On 10 May 2006 23:37:34 GMT,
[email protected]
wrote:
>Someone writes:
>
>>> I don't know how much 8 versus 10 degrees would affect
>>> calculations.
>
>>> Jobst has suggested testing a pair of spokes with carefully matched
>>> tension. I'm wondering about using another rim and higher initial
>>> tension and so forth, so please mention any changes that you think
>>> might be interesting.
>
>>> Of course, anyone with a tension gauge, some weights, some rope, a
>>> ceiling, and a way to anchor a bike safely to a workbench can test
>>> their own wheels.
>
>>> A padded clamp can squeeze the opposite pair of spokes for a rough
>>> check of the effect of imbalance.
>
>> I don't understand why the tension applied to those two spokes is
>> not dispersed throughout the wheel. It seems like this whole thread
>> treats the hub as if it were somehow fixed, when it actually floats
>> in the middle of the wheel, suspended by all the spokes. It seems
>> like applying all that force to those two spokes should change the
>> tension to some degree in all the spokes on the wheel.
>
>Yes, it does that but I think the thrust of this experiment has gotten
>lost. Let me try and isolate what I think has been tested from the
>outset of this experiment.
>
>Knowing the span (exposed length) of a spoke and its deflection for a
>known force applied at midspan, we can assess its tension in the
>deflected position. We can also readily measure spoke tension before
>deflecting the spoke. In this way we can assess the over-stress that
>is intended to cause high stress points spokes to yield.
>
>When spokes yield at high tension (stretch permanently) they do so
>only at the high stress points while the rest of the spoke is far from
>yield, however, local spots (where spokes would otherwise fail in
>fatigue) they yield and do not return to original length being relaxed
>of that stress. A good example of this is that such a spoke, when
>tensile tested does not break at the elbow or threads, but at some
>place in mid span.
>
>Typically, to assess tension at the height of a stress relief
>deflection, one can assess the geometry of, (the Carl Fogel test) for
>instance, a 100 lb lateral load deflecting a 300 mm long spoke 20 mm
>at midspan, which results in a spoke tension of 757 lb. That tension
>is significantly higher than at rest, which is usually around 200 lb
>and will yield any high stress locations in a spoke.
>
>[dimensions are approximate and rounded)
>
>What seems most confusing about this is that spokes do not physically
>stretch over their length but rather the rim gives inward. That does
>not alter the relationship of the sides of the force triangle. It
>could just as well be done between fixed anchors with a more elastic
>spoke instead of a fixed length spoke between elastic anchors (the
>rim). Spoke tension is derived from its midspan deflection by the
>lateral force applied there.
>
>Jobst Brandt
Dear Jobst,
"Typically, to assess tension at the height of a stress
relief deflection, one can assess the geometry of, (the Carl
Fogel test) for instance, a 100 lb lateral load deflecting a
300 mm long spoke 20 mm at midspan, which results in a spoke
tension of 757 lb."
???
Possibly I'm misunderstanding you, but whatever you're
trying to say could be read as saying that my test resulted
in a spoke tension of 757 pounds.
That's almost 500 pounds higher than what I measured.
The point of the experiment was to test the theoretical
prediction that Joe Riel had already cast doubt on by adding
rim stiffness--the geometric calculation that you mention
works only if the rim and hub fail to yield at all.
(They also require that the spoke rise to a high triangular
point, not the broader and lower curve of a hand-squeeze,
but that's another matter.)
Here's what I wrote describing the results of the second
experiment:
"Adding a 100-lb weight raised the upper/lower spoke tension
from 185/144 lbs to 263/273 lbs, only about 80 to 130
pounds."
These tensions were measured with a Park tension gauge.
The two spokes started with a tension of 185 and 144 pounds.
With a 100 lb weight hanging on a rope from the upper spoke
of a parallel horizontal pair and the lower spoke held up by
a ceiling rope, the two spokes rose to only about 270 pounds
of tension. Clamping the opposite pair of spokes made
scarcely any difference.
That's almost 500 pounds short of the 757 pound calculation.
Another way to look at it is that the 757 pound theoreticl
calculation is almost three times higher than the measured
result.
Possibly you didn't mean to give the impression that my test
agreed with the 757-pound calculation, but it's worth
emphasizing that my test basically showed that such
calculations that ignored rim stiffness grossly inflated the
tension rise by a factor of three in the single wheel tested
and that Joe Riel's calculations that included rim stiffness
were a much better model.
Cheers,
Carl Fogel
