It's unlikely to fail at the midsection,
>> and if it is able to flex there, is less likely to fail at the elbow.

>
> I don't see how that goes against my point.


I probably didn't understand your point.

> The mere fact that it's a good thing that there's some bending going
> on means that the natural spoke line inherent in the materials (elbow
> bend, etc.) is not optimized.


I'm not sure I understand it now.

The way I see it bending at the elbow is bad. Good spoke line at the
elbow (and correct shank length) reduces elbow bending. Flexible
midsection also reduces elbow bending.

Now I thought your point was: correcting the spoke line will reduce
elbow flex in use, so why does jim beam tell us not to correct the spoke
line?

I don't think jim beam has ever denied that poor spoke line
(specifically spoke not flush) contributes to fatigue. It was from him
that I first heard about the long-shank problems with DT spokes. The
point is not that a good spoke line is not desirable, just how should it
be achieved: manually (thumb, crankarm, etc.), or just tension the
spokes and let them do their thing.

One thing though is that jim beam has said IIRC that premature spoke
line correction may actually promote fatigue. I'm less clear on the
reasons for that.
 
On Thu, 08 Feb 2007 13:26:38 -0600, Gary Young <[email protected]>
wrote:

>On Wed, 07 Feb 2007 18:38:10 +0000, Kinky Cowboy wrote:
>
>> On 7 Feb 2007 06:07:53 -0800, "Johan Bornman" <[email protected]>
>> wrote:
>>
>>>
>>>Why would the company think that it's bladed spoke will last longer
>>>than the round one?

>>
>> Because the two spokes have different metallurgy. The production
>> process for the CX-Ray is much more involved than the basic "take a
>> bit of wire and bang one end into a mushroom" of a cheap straight
>> guage spoke.

>
>The Sapim website says that all of their butted spokes are made out of
>Inox 18/8 and are cold forged. Yet the CX-Ray supposedly has double the
>fatigue life of a 2.0/1.5 double butted spoke. What sort of magic are they
>using?
>


AFAIK, the CX-Rays get a different post forming heat treatment
compared with the other spokes. Whatever magic they're using, they
quote a higher UTS for the CX-Rays than their other spokes, so they
must be doing something different to the metal to make that so. They
also quote a longer fatigue life; assuming their test protocol
compares like with like, it doesn't really matter what they've done to
achieve this. The answer to the OP's question is actually "because
they've tested them, and found it to be so"

They can sprinkle fairy dust on the drawing dies for all I care, as
long as they consistently and repeatably make spokes which last longer
than the competition.

Of course, they might have fooled all the people who have spent 4
times as much on a CX-ray compared with a good round double butted
spoke. That would seem to include all the pro team mechanics who
meticulously replace the original spokes in their sponsor-supplied
wheels with CX-Rays.

I don't think fatigue life is a major consideration for most CX-Ray
users (maybe the DH MTB world champions who have won on CX-Rays give
their spokes a beating, but most CX-Rays are on time trial wheels)
Customers who bought CX-Ray equipped wheels from Hed, Zipp, Corima,
American Classic, Velocity, Planet-X, X-treme, Speedcomposites, Tune,
Rolf, Cane Creek, Roland-Werk, Reynolds (who seem to have downgraded
to DT for 2007) or one of the many shop wheelbuilders who use and
recommend CX-rays were probably more impressed by the low weight and
excellent aerodynamic performance of the spokes.

Kinky Cowboy*

*Batteries not included
May contain traces of nuts
Your milage may vary
 
On Thu, 08 Feb 2007 14:45:32 -0600, Gary Young <[email protected]>
wrote:

>On Thu, 08 Feb 2007 12:50:20 -0700, carlfogel wrote:
>
>> On Thu, 08 Feb 2007 11:50:03 -0600, A Muzi <[email protected]>
>> wrote:
>>
>>>>>>>> Dear Johan,
>>>>>>>> Your question has come up before--it would be nice if Sapim would
>>>>>>>> include a description of its testing.
>>>>>>>> The Sapim spoke fatigue tests must use exaggerated stress cycles
>>>>>>>> because it takes literally years to break spokes in continuous testing
>>>>>>>> under ordinary low stress cycles--no spoke manufacturer sells spokes
>>>>>>>> that are expected to fail in only 2,000 km of normal riding.
>>>
>>>>>>> I have no doubt that it is an exaggerated test cycle. But don't they
>>>>>>> know in marketing hype one doesn't cite exaggerated examples, but best
>>>>>>> case scenarios. I have SAPIM-built wheels which have done closer to 40
>>>>>>> 000 kms. Why would a company misrepresent itself so much? I don't
>>>>>>> understand and I don't suppose there is a rational answer for this.
>>>
>>>jim beam wrote:
>>>>>> there /is/ a rational answer for this - almost all fatigue testing is
>>>>>> done on an accelerated basis. since that's standard practice, their
>>>>>> reporting is simply consistent with standard practice.-
>>>
>>>> Johan Bornman wrote:
>>>>> You miss the point, being scientifically rational doesn't make sense
>>>>> in this brochure.
>>>>
>>>jim beam wrote:
>>>> it makes more sense than being scientifically irrational or making
>>>> claims like "our spokes last forever if you don't use them", which is
>>>> basically where you end up if you exclude presentation of normal data.
>>>
>>>> Johan Bornman wrote:
>>>>> It is intended for consumers who can quickly do the
>>>>> sums and discover that SAPIM kind of suggests a useful life of 1 000
>>>>> kms for wheels built with its spokes.
>>>
>>>> Johan Bornman wrote:
>>>>> A good copywriter with the
>>>>> company's interest at heart would have phrased the exercise completely
>>>>> differently.
>>>
>>>jim beam wrote:
>>>> bottom line, while /you/ may not appreciate the format, the fact is,
>>>> sapim are the only ones that dare to present fatigue data. that should
>>>> tell you more than anything else.
>>>
>>>Maybe.
>>>Or were those numbers pulled out of the air? "2000k lifetime" is
>>>nonsensical to anyone with familiarity. But any ad writer would eagerly
>>>write "one million cycles" for any product - especially if he weren't
>>>burdened by knowledge of bikes or spokes.
>>>
>>>If they meant 1MM cycles at an extreme load, why no mention of that? If
>>>that was intended, it's simple in English:
>>>"Our product will bear up to 1MM cycles of extreme load"
>>>"Our product sustained 1MM cycles of industry standard fatigue testing"
>>> "Our competitor's product failed at merely 985,00 cycles"
>>>
>>>That is not what they wrote.
>>>
>>>I was also curious about the smaller-section aero spoke showing 2.3x
>>>longevity. That does not make any sense at all as _less_ material just
>>>couldn't bear _2.3x_ the load cycles. And the values are nice and round,
>>>too, 1,000,000 and 2,300,000. Convenient numbers, those, as they just
>>>happen to fit the price curve but do they relate to the actual product?
>>>
>>>It is not at all clear to me.

>>
>> Dear Andrew,
>>
>> Here's the data, culled from the Sapim site's individual pages,
>> ordered by cycles to failure:
>>
>> per n/mm^2
>> 64 x middle test cycles
>> model 260mm hub x middle x rim strength to failure
>> ----- ---- --- --------- --- -------- ---------
>> cxray 278g 2.0 x (0.9 x 2.3) x 2.0 1600 3,500,000
>> strong 430g 2.3 x 2.0 x 2.0 1400 1,600,000
>> laser 279g 2.0 x 1.5 x 2.0 1500 1,250,000
>> cx 423g 2.0 x (1.3 x 2.8) x 2.0 1200 1,220,000
>> race 360g 2.0 x 1.8 x 2.0 1350 980,000
>> leader 424g 2.0 x 2.0 x 2.0 1080 870,000
>>
>> Note that the cycles to failure aren't all that rounded.
>>
>> Judging by weight, the cxray aero spokes (278 grams) are probably
>> flattened laser butted spokes (279 grams).
>>
>> The cx aero spokes (423 grams) are probably flattened straight leader
>> spokes (424 grams).
>>
>> Since the round spokes are drawn to thin their midsections, their
>> midsections work harden and strengthen, with the n/mm^2 strength
>> rising (1080, 1350, 1500) as they grow thinner (2.0, 1.8, 1.5).

>
>What does the strength of the middle section have to do with fatigue?


[snip]

Dear Gary,

Interspersed comments not only make further replies difficult, but
often reveal a reluctance to read the whole post:

"The aero spokes also show greater midsection strength than the round
spokes from which they were probably stamped, again due to work
hardening, but this doesn't affect the unworked elbows where they
probably fail."

I was pointing out the consistency of the Sapim data. The more the
spokes are flattened or drawn, the greater the strength of their
mid-sections.

Cheers,

Carl Fogel
 
On Thu, 08 Feb 2007 17:46:03 -0700, carlfogel wrote:

> On Thu, 08 Feb 2007 14:45:32 -0600, Gary Young <[email protected]>
> wrote:
>
>>On Thu, 08 Feb 2007 12:50:20 -0700, carlfogel wrote:
>>
>>> On Thu, 08 Feb 2007 11:50:03 -0600, A Muzi <[email protected]>
>>> wrote:
>>>
>>>>>>>>> Dear Johan,
>>>>>>>>> Your question has come up before--it would be nice if Sapim would
>>>>>>>>> include a description of its testing.
>>>>>>>>> The Sapim spoke fatigue tests must use exaggerated stress cycles
>>>>>>>>> because it takes literally years to break spokes in continuous testing
>>>>>>>>> under ordinary low stress cycles--no spoke manufacturer sells spokes
>>>>>>>>> that are expected to fail in only 2,000 km of normal riding.
>>>>
>>>>>>>> I have no doubt that it is an exaggerated test cycle. But don't they
>>>>>>>> know in marketing hype one doesn't cite exaggerated examples, but best
>>>>>>>> case scenarios. I have SAPIM-built wheels which have done closer to 40
>>>>>>>> 000 kms. Why would a company misrepresent itself so much? I don't
>>>>>>>> understand and I don't suppose there is a rational answer for this.
>>>>
>>>>jim beam wrote:
>>>>>>> there /is/ a rational answer for this - almost all fatigue testing is
>>>>>>> done on an accelerated basis. since that's standard practice, their
>>>>>>> reporting is simply consistent with standard practice.-
>>>>
>>>>> Johan Bornman wrote:
>>>>>> You miss the point, being scientifically rational doesn't make sense
>>>>>> in this brochure.
>>>>>
>>>>jim beam wrote:
>>>>> it makes more sense than being scientifically irrational or making
>>>>> claims like "our spokes last forever if you don't use them", which is
>>>>> basically where you end up if you exclude presentation of normal data.
>>>>
>>>>> Johan Bornman wrote:
>>>>>> It is intended for consumers who can quickly do the
>>>>>> sums and discover that SAPIM kind of suggests a useful life of 1 000
>>>>>> kms for wheels built with its spokes.
>>>>
>>>>> Johan Bornman wrote:
>>>>>> A good copywriter with the
>>>>>> company's interest at heart would have phrased the exercise completely
>>>>>> differently.
>>>>
>>>>jim beam wrote:
>>>>> bottom line, while /you/ may not appreciate the format, the fact is,
>>>>> sapim are the only ones that dare to present fatigue data. that should
>>>>> tell you more than anything else.
>>>>
>>>>Maybe.
>>>>Or were those numbers pulled out of the air? "2000k lifetime" is
>>>>nonsensical to anyone with familiarity. But any ad writer would eagerly
>>>>write "one million cycles" for any product - especially if he weren't
>>>>burdened by knowledge of bikes or spokes.
>>>>
>>>>If they meant 1MM cycles at an extreme load, why no mention of that? If
>>>>that was intended, it's simple in English:
>>>>"Our product will bear up to 1MM cycles of extreme load"
>>>>"Our product sustained 1MM cycles of industry standard fatigue testing"
>>>> "Our competitor's product failed at merely 985,00 cycles"
>>>>
>>>>That is not what they wrote.
>>>>
>>>>I was also curious about the smaller-section aero spoke showing 2.3x
>>>>longevity. That does not make any sense at all as _less_ material just
>>>>couldn't bear _2.3x_ the load cycles. And the values are nice and round,
>>>>too, 1,000,000 and 2,300,000. Convenient numbers, those, as they just
>>>>happen to fit the price curve but do they relate to the actual product?
>>>>
>>>>It is not at all clear to me.
>>>
>>> Dear Andrew,
>>>
>>> Here's the data, culled from the Sapim site's individual pages,
>>> ordered by cycles to failure:
>>>
>>> per n/mm^2
>>> 64 x middle test cycles
>>> model 260mm hub x middle x rim strength to failure
>>> ----- ---- --- --------- --- -------- ---------
>>> cxray 278g 2.0 x (0.9 x 2.3) x 2.0 1600 3,500,000
>>> strong 430g 2.3 x 2.0 x 2.0 1400 1,600,000
>>> laser 279g 2.0 x 1.5 x 2.0 1500 1,250,000
>>> cx 423g 2.0 x (1.3 x 2.8) x 2.0 1200 1,220,000
>>> race 360g 2.0 x 1.8 x 2.0 1350 980,000
>>> leader 424g 2.0 x 2.0 x 2.0 1080 870,000
>>>
>>> Note that the cycles to failure aren't all that rounded.
>>>
>>> Judging by weight, the cxray aero spokes (278 grams) are probably
>>> flattened laser butted spokes (279 grams).
>>>
>>> The cx aero spokes (423 grams) are probably flattened straight leader
>>> spokes (424 grams).
>>>
>>> Since the round spokes are drawn to thin their midsections, their
>>> midsections work harden and strengthen, with the n/mm^2 strength
>>> rising (1080, 1350, 1500) as they grow thinner (2.0, 1.8, 1.5).

>>
>>What does the strength of the middle section have to do with fatigue?

>
> [snip]
>
> Dear Gary,
>
> Interspersed comments not only make further replies difficult, but
> often reveal a reluctance to read the whole post:
>
> "The aero spokes also show greater midsection strength than the round
> spokes from which they were probably stamped, again due to work
> hardening, but this doesn't affect the unworked elbows where they
> probably fail."


Actually, I did read the whole post, but I may have been misreading that
sentence as saying that spokes are vulnerable at the elbows because they
are not work hardened in that area (which is different from saying
that midsection strength has nothing to do with failure of elbows).

>
> I was pointing out the consistency of the Sapim data. The more the
> spokes are flattened or drawn, the greater the strength of their
> mid-sections.


As I mentioned, I'd like to see an explanation of the strength results for
the Strong spokes, which appear inconsistent on the face of things. I
guess I'm hoping an expert can explain if the fact that the Strong is
drawn down slightly more than the Race (0.3 versus 0.2 mm) can compensate
for the advantage the Race has in being 0.2 mm thinner in its midsection.

>
> Cheers,
>
> Carl Fogel
 
On Feb 8, 12:50 pm, [email protected] wrote:
> Jim Beam's explanation for why the flattened aero spokes last longer
> makes sense--their angle to the elbow lets them put less bending
> strain on the elbow. That is, when an aero spoke bends at the elbow,
> it probably does less bending at the elbow and more bending at the
> thin section than a straight-section spoke, spreading the stress out.
>
> The round spokes (leader 2.0, race 1.8, laser 1.5) probably last
> longer in testing as their midsections grow thinner for the same
> reason--as the thinner midsection bends a little more, it puts a
> little less bending strain on the elbow, spreading the bending stress
> over more spoke.


Thanks for the info Carl, but my conclusion from this is that the test
is not a good representation of a wheel. In a well built wheel the
spoke elbow is not going to be moving around (changing angle)... and
if the spoke elbow isn't moving, then the bending moment of the middle
of the spoke is irrelevant. Didn't jim say the load cycle was 0-80kg?
Going to zero will surely amplify the elbow effect if it is not
completely in line with the load path. If it had been 50-150kg I'd be
more convinced. Under normal loads that a wheel sees, the spoke will
always be in tension and the elbow will be "stuck" in place. Sapim got
a results that favor low bending moments in the center of the spoke,
but I don't believe that has anything to do with reality. Unless Sapim
did something "magical" with the CX-Ray, it's real performance in a
wheel should be the same as the Laser. Reducing the cross section in
the middle of the spoke *does* help... but the bending moment is
irrelevant.
 
On Thu, 08 Feb 2007 16:46:22 -0600, Ben C wrote:

> On 2007-02-08, Gary Young <[email protected]> wrote:
>> On Thu, 08 Feb 2007 15:45:40 -0600, Ben C wrote:
>>
>>> On 2007-02-08, Gary Young <[email protected]> wrote: [...]
>>>> I find it interesting that "jim beam" is advancing his bending moment
>>>> thesis, since it seems to show the importance of correcting the spoke
>>>> line, something he's dismissed in the past. He's suggesting that the
>>>> spoke that bends most to have the right spoke line is the one that's
>>>> going to last the longest.
>>>
>>> I think the distinction is between bending in the midsection and
>>> bending at the elbow. If the spoke flexes a bit more at the midsection
>>> it can flex a bit less at the elbow. It's unlikely to fail at the
>>> midsection, and if it is able to flex there, is less likely to fail at
>>> the elbow.

>>
>> I don't see how that goes against my point.

>
> I probably didn't understand your point.
>
>> The mere fact that it's a good thing that there's some bending going on
>> means that the natural spoke line inherent in the materials (elbow
>> bend, etc.) is not optimized.

>
> I'm not sure I understand it now.
>
> The way I see it bending at the elbow is bad. Good spoke line at the
> elbow (and correct shank length) reduces elbow bending. Flexible
> midsection also reduces elbow bending.
>
> Now I thought your point was: correcting the spoke line will reduce
> elbow flex in use, so why does jim beam tell us not to correct the spoke
> line?
>
> I don't think jim beam has ever denied that poor spoke line
> (specifically spoke not flush) contributes to fatigue. It was from him
> that I first heard about the long-shank problems with DT spokes. The
> point is not that a good spoke line is not desirable, just how should it
> be achieved: manually (thumb, crankarm, etc.), or just tension the
> spokes and let them do their thing.
>
> One thing though is that jim beam has said IIRC that premature spoke
> line correction may actually promote fatigue. I'm less clear on the
> reasons for that.


Because bending the spoke to yield introduces stresses (whether he would
call them residual or not I don't know) that he doesn't think stress
relieving removes.

Maybe this will make my main point more clear:

Imagine you have spokes with a perfect line -- they sit snugly in the
holes, hug closely to the flange as they exit the hub, and enter the
nipples at the same angle at which the nipples are seated in the rim. In
that case, where would the bending stress come from? In what way would it
matter if the spoke is straight gauge or butted? The spoke would change in
length as it is loaded and unloaded, but would it bend? (I confess that my
theory is all wet if such axial changes are considered bending.) Do
straight-pull, elbow-less spokes experience bending stress?

Both Jobst and "jim beam" agree that butted spokes have better fatigue
life, but Jobst ascribes it to something I don't think involves bending.
On page 47 of the Bicycle Wheel, he says:

"[T]he most valuable contribution of swaging is that peak stresses are
absorbed in the straight midsection rather than concentrated in the
threads and elbow, thereby substantially reducing fatigue failures. Swaged
spokes act like strain screws commonly used in high-performance machinery."

If I understand Jobst correctly, butted spokes give that benefit
whether they have elbows or are straight-pull, because both types of spoke
are subject to peak stresses as the wheel makes its revolutions.

I believe "jim beam" has agreed that butted spokes do absorb peak
stresses, but has said that the more important benefit of butted
spokes is that they reduce bending stresses. Where do those stresses come
from unless there is a problem with the spoke line?

He has repeatedly argued that if one builds wheels according to
manufacturer spec, the spoke line will be optimized. That can't be
completely true, or there would be no bending stresses that would
give butted spokes an advantage over straight-gauge spokes.

Now I suppose he can argue back that a slight misalignment is bad, but a
minor problem. That doesn't seem to be supported by the Sapim data, which
show a threefold increase in fatigue life in some cases, allegedly
because butted spokes reduce bending stresses.

Either Sapim's data is bad and/or jim beam's theory is bad (if you ask
me it's probably both).
 
Gary Young wrote:
> On Thu, 08 Feb 2007 16:46:22 -0600, Ben C wrote:
>
>> On 2007-02-08, Gary Young <[email protected]> wrote:
>>> On Thu, 08 Feb 2007 15:45:40 -0600, Ben C wrote:
>>>
>>>> On 2007-02-08, Gary Young <[email protected]> wrote: [...]
>>>>> I find it interesting that "jim beam" is advancing his bending moment
>>>>> thesis, since it seems to show the importance of correcting the spoke
>>>>> line, something he's dismissed in the past. He's suggesting that the
>>>>> spoke that bends most to have the right spoke line is the one that's
>>>>> going to last the longest.
>>>> I think the distinction is between bending in the midsection and
>>>> bending at the elbow. If the spoke flexes a bit more at the midsection
>>>> it can flex a bit less at the elbow. It's unlikely to fail at the
>>>> midsection, and if it is able to flex there, is less likely to fail at
>>>> the elbow.
>>> I don't see how that goes against my point.

>> I probably didn't understand your point.
>>
>>> The mere fact that it's a good thing that there's some bending going on
>>> means that the natural spoke line inherent in the materials (elbow
>>> bend, etc.) is not optimized.

>> I'm not sure I understand it now.
>>
>> The way I see it bending at the elbow is bad. Good spoke line at the
>> elbow (and correct shank length) reduces elbow bending. Flexible
>> midsection also reduces elbow bending.
>>
>> Now I thought your point was: correcting the spoke line will reduce
>> elbow flex in use, so why does jim beam tell us not to correct the spoke
>> line?
>>
>> I don't think jim beam has ever denied that poor spoke line
>> (specifically spoke not flush) contributes to fatigue. It was from him
>> that I first heard about the long-shank problems with DT spokes. The
>> point is not that a good spoke line is not desirable, just how should it
>> be achieved: manually (thumb, crankarm, etc.), or just tension the
>> spokes and let them do their thing.
>>
>> One thing though is that jim beam has said IIRC that premature spoke
>> line correction may actually promote fatigue. I'm less clear on the
>> reasons for that.

>
> Because bending the spoke to yield introduces stresses (whether he would
> call them residual or not I don't know) that he doesn't think stress
> relieving removes.
>
> Maybe this will make my main point more clear:
>
> Imagine you have spokes with a perfect line -- they sit snugly in the
> holes, hug closely to the flange as they exit the hub, and enter the
> nipples at the same angle at which the nipples are seated in the rim. In
> that case, where would the bending stress come from? In what way would it
> matter if the spoke is straight gauge or butted? The spoke would change in
> length as it is loaded and unloaded, but would it bend? (I confess that my
> theory is all wet if such axial changes are considered bending.) Do
> straight-pull, elbow-less spokes experience bending stress?
>
> Both Jobst and "jim beam" agree that butted spokes have better fatigue
> life, but Jobst ascribes it to something I don't think involves bending.
> On page 47 of the Bicycle Wheel, he says:
>
> "[T]he most valuable contribution of swaging is that peak stresses are
> absorbed in the straight midsection rather than concentrated in the
> threads and elbow, thereby substantially reducing fatigue failures. Swaged
> spokes act like strain screws commonly used in high-performance machinery."
>
> If I understand Jobst correctly, butted spokes give that benefit
> whether they have elbows or are straight-pull, because both types of spoke
> are subject to peak stresses as the wheel makes its revolutions.
>
> I believe "jim beam" has agreed that butted spokes do absorb peak
> stresses, but has said that the more important benefit of butted
> spokes is that they reduce bending stresses. Where do those stresses come
> from unless there is a problem with the spoke line?
>
> He has repeatedly argued that if one builds wheels according to
> manufacturer spec, the spoke line will be optimized. That can't be
> completely true, or there would be no bending stresses that would
> give butted spokes an advantage over straight-gauge spokes.
>
> Now I suppose he can argue back that a slight misalignment is bad, but a
> minor problem. That doesn't seem to be supported by the Sapim data, which
> show a threefold increase in fatigue life in some cases, allegedly
> because butted spokes reduce bending stresses.
>
> Either Sapim's data is bad and/or jim beam's theory is bad (if you ask
> me it's probably both).


gary, you need to understand a fundamental principle of mechanics: the
shortest distance between two points is a straight line. with a spoke,
one end is /not/ anchored axial to the spoke, the straight line, it's
anchored about 3mm outboard. no straight line means there fundamentally
/must/ be bending at the transition point, the spoke elbow.

now, if you understand that principle, think next on how to address the
/degree/ of bending. one component of this, in addition to offset,
spoke diameter, material, etc., is the stiffness of the long part of the
spoke, because that's the bit against which the bending moment is
exerted. if that too is stiff, the stress at the elbow can rise. if
it's not, it can't. obviously, skinny spokes offer much less resistance
to the bending moment, so stress doesn't rise as much, hence less fatigue.
 
On Feb 8, 9:50 pm, [email protected] wrote:
>
> Jim Beam's explanation for why the flattened aero spokes last longer
> makes sense--their angle to the elbow lets them put less bending
> strain on the elbow. That is, when an aero spoke bends at the elbow,
> it probably does less bending at the elbow and more bending at the
> thin section than a straight-section spoke, spreading the stress out.
>

That's not the way I see it. At the elbow and 30mm onwards, these
(flattened aero) spokes are still plain round spokes. I think it does
as much bending at the elbow, for a given tension, as a straight-gauge
or butted spoke.

JB
 
The CX-Ray has a lesser cross-sectional area in the middle of the spoke then
the Race spoke, so for a given stress used in testing the elongation will be
greater there for the CX-Ray, and there will be less strain (in the
engineering sense) at the head of the spoke, which is where most spoke
breakages occur. Hence the greater expected life of the ligher spoke.

Nick

"Johan Bornman" <[email protected]> wrote in message
news:[email protected]...
> Jim Beam writes:
>
>> what does weight have to do with it other than being a side-effect of
>> having more material?

>
> Everything, the more weight for a given length, the more material. I
> assume (with confidence) that they use the same material - 18/8
> stainless steel, as the brochure states.
 
On Thu, 08 Feb 2007 19:30:35 -0600, Gary Young <[email protected]>
wrote:

>On Thu, 08 Feb 2007 17:46:03 -0700, carlfogel wrote:
>
>> On Thu, 08 Feb 2007 14:45:32 -0600, Gary Young <[email protected]>
>> wrote:
>>
>>>On Thu, 08 Feb 2007 12:50:20 -0700, carlfogel wrote:
>>>
>>>> On Thu, 08 Feb 2007 11:50:03 -0600, A Muzi <[email protected]>
>>>> wrote:
>>>>
>>>>>>>>>> Dear Johan,
>>>>>>>>>> Your question has come up before--it would be nice if Sapim would
>>>>>>>>>> include a description of its testing.
>>>>>>>>>> The Sapim spoke fatigue tests must use exaggerated stress cycles
>>>>>>>>>> because it takes literally years to break spokes in continuous testing
>>>>>>>>>> under ordinary low stress cycles--no spoke manufacturer sells spokes
>>>>>>>>>> that are expected to fail in only 2,000 km of normal riding.
>>>>>
>>>>>>>>> I have no doubt that it is an exaggerated test cycle. But don't they
>>>>>>>>> know in marketing hype one doesn't cite exaggerated examples, but best
>>>>>>>>> case scenarios. I have SAPIM-built wheels which have done closer to 40
>>>>>>>>> 000 kms. Why would a company misrepresent itself so much? I don't
>>>>>>>>> understand and I don't suppose there is a rational answer for this.
>>>>>
>>>>>jim beam wrote:
>>>>>>>> there /is/ a rational answer for this - almost all fatigue testing is
>>>>>>>> done on an accelerated basis. since that's standard practice, their
>>>>>>>> reporting is simply consistent with standard practice.-
>>>>>
>>>>>> Johan Bornman wrote:
>>>>>>> You miss the point, being scientifically rational doesn't make sense
>>>>>>> in this brochure.
>>>>>>
>>>>>jim beam wrote:
>>>>>> it makes more sense than being scientifically irrational or making
>>>>>> claims like "our spokes last forever if you don't use them", which is
>>>>>> basically where you end up if you exclude presentation of normal data.
>>>>>
>>>>>> Johan Bornman wrote:
>>>>>>> It is intended for consumers who can quickly do the
>>>>>>> sums and discover that SAPIM kind of suggests a useful life of 1 000
>>>>>>> kms for wheels built with its spokes.
>>>>>
>>>>>> Johan Bornman wrote:
>>>>>>> A good copywriter with the
>>>>>>> company's interest at heart would have phrased the exercise completely
>>>>>>> differently.
>>>>>
>>>>>jim beam wrote:
>>>>>> bottom line, while /you/ may not appreciate the format, the fact is,
>>>>>> sapim are the only ones that dare to present fatigue data. that should
>>>>>> tell you more than anything else.
>>>>>
>>>>>Maybe.
>>>>>Or were those numbers pulled out of the air? "2000k lifetime" is
>>>>>nonsensical to anyone with familiarity. But any ad writer would eagerly
>>>>>write "one million cycles" for any product - especially if he weren't
>>>>>burdened by knowledge of bikes or spokes.
>>>>>
>>>>>If they meant 1MM cycles at an extreme load, why no mention of that? If
>>>>>that was intended, it's simple in English:
>>>>>"Our product will bear up to 1MM cycles of extreme load"
>>>>>"Our product sustained 1MM cycles of industry standard fatigue testing"
>>>>> "Our competitor's product failed at merely 985,00 cycles"
>>>>>
>>>>>That is not what they wrote.
>>>>>
>>>>>I was also curious about the smaller-section aero spoke showing 2.3x
>>>>>longevity. That does not make any sense at all as _less_ material just
>>>>>couldn't bear _2.3x_ the load cycles. And the values are nice and round,
>>>>>too, 1,000,000 and 2,300,000. Convenient numbers, those, as they just
>>>>>happen to fit the price curve but do they relate to the actual product?
>>>>>
>>>>>It is not at all clear to me.
>>>>
>>>> Dear Andrew,
>>>>
>>>> Here's the data, culled from the Sapim site's individual pages,
>>>> ordered by cycles to failure:
>>>>
>>>> per n/mm^2
>>>> 64 x middle test cycles
>>>> model 260mm hub x middle x rim strength to failure
>>>> ----- ---- --- --------- --- -------- ---------
>>>> cxray 278g 2.0 x (0.9 x 2.3) x 2.0 1600 3,500,000
>>>> strong 430g 2.3 x 2.0 x 2.0 1400 1,600,000
>>>> laser 279g 2.0 x 1.5 x 2.0 1500 1,250,000
>>>> cx 423g 2.0 x (1.3 x 2.8) x 2.0 1200 1,220,000
>>>> race 360g 2.0 x 1.8 x 2.0 1350 980,000
>>>> leader 424g 2.0 x 2.0 x 2.0 1080 870,000
>>>>
>>>> Note that the cycles to failure aren't all that rounded.
>>>>
>>>> Judging by weight, the cxray aero spokes (278 grams) are probably
>>>> flattened laser butted spokes (279 grams).
>>>>
>>>> The cx aero spokes (423 grams) are probably flattened straight leader
>>>> spokes (424 grams).
>>>>
>>>> Since the round spokes are drawn to thin their midsections, their
>>>> midsections work harden and strengthen, with the n/mm^2 strength
>>>> rising (1080, 1350, 1500) as they grow thinner (2.0, 1.8, 1.5).
>>>
>>>What does the strength of the middle section have to do with fatigue?

>>
>> [snip]
>>
>> Dear Gary,
>>
>> Interspersed comments not only make further replies difficult, but
>> often reveal a reluctance to read the whole post:
>>
>> "The aero spokes also show greater midsection strength than the round
>> spokes from which they were probably stamped, again due to work
>> hardening, but this doesn't affect the unworked elbows where they
>> probably fail."

>
>Actually, I did read the whole post, but I may have been misreading that
>sentence as saying that spokes are vulnerable at the elbows because they
>are not work hardened in that area (which is different from saying
>that midsection strength has nothing to do with failure of elbows).
>
>>
>> I was pointing out the consistency of the Sapim data. The more the
>> spokes are flattened or drawn, the greater the strength of their
>> mid-sections.

>
>As I mentioned, I'd like to see an explanation of the strength results for
>the Strong spokes, which appear inconsistent on the face of things. I
>guess I'm hoping an expert can explain if the fact that the Strong is
>drawn down slightly more than the Race (0.3 versus 0.2 mm) can compensate
>for the advantage the Race has in being 0.2 mm thinner in its midsection.
>
>>
>> Cheers,
>>
>> Carl Fogel


Dear Gary,

The midsection strength and the fatigue resistance are probably
separate matters.

The Strong probably resists fatigue better because its elbow is 2.3 mm
thick, while the other spokes use 2.0 mm elbows.

As for midsection work-hardening strength increases, I suspect that
calculating the work-hardening effect of drawing different initial
spoke thicknesses down different amounts is trickier than a simple
comparison of raw mm reduction, or even of cross section.

For one thing, the different-thickness straight spoke blanks probably
start out with different initial strengths due to the work hardening
effect that got them to their initial size. That is, a 2.3 mm straight
spoke probably starts out with a different strength than a 2.0 mm
straight spoke.

Cheers,

Carl Fogel
 
On 8 Feb 2007 22:08:47 -0800, "Johan Bornman" <[email protected]>
wrote:

>On Feb 8, 9:50 pm, [email protected] wrote:
>>
>> Jim Beam's explanation for why the flattened aero spokes last longer
>> makes sense--their angle to the elbow lets them put less bending
>> strain on the elbow. That is, when an aero spoke bends at the elbow,
>> it probably does less bending at the elbow and more bending at the
>> thin section than a straight-section spoke, spreading the stress out.
>>

>That's not the way I see it. At the elbow and 30mm onwards, these
>(flattened aero) spokes are still plain round spokes. I think it does
>as much bending at the elbow, for a given tension, as a straight-gauge
>or butted spoke.
>
>JB


Dear Johan,

The thick, identical 2.0 mm elbows must bend against some resistance.

If the resistance to bending in that plane is a straight spoke
section, then the resistance must be greater than if it is provided by
a blade angled for easy bending. This spreads the total stress out
over more material.

To exaggerate the effect, think of a spoke with a zig-zag elbow:

A A______________
/\_________ versus /
/ B /

Elbows A & B should last longer because the stress is spread out over
two bends, but the sharp angles would probably ruin things.

When the angles are scaled down to the almost imperceptible flexing of
a normal elbow and the round/blade junction, there seems to be a large
and measurable improvement in fatigue resistance over a round/round
junction.

Cheers,

Carl Fogel
 
On Fri, 09 Feb 2007 00:20:26 -0700, carlfogel wrote:

> On Thu, 08 Feb 2007 19:30:35 -0600, Gary Young <[email protected]>
> wrote:
>
>>On Thu, 08 Feb 2007 17:46:03 -0700, carlfogel wrote:
>>
>>> On Thu, 08 Feb 2007 14:45:32 -0600, Gary Young <[email protected]>
>>> wrote:
>>>
>>>>On Thu, 08 Feb 2007 12:50:20 -0700, carlfogel wrote:
>>>>
>>>>> On Thu, 08 Feb 2007 11:50:03 -0600, A Muzi <[email protected]>
>>>>> wrote:
>>>>>
>>>>>>>>>>> Dear Johan,
>>>>>>>>>>> Your question has come up before--it would be nice if Sapim would
>>>>>>>>>>> include a description of its testing.
>>>>>>>>>>> The Sapim spoke fatigue tests must use exaggerated stress cycles
>>>>>>>>>>> because it takes literally years to break spokes in continuous testing
>>>>>>>>>>> under ordinary low stress cycles--no spoke manufacturer sells spokes
>>>>>>>>>>> that are expected to fail in only 2,000 km of normal riding.
>>>>>>
>>>>>>>>>> I have no doubt that it is an exaggerated test cycle. But don't they
>>>>>>>>>> know in marketing hype one doesn't cite exaggerated examples, but best
>>>>>>>>>> case scenarios. I have SAPIM-built wheels which have done closer to 40
>>>>>>>>>> 000 kms. Why would a company misrepresent itself so much? I don't
>>>>>>>>>> understand and I don't suppose there is a rational answer for this.
>>>>>>
>>>>>>jim beam wrote:
>>>>>>>>> there /is/ a rational answer for this - almost all fatigue testing is
>>>>>>>>> done on an accelerated basis. since that's standard practice, their
>>>>>>>>> reporting is simply consistent with standard practice.-
>>>>>>
>>>>>>> Johan Bornman wrote:
>>>>>>>> You miss the point, being scientifically rational doesn't make sense
>>>>>>>> in this brochure.
>>>>>>>
>>>>>>jim beam wrote:
>>>>>>> it makes more sense than being scientifically irrational or making
>>>>>>> claims like "our spokes last forever if you don't use them", which is
>>>>>>> basically where you end up if you exclude presentation of normal data.
>>>>>>
>>>>>>> Johan Bornman wrote:
>>>>>>>> It is intended for consumers who can quickly do the
>>>>>>>> sums and discover that SAPIM kind of suggests a useful life of 1 000
>>>>>>>> kms for wheels built with its spokes.
>>>>>>
>>>>>>> Johan Bornman wrote:
>>>>>>>> A good copywriter with the
>>>>>>>> company's interest at heart would have phrased the exercise completely
>>>>>>>> differently.
>>>>>>
>>>>>>jim beam wrote:
>>>>>>> bottom line, while /you/ may not appreciate the format, the fact is,
>>>>>>> sapim are the only ones that dare to present fatigue data. that should
>>>>>>> tell you more than anything else.
>>>>>>
>>>>>>Maybe.
>>>>>>Or were those numbers pulled out of the air? "2000k lifetime" is
>>>>>>nonsensical to anyone with familiarity. But any ad writer would eagerly
>>>>>>write "one million cycles" for any product - especially if he weren't
>>>>>>burdened by knowledge of bikes or spokes.
>>>>>>
>>>>>>If they meant 1MM cycles at an extreme load, why no mention of that? If
>>>>>>that was intended, it's simple in English:
>>>>>>"Our product will bear up to 1MM cycles of extreme load"
>>>>>>"Our product sustained 1MM cycles of industry standard fatigue testing"
>>>>>> "Our competitor's product failed at merely 985,00 cycles"
>>>>>>
>>>>>>That is not what they wrote.
>>>>>>
>>>>>>I was also curious about the smaller-section aero spoke showing 2.3x
>>>>>>longevity. That does not make any sense at all as _less_ material just
>>>>>>couldn't bear _2.3x_ the load cycles. And the values are nice and round,
>>>>>>too, 1,000,000 and 2,300,000. Convenient numbers, those, as they just
>>>>>>happen to fit the price curve but do they relate to the actual product?
>>>>>>
>>>>>>It is not at all clear to me.
>>>>>
>>>>> Dear Andrew,
>>>>>
>>>>> Here's the data, culled from the Sapim site's individual pages,
>>>>> ordered by cycles to failure:
>>>>>
>>>>> per n/mm^2
>>>>> 64 x middle test cycles
>>>>> model 260mm hub x middle x rim strength to failure
>>>>> ----- ---- --- --------- --- -------- ---------
>>>>> cxray 278g 2.0 x (0.9 x 2.3) x 2.0 1600 3,500,000
>>>>> strong 430g 2.3 x 2.0 x 2.0 1400 1,600,000
>>>>> laser 279g 2.0 x 1.5 x 2.0 1500 1,250,000
>>>>> cx 423g 2.0 x (1.3 x 2.8) x 2.0 1200 1,220,000
>>>>> race 360g 2.0 x 1.8 x 2.0 1350 980,000
>>>>> leader 424g 2.0 x 2.0 x 2.0 1080 870,000
>>>>>
>>>>> Note that the cycles to failure aren't all that rounded.
>>>>>
>>>>> Judging by weight, the cxray aero spokes (278 grams) are probably
>>>>> flattened laser butted spokes (279 grams).
>>>>>
>>>>> The cx aero spokes (423 grams) are probably flattened straight leader
>>>>> spokes (424 grams).
>>>>>
>>>>> Since the round spokes are drawn to thin their midsections, their
>>>>> midsections work harden and strengthen, with the n/mm^2 strength
>>>>> rising (1080, 1350, 1500) as they grow thinner (2.0, 1.8, 1.5).
>>>>
>>>>What does the strength of the middle section have to do with fatigue?
>>>
>>> [snip]
>>>
>>> Dear Gary,
>>>
>>> Interspersed comments not only make further replies difficult, but
>>> often reveal a reluctance to read the whole post:
>>>
>>> "The aero spokes also show greater midsection strength than the round
>>> spokes from which they were probably stamped, again due to work
>>> hardening, but this doesn't affect the unworked elbows where they
>>> probably fail."

>>
>>Actually, I did read the whole post, but I may have been misreading that
>>sentence as saying that spokes are vulnerable at the elbows because they
>>are not work hardened in that area (which is different from saying
>>that midsection strength has nothing to do with failure of elbows).
>>
>>>
>>> I was pointing out the consistency of the Sapim data. The more the
>>> spokes are flattened or drawn, the greater the strength of their
>>> mid-sections.

>>
>>As I mentioned, I'd like to see an explanation of the strength results for
>>the Strong spokes, which appear inconsistent on the face of things. I
>>guess I'm hoping an expert can explain if the fact that the Strong is
>>drawn down slightly more than the Race (0.3 versus 0.2 mm) can compensate
>>for the advantage the Race has in being 0.2 mm thinner in its midsection.
>>
>>>
>>> Cheers,
>>>
>>> Carl Fogel

>
> Dear Gary,
>
> The midsection strength and the fatigue resistance are probably
> separate matters.
>
> The Strong probably resists fatigue better because its elbow is 2.3 mm
> thick, while the other spokes use 2.0 mm elbows.


Perhaps, but the brochure I linked to in another reply says that the
Strong derives its advantage from fitting in the spoke holes more
snugly than spokes with thinner elbows. That suggests to me that the
fatigue profile of other Sapim spokes could be improved by using hubs
with smaller spoke holes.
>
> As for midsection work-hardening strength increases, I suspect that
> calculating the work-hardening effect of drawing different initial spoke
> thicknesses down different amounts is trickier than a simple comparison
> of raw mm reduction, or even of cross section.
>

I agree that it may be complicated. It's certainly beyond me. That's why I
put a call out to experts.

> For one thing, the different-thickness straight spoke blanks probably
> start out with different initial strengths due to the work hardening
> effect that got them to their initial size. That is, a 2.3 mm straight
> spoke probably starts out with a different strength than a 2.0 mm
> straight spoke.
>

Are you suggesting that the Strong spokes are work-hardened up from a
smaller diameter? Do you really think that's plausible? Why not just start
with a 2.3 mm wire and then draw down the thin section?

> Cheers,
>
> Carl Fogel
 
Nick Payne wrote:
> The CX-Ray has a lesser cross-sectional area in the middle of the spoke then
> the Race spoke, so for a given stress used in testing the elongation will be
> greater there for the CX-Ray, and there will be less strain (in the
> engineering sense) at the head of the spoke,


if you apply a 100lb load, you apply a 100lb load, regardless of how
much elasticity there is further up the line. the difference in fatigue
lives /could/ come from greater elasticity in that it would affect how
the rim distorts and thus how much stress cycles for any individual
spoke, but that wouldn't affect fatigue life for a spoke in a testing
machine. otoh, elbow bending is proven by the fact that spokes fatigue
there, so what can affect the reaction to bending moment and thus
fatigue life? spoke bending stiffness. that's why many pre-built [i.e.
researched and redesigned] wheels use straight pull spokes - eliminates
the bend.

> which is where most spoke
> breakages occur. Hence the greater expected life of the ligher spoke.
>
> Nick
>
> "Johan Bornman" <[email protected]> wrote in message
> news:[email protected]...
>> Jim Beam writes:
>>
>>> what does weight have to do with it other than being a side-effect of
>>> having more material?

>> Everything, the more weight for a given length, the more material. I
>> assume (with confidence) that they use the same material - 18/8
>> stainless steel, as the brochure states.

>
>
 
On Thu, 08 Feb 2007 21:47:13 -0800, jim beam wrote:

> Gary Young wrote:
>> On Thu, 08 Feb 2007 16:46:22 -0600, Ben C wrote:
>>
>>> On 2007-02-08, Gary Young <[email protected]> wrote:
>>>> On Thu, 08 Feb 2007 15:45:40 -0600, Ben C wrote:
>>>>
>>>>> On 2007-02-08, Gary Young <[email protected]> wrote: [...]
>>>>>> I find it interesting that "jim beam" is advancing his bending moment
>>>>>> thesis, since it seems to show the importance of correcting the spoke
>>>>>> line, something he's dismissed in the past. He's suggesting that the
>>>>>> spoke that bends most to have the right spoke line is the one that's
>>>>>> going to last the longest.
>>>>> I think the distinction is between bending in the midsection and
>>>>> bending at the elbow. If the spoke flexes a bit more at the midsection
>>>>> it can flex a bit less at the elbow. It's unlikely to fail at the
>>>>> midsection, and if it is able to flex there, is less likely to fail at
>>>>> the elbow.
>>>> I don't see how that goes against my point.
>>> I probably didn't understand your point.
>>>
>>>> The mere fact that it's a good thing that there's some bending going on
>>>> means that the natural spoke line inherent in the materials (elbow
>>>> bend, etc.) is not optimized.
>>> I'm not sure I understand it now.
>>>
>>> The way I see it bending at the elbow is bad. Good spoke line at the
>>> elbow (and correct shank length) reduces elbow bending. Flexible
>>> midsection also reduces elbow bending.
>>>
>>> Now I thought your point was: correcting the spoke line will reduce
>>> elbow flex in use, so why does jim beam tell us not to correct the spoke
>>> line?
>>>
>>> I don't think jim beam has ever denied that poor spoke line
>>> (specifically spoke not flush) contributes to fatigue. It was from him
>>> that I first heard about the long-shank problems with DT spokes. The
>>> point is not that a good spoke line is not desirable, just how should it
>>> be achieved: manually (thumb, crankarm, etc.), or just tension the
>>> spokes and let them do their thing.
>>>
>>> One thing though is that jim beam has said IIRC that premature spoke
>>> line correction may actually promote fatigue. I'm less clear on the
>>> reasons for that.

>>
>> Because bending the spoke to yield introduces stresses (whether he would
>> call them residual or not I don't know) that he doesn't think stress
>> relieving removes.
>>
>> Maybe this will make my main point more clear:
>>
>> Imagine you have spokes with a perfect line -- they sit snugly in the
>> holes, hug closely to the flange as they exit the hub, and enter the
>> nipples at the same angle at which the nipples are seated in the rim. In
>> that case, where would the bending stress come from? In what way would it
>> matter if the spoke is straight gauge or butted? The spoke would change in
>> length as it is loaded and unloaded, but would it bend? (I confess that my
>> theory is all wet if such axial changes are considered bending.) Do
>> straight-pull, elbow-less spokes experience bending stress?
>>
>> Both Jobst and "jim beam" agree that butted spokes have better fatigue
>> life, but Jobst ascribes it to something I don't think involves bending.
>> On page 47 of the Bicycle Wheel, he says:
>>
>> "[T]he most valuable contribution of swaging is that peak stresses are
>> absorbed in the straight midsection rather than concentrated in the
>> threads and elbow, thereby substantially reducing fatigue failures. Swaged
>> spokes act like strain screws commonly used in high-performance machinery."
>>
>> If I understand Jobst correctly, butted spokes give that benefit
>> whether they have elbows or are straight-pull, because both types of spoke
>> are subject to peak stresses as the wheel makes its revolutions.
>>
>> I believe "jim beam" has agreed that butted spokes do absorb peak
>> stresses, but has said that the more important benefit of butted
>> spokes is that they reduce bending stresses. Where do those stresses come
>> from unless there is a problem with the spoke line?
>>
>> He has repeatedly argued that if one builds wheels according to
>> manufacturer spec, the spoke line will be optimized. That can't be
>> completely true, or there would be no bending stresses that would
>> give butted spokes an advantage over straight-gauge spokes.
>>
>> Now I suppose he can argue back that a slight misalignment is bad, but a
>> minor problem. That doesn't seem to be supported by the Sapim data, which
>> show a threefold increase in fatigue life in some cases, allegedly
>> because butted spokes reduce bending stresses.
>>
>> Either Sapim's data is bad and/or jim beam's theory is bad (if you ask
>> me it's probably both).

>
> gary, you need to understand a fundamental principle of mechanics: the
> shortest distance between two points is a straight line.


Condescending lesson -- on geometry really, not mechanics -- duly noted
and then ignored as being an issue never in dispute.

with a spoke,
> one end is /not/ anchored axial to the spoke, the straight line, it's
> anchored about 3mm outboard.


Yes, of course. That's why Jobst recommends correcting the spoke line.

no straight line means there fundamentally
> /must/ be bending at the transition point, the spoke elbow.
>

There must be bending even if the elbow is supported by the flange all
along its length? Please don't respond that the elbow is never
perfectly supported by the flange. I'm aware of that. What I want to
know is if you really think that there "fundamentally /must/ be
bending at the transition point." "Fundamentally must" suggests that there
will be bending even in ideal conditions. Is that what you mean to say?

> now, if you understand that principle, think next on how to address the
> /degree/ of bending. one component of this, in addition to offset,
> spoke diameter, material, etc., is the stiffness of the long part of the
> spoke, because that's the bit against which the bending moment is
> exerted. if that too is stiff, the stress at the elbow can rise. if
> it's not, it can't. obviously, skinny spokes offer much less resistance
> to the bending moment, so stress doesn't rise as much, hence less
> fatigue.


I'd like to think more about this, and it would help if I have your
answers to my question above.