Re: Spoke Failure--surface quality or stress relief?



On Sat, 09 Apr 2005 04:01:10 GMT, Werehatrack
<[email protected]> wrote:

[snip]

>And in my experience, the few failures I've seen at the nipple end
>were clearly due to the spoke being stressed beyond the yield point of
>the threaded section; the failure was obviously caused by tension in
>excess of the spoke's load capacity. (I will note that I have yet to
>see a spoke break at that point without a spokejam being involved,
>also.)


[snip]

Dear Werehatrack,

I've never even had a spoke fail at the nipple-end, but the
1984-1985 Wheelsmith spoke testing at Stanford found that
about 10% of tensioned spokes cycled to destruction (8 out
of 76) broke at the nipple instead of the elbow:

http://www.duke.edu/~hpgavin/papers/HPGavin-Wheel-Paper.pdf

It would be interesting to find out if some change in the
manufacturing process has led to a greater reduction in the
nipple-end failures in the last twenty years, or if I've
just been lucky.

Carl Fogel
 
Carl Fogel writes:

>> And in my experience, the few failures I've seen at the nipple end
>> were clearly due to the spoke being stressed beyond the yield point
>> of the threaded section; the failure was obviously caused by
>> tension in excess of the spoke's load capacity. (I will note that
>> I have yet to see a spoke break at that point without a spokejam
>> being involved, also.)


> I've never even had a spoke fail at the nipple-end, but the
> 1984-1985 Wheelsmith spoke testing at Stanford found that about 10%
> of tensioned spokes cycled to destruction (8 out of 76) broke at the
> nipple instead of the elbow:


> http://www.duke.edu/~hpgavin/papers/HPGavin-Wheel-Paper.pdf


> It would be interesting to find out if some change in the
> manufacturing process has led to a greater reduction in the
> nipple-end failures in the last twenty years, or if I've just been
> lucky.


Few people use large flange hubs with tangential spoking (that give a
larger spoke entry angle to the nipple) these days. It was my
experience that large flange hubs were popular before the 1980's and
most of the spoke thread failures I encountered were on such wheels.

Of course DT, Sapim and Wheelsmith among other good brands replaced
Robergel, Stella, Redialli, Berg and others that had poorer endurance.
Just the same, the percentage of such failures, regardless of how few,
probably hasn't changed much except for the absence of large flange
hubs and tangential spoking on them. This was done by people who
believed rider torque was a contributor to fatigue failures. It is
not (see "the Bicycle Wheel")

[email protected]
 
[email protected] (Sat, 09 Apr 2005 13:41:45) wrote:
> I've never even had a spoke fail at the nipple-end, but the
> 1984-1985 Wheelsmith spoke testing at Stanford found that about 10%
> of tensioned spokes cycled to destruction (8 out of 76) broke at the
> nipple instead of the elbow:


> http://www.duke.edu/~hpgavin/papers/ HPGavin-Wheel-Paper.pdf


> It would be interesting to find out if some change in the
> manufacturing process has led to a greater reduction in the nipple-
> end failures in the last twenty years, or if I've just been lucky.


I have an anecdote for you. Last year I built a wheel with a flexy rim
(MA-2), 32 straight gauge 1.8 mm spokes, and fairly high dish (Shimano
8/9 spd hub). I even used Jobstian methods to find the max tension and
stress relieve. To shorten a long story, I eventually replaced the drive
side spokes with 2 mm straight gauge. That was after 4-5 broken spokes,
all at the nipple, all on the drive side, and all JRA. No problems with
the wheel since.

Maybe I got a bad batch, maybe the tension was high enough to fatigue,
or maybe... something else. I was curious, but didn't have time to
investigate. At any rate, spoke failures at the nipple have not become
extinct.

--
Dave
 
> This was done by people who
> believed rider torque was a contributor to fatigue failures. It is
> not (see "the Bicycle Wheel")


Since I don't have a copy of it handy, do you mean cyclic fatigue failures,
or do you mean the final failure factor itself?

--
Phil, Squid-in-Training
 
Phil Lee writes:

>> This was done by people who believed rider torque was a contributor
>> to fatigue failures. It is not (see "the Bicycle Wheel")


> Since I don't have a copy of it handy, do you mean cyclic fatigue failures,
> or do you mean the final failure factor itself?


Pedaling torque loads are down in the "noise" being distributed among
all spokes. stress cycles from rider weight are bay far the greatest
load spokes see in cyclic loading. When spokes break from hitting an
obstacle, it is not the obstacle but the return to full tension after
the bump that breaks the spoke. All the significant loads on bicycle
spokes are tension relaxation that are followed by a return to full
tension.

[email protected]
 
[email protected] wrote:
> Phil Lee writes:
>
>
>>>This was done by people who believed rider torque was a contributor
>>>to fatigue failures. It is not (see "the Bicycle Wheel")

>
>
>>Since I don't have a copy of it handy, do you mean cyclic fatigue failures,
>>or do you mean the final failure factor itself?

>
>
> Pedaling torque loads are down in the "noise" being distributed among
> all spokes. stress cycles from rider weight are bay far the greatest
> load spokes see in cyclic loading. When spokes break from hitting an
> obstacle, it is not the obstacle but the return to full tension after
> the bump that breaks the spoke. All the significant loads on bicycle
> spokes are tension relaxation that are followed by a return to full
> tension.


except when there's lateral loads. you know, like when you have a dual
pivot brake caliper scraping when you're climbing a hill...

>
> [email protected]
 
Joe Riel <[email protected]> writes:

> Anecdotal evidence, be that what it may, is that scraping is primarily
> a problem with flexible frames. That is, the scraping is a result of
> the frame bending rather than the wheel bending. However, let's
> assume that that isn't the case, that it is entirely due to the wheel
> bending, and estimate an upper bound on tension increase. A quick
> test on my road bike (32 spokes) indicates that it doesn't take much
> force to move the wheel far enough to touch the brake pad. How much
> does the tension in the spokes increase? By plucking the spokes, the
> change in pitch seems to be less than a semitone (though my built-in
> pitch meter is notoriously out of calibration). Assume it is a full
> semitone (1/12 of an octave). Then the percent tension increase is
> (1 + 1/12)^2 - 1 = 15/144 = 17%. That's significant, but not hugely so.


The tension increase is actually even less: 12%. Remember that the
musical scale is logarithmic with a semitone representing scaling
frequency by 2^(1/12). Tension is proportional to frequency squared,
so a semitone represents scaling tension by 4^(1/12), or about 12.25%.
 
Wild Turkey writes:

>>>> This was done by people who believed rider torque was a
>>>> contributor to fatigue failures. It is not (see "the Bicycle
>>>> Wheel")


>>> Since I don't have a copy of it handy, do you mean cyclic fatigue

failures,>>> or do you mean the final failure factor itself?

>> Pedaling torque loads are down in the "noise" being distributed
>> among all spokes. stress cycles from rider weight are bay far the
>> greatest load spokes see in cyclic loading. When spokes break from
>> hitting an obstacle, it is not the obstacle but the return to full
>> tension after the bump that breaks the spoke. All the significant
>> loads on bicycle spokes are tension relaxation that are followed by
>> a return to full tension.


> except when there's lateral loads. you know, like when you have a
> dual pivot brake caliper scraping when you're climbing a hill...


You forget fast. This was explained in repetitive exchanges here not
long ago. When standing and pedaling, the lean of the bicycle does
not exceed the lateral spoke angle in a wheel so spokes on both sides
of the wheel in the load affected zone are slackened with the center
of pressure remaining between flanges, the high side losing more
tension than the low side. Tension on the remaining spokes remains
unchanged except for a small bias that arises from the forces of
lateral rim bending, shown in "the Bicycle Wheel" under the subject of
wheel collapse. The required lateral force can be assessed by
manually pushing against the wheel at the seat stays and sensing the
force it takes to move 1/10" clearance of a brake pad.

(At a dual pivot ratio of 5.6:1 that is 1 1/8 inch at the hand lever.)

Now stop harping on your forgetfulness!

[email protected]
 
[email protected] wrote:
> Wild Turkey writes:
>
>
>>>>>This was done by people who believed rider torque was a
>>>>>contributor to fatigue failures. It is not (see "the Bicycle
>>>>>Wheel")

>
>
>>>>Since I don't have a copy of it handy, do you mean cyclic fatigue

>
> failures,>>> or do you mean the final failure factor itself?
>
>
>>>Pedaling torque loads are down in the "noise" being distributed
>>>among all spokes. stress cycles from rider weight are bay far the
>>>greatest load spokes see in cyclic loading. When spokes break from
>>>hitting an obstacle, it is not the obstacle but the return to full
>>>tension after the bump that breaks the spoke. All the significant
>>>loads on bicycle spokes are tension relaxation that are followed by
>>>a return to full tension.

>
>
>>except when there's lateral loads. you know, like when you have a
>>dual pivot brake caliper scraping when you're climbing a hill...

>
>
> You forget fast. This was explained in repetitive exchanges here not
> long ago. When standing and pedaling, the lean of the bicycle does
> not exceed the lateral spoke angle in a wheel so spokes on both sides
> of the wheel in the load affected zone are slackened with the center
> of pressure remaining between flanges, the high side losing more
> tension than the low side. Tension on the remaining spokes remains
> unchanged except for a small bias that arises from the forces of
> lateral rim bending, shown in "the Bicycle Wheel" under the subject of
> wheel collapse. The required lateral force can be assessed by
> manually pushing against the wheel at the seat stays and sensing the
> force it takes to move 1/10" clearance of a brake pad.


that's the biggest pile of used buffalo food i've seen in a long while.
you reckon that a campy 10s hub with only 15.2mm offset drive side
rear always has the rim center remains between the flanges??? you're a
fraud! how anyone can so shamelessly and transparently fudge reality
like this is beyond me. you might think you can generally get away with
it on a more specialized subject like fatigue, but on something as
easily tested as this??? do you /really/ hold everyone on this planet
in such utter contempt that you think you can fabricate this nonsense
and get away with it?

anyway, regardless of flanges, /any/ deviation from center /does/ result
in an increase in tension on the side from which the lateral force
originates. do the ping test. you can't pull this fraud off jobst -
this test is too easily performed by the layman.

>
> (At a dual pivot ratio of 5.6:1 that is 1 1/8 inch at the hand lever.)


ah, the old trick - a quite true but /utterly/ irrelevant fact, as if it
had /anything/ to do with spoke tension.

>
> Now stop harping on your forgetfulness!
>
> [email protected]
 
Jim Smith <[email protected]> writes:

> The tension increase is actually even less: 12%. Remember that the
> musical scale is logarithmic with a semitone representing scaling
> frequency by 2^(1/12). Tension is proportional to frequency squared,
> so a semitone represents scaling tension by 4^(1/12), or about 12.25%.


Thanks, I don't know what I was thinking there.

Joe
 

Similar threads

M
Replies
63
Views
1K
D