Aaargh!

Quoted message said:

On Mon, 8 Jan 2007 05:14:22 +0000 (UTC), [email hidden]

(Luns Tee) said:

In article <[email hidden]>,

Quoted message said:

Given the same dimensions for the short spoke span in the gauge, the
different deflections must reflect how stiff or how elastic the spoke
material is.

The difference in bending stiffness does affect the measurement,
but there are oddities to DT's calibration charts that are fishy to me.

The 2.0mm titanium spoke is half as stiff in bending as a steel
spoke. It should deflect more than steel, and for a given range of
tensions, the range of deflections should be greater than for steel.

DT's charts do indeed indicate a greater range of deflections
for titanium, however note that the difference between titanium and
steel varies from the the titanium spoke having more deflection at low
tensions (smaller reading) as it should, to the steel spoke having
less deflection at high tensions, which would make no sense.

My best guess as to what's going on is that there's a constant
offset being introduced into the measurement, with the steel spoke's
measured deflection being a fixed amount more than what it should be,
and/or the titanium's being less. For larger displacements (low
tension), these are overwhelmed by the actual displacement
differences, but at high tensions, the offsets are overwhelming the
actual difference.

The steel spoke displacement is at most only 0.02mm greater than
the titanium in their tables. This is well within the range of offsets
observable from a bend/kink in the spoke, and is the offset that only
Jobst's design can zero for.

Other weirdness in the chart: compare the columns for
1.8/1.6 and 2.0/1.8 spokes. Comparing the straight gauge 1.8 to the
2.0/1.8 spoke, or the 1.8/1.5 with 2.0/1.5 spokes should give you some
feel for the tolerances in these measurement values.

Quoted message said:

Possibly the companies are applying complicated physics
equations, but I think that it's likely just measuring known loads.

I won't speak for what DT does, but the spreadsheet that Park
provides for their gauge does have equations embedded in it (credited to
C.S. Howat). The equation used has no basis in physics as far as I can
tell, but is based on observing that the measured curves, when
plotted on a semi-log scale, are curved somewhat like a parabola. This
may work reasonably well over some range of measurements, but wouldn't
extrapolate well, nor is it able to predict a-priori what the curve
for one type of spoke should be based on measurements for another.

This isn't to single out Park though. The calibration sheets
that FSA provides also give curves that are based on measurements. As
with Titanium vs. Stainless in DT's charts, the curves on FSA's charts
sometimes cross too, when I don't think they should. This is the
result of the curves being, I think, a polynomial fit of the measured
data. While it's a convenient fit to perform in Excel, and produces
smooth pretty curves, it doesn't reflect what the data should be doing.

-Luns

Dear Luns,

Aha! I went back to the DT pdf and proofread it!

The chart at the bottom has not one but two typos that goof up most of
the far right titantium 2 mm column.

I compared that weird column with the data in the individual section
for the 2mm titanium, which appears earlier in the pdf.

At 700n tension, the copy at the bottom skips a value. There should be
a 1.68 mm between 1.60 and 1.75, where I commented "big jump".

At 2000n tension, DT goofed again on the big chart at the bottom. It
should read 2.76 mm, not the 2.86 mm that duplicates the next entry.

I emailed DT Swiss USA about the typos. Anyone can make a tpyo, but
not everyone will fix a typo, so it will be interesting to see what
happens. I wonder if the tool comes with a plastic chart that
duplicates the errors, or if only the big chart at the bottom of the
pdf has the mistakes?

I may not be able to resist checking the whole table to see if there
are other errors.

Cheers,

Carl Fogel

I figured out a quick way to check the rest of the table and saw now
other large errors.

A nice reply from DT Swiss USA thanked me and explained:

"We actually quit making our ti spokes 3 - 4 years ago now, and those
tension readings would not be accurate on other brands of ti spokes
since ours were a mix of ti and other materials to reduce stretch."

So things just get more and more complicated.

CF

Quoted message said:

Quoted message said:

On Mon, 8 Jan 2007 05:14:22 +0000 (UTC), [email hidden]

(Luns Tee) said:

In article <[email hidden]>,
<[email hidden]> wrote:
>Given the same dimensions for the short spoke span in the gauge, the
>different deflections must reflect how stiff or how elastic the spoke
>material is.

The difference in bending stiffness does affect the measurement,
but there are oddities to DT's calibration charts that are fishy to me.

The 2.0mm titanium spoke is half as stiff in bending as a steel
spoke. It should deflect more than steel, and for a given range of
tensions, the range of deflections should be greater than for steel.

DT's charts do indeed indicate a greater range of deflections
for titanium, however note that the difference between titanium and
steel varies from the the titanium spoke having more deflection at low
tensions (smaller reading) as it should, to the steel spoke having
less deflection at high tensions, which would make no sense.

My best guess as to what's going on is that there's a constant
offset being introduced into the measurement, with the steel spoke's
measured deflection being a fixed amount more than what it should be,
and/or the titanium's being less. For larger displacements (low
tension), these are overwhelmed by the actual displacement
differences, but at high tensions, the offsets are overwhelming the
actual difference.

The steel spoke displacement is at most only 0.02mm greater than
the titanium in their tables. This is well within the range of offsets
observable from a bend/kink in the spoke, and is the offset that only
Jobst's design can zero for.

Other weirdness in the chart: compare the columns for
1.8/1.6 and 2.0/1.8 spokes. Comparing the straight gauge 1.8 to the
2.0/1.8 spoke, or the 1.8/1.5 with 2.0/1.5 spokes should give you some
feel for the tolerances in these measurement values.

>Possibly the companies are applying complicated physics
>equations, but I think that it's likely just measuring known loads.

I won't speak for what DT does, but the spreadsheet that Park
provides for their gauge does have equations embedded in it (credited to
C.S. Howat). The equation used has no basis in physics as far as I can
tell, but is based on observing that the measured curves, when
plotted on a semi-log scale, are curved somewhat like a parabola. This
may work reasonably well over some range of measurements, but wouldn't
extrapolate well, nor is it able to predict a-priori what the curve
for one type of spoke should be based on measurements for another.

This isn't to single out Park though. The calibration sheets
that FSA provides also give curves that are based on measurements. As
with Titanium vs. Stainless in DT's charts, the curves on FSA's charts
sometimes cross too, when I don't think they should. This is the
result of the curves being, I think, a polynomial fit of the measured
data. While it's a convenient fit to perform in Excel, and produces
smooth pretty curves, it doesn't reflect what the data should be doing.

-Luns

Dear Luns,

Aha! I went back to the DT pdf and proofread it!

The chart at the bottom has not one but two typos that goof up most of
the far right titantium 2 mm column.

I compared that weird column with the data in the individual section
for the 2mm titanium, which appears earlier in the pdf.

At 700n tension, the copy at the bottom skips a value. There should be
a 1.68 mm between 1.60 and 1.75, where I commented "big jump".

At 2000n tension, DT goofed again on the big chart at the bottom. It
should read 2.76 mm, not the 2.86 mm that duplicates the next entry.

I emailed DT Swiss USA about the typos. Anyone can make a tpyo, but
not everyone will fix a typo, so it will be interesting to see what
happens. I wonder if the tool comes with a plastic chart that
duplicates the errors, or if only the big chart at the bottom of the
pdf has the mistakes?

I may not be able to resist checking the whole table to see if there
are other errors.

Cheers,

Carl Fogel

I figured out a quick way to check the rest of the table and saw now
other large errors.

A nice reply from DT Swiss USA thanked me and explained:

"We actually quit making our ti spokes 3 - 4 years ago now, and those
tension readings would not be accurate on other brands of ti spokes
since ours were a mix of ti and other materials to reduce stretch."

So things just get more and more complicated.

CF

Oops--"saw no other large errors"

CR

Quoted message said:

Pat Lamb said:

Quoted message said:

Aaargh!

By exhaustive testing, I've just determined that if you drop a Park
tension gauge onto a concrete floor from a height of six feet, the
gauge will begin to read ~250 pounds of tension instead of the ~190
pounds that it used to read when applied to a spoke hanging from the
ceiling and supporting ~190 pounds of weights.

Was this truly exhaustive testing? Have you been able to replicate the
results?

Well, only twelve data points, but I was exhausted after filling,
hauling, and pouring eleven 10 kg buckets of water into a trash barrel
after first building and setting up a 138-lb platform:

I meant, did you replicate the drop test results? 🙂

On Tue, 9 Jan 2007 09:16:47 -0800, "Bill Sornson" <[email hidden]>

Quoted message said:

Quoted message said:

John Forrest Tomlinson writes:

So...the rule is that off-topic stuff is OK as long as /you/ (the cool
people) agree with and/or like it?

Good thing I plonked JFT long ago; must have missed lots of gay pron! LOL

You have a very juvenille sense of humour.
--
JT
****************************
Remove "remove" to reply
Visit http://www.jt10000.com
****************************

Mark Hickey said:

John Forrest Tomlinson said:

Even scarier than voting for him two years ago is that there are
*still* people who support him. Even in this newsgroup.

Scarier than that are the people who honestly can't imagine any one
individual doing what the majority did or does. Waaaay out of touch.

Speaking for myself, I could certainly imagine that behavior. After
all, Mark, I regularly read what you write, and I regularly talk to at
least three others whose ideas are depressingly similar. My hopes for
your rationality have long since expired.

So it's not a failure to imagine. It's just dismay that such dedicated
self-delusion persists.

Quoted message said:

Doomed to go through life not understanding what's going on all around
them. Sadder yet, they like it that way, and prefer to stay ignorant.

Who are you talking about now? The 80% of the American public that now
disapproves the war? Or your final, faithful 20%?

As I've said before: You remind me of my dear, deluded mother-in-law.
She maintained Nixon was innocent until the day before he resigned.
(Hey, _someone_ has to be the last person to come to their senses!)

Quoted message said:

But really I just wanted to tell the troll patrol, "no cigar"... go
ahead and post off-topic and get circular agreement going. I'm
killfiling the thread as soon as I hit the send button.

Of _course_ you are! Any other behavior would carry too much risk of
your admitting you were obviously, blatantly, hard-headedly wrong!

- Frank Krygowski