wrong? (long?)

[Richard Moody]

Tim Keenan (M.Sc.) wrote---
>That Nickel is according to a CRC hand book, " having infinite
magnetic susceptibility, >along with Iron and Cobalt--only"

Another technician told me nickel was magnetic.  However we could not
demonstrate it on American nickels, but he had another piece of
nickel to prove it.  I don't know if earlier nickels are nickel
enough to be attracted.  While we are on the subject of wrongs here's
my nickel's worth.

There have been several suggestions that amplitude changes frequency,
esp
in the upper partials.  The greater the the amp. the more sharp the
freq.  because the tension is
increased. In strings of course. I don't think this this applies to
loud speaker tones.
	 I thought amplitude was independent of pitch.
You can see what a state music would be in if loudness affected
pitch. I am willing to grant there may be momentary sub-sensory
phenomena, machines perhaps can detect, but as long as music is
listened to
and enjoyed by human ears, these are the priorities of my concern as
a technician. .
	Of course it makes sense that if you stretch a
string, as in getting ready to pluck, the tension is increased.

 Take two identical strings, pull one to 3mm,
the other to 6mm.  Let go. If the two strings are at the same freq
there should
be no beats.
	So as far as the amplitude affecting the string tension,
 "perhaps it does", but that tension also changes at the
same rate of the frequency.  ; )

	A while back an experiment was performend to observe the partials as
affected by voicing.   Data from sensitive instruments was collected
and sure enough the partials showed a change, the softer hammer
produced the higher partials,  so the conclusion was that yes voicing
does affect higher partials.  However the original hypothesis was
that harder hammers should produce higher partials because they would
have greater amplitude. Hmm.
	 Another interesting variation  was a 10% difference in the cent
reading of the partial produced by the same hammer at the same
volume. This was on a machine that claims to be accurate to .1 cent.
To the statistics 101 challanged that means  give or take .1 cent off
the reading to be within the range of accuracy, right?   OK so the
machine was within its limits (I think) so the data was averaged, and
the  difference was 2.5% or two and a half hundredths.   And if you
realize the value of a cent, 100th (close enuf) of a semitone, you
are comprehending 2.5% of 3%. (the readings involved a
partial in the 3 cent sharp range).
	Is 2.5% difference considered statistically significant? Any
statistitions on the list?  Perhaps the data could be "organized" in
a way that would be more favorable to the experiment.
	So from the data as I read it, I conclude voicing so far has not
been proven to change the frequency of the partials. Considering the
machine can be off by .1 I wonder about the validity of the over all
data being different only by .025 as measured by that machine? The
question I have though, is if the partials do go sharp from more
amplitude, what happens to them with less?  And of course how does
this affect the practice of tuning?

	So I offer another hypothesis, that nothing can change the frequency
of partials of a piano wire, execpt the wire itself. (The fundamental
is considered the first partial)  A partial is caused by a physical
segment of the vibrating wire, this segment does not change, unless
the physical characteristics of the wire are changed.   Whether it is
plucked, bowed or struck, or made to resonate, the partials will not
change at any amplitude.  A second rule might be that you can't
change one partial with out changing them all. To ask a question, can
amplitude change some partials more or less than others?  And the
question of less amplitude.   The energy of a vibrating
piano wire is evansecent.  If the partials sharpen from increased
amplitude, what happens to them with decreasing amplitude?

	What happens I think was described by Helmholtz.  In a piano, the
volume (amplitude) of the partials is determined foremost by the
velocity of the hammer, then by the condition of the  hammer, and
where the string is struck.  The reason a string
sounds bright is because the hammer is hard.  In a hard hammer the
higher partials are excited more than the lower ones.  Imagine a
thumb tack in a hammer. (a lot of us don't have
to).
	In theory, (also Helmholtz i think) the ideal rebound from a string
is when the hammer leaves the string at its highest amplitude. That
would be the apex of the fundamental. The ideal hammer rebounding
from a string at its (ideal)
apex produces in theory an ideal tone of the fundamental comprising
50% of the volume and the partials making up the other half.
 How this could be measured, is beyond me, but it does help in
visualizing the concept.
	Now if the hammer leaves the string before this apex is reached, it
must be rebounding early, because it it too hard. Therefore the apex
of the fundamental was  not reached, but the apexes of rest of the
partials were.  Thus a tone with a greater volume of upper partials.
  Suppose it doesn't rebound until after the apex is reached?  The
string would be using its energy to throw off the hammer, less
energy, less amplitude.  Some would say that upper partials are
additionally dampened by the extra contact with the string.

	Any how that's the theory.  There are suppposed to be movies in slow
motion of hammers hitting strings.  With mpeg shouldn't be too long
before they are on cd or a web site.  I wonder if vibrating piano
wires were looked at and photographed with strobe lights?
Richard Moody




----------
> From: Tim Keenan & Rebecca Counts <keenan.counts@sympatico.ca>
> To: pianotech@byu.edu
> Subject: Re: wrong?
> Date: Sunday, April 13, 1997 9:39 AM
>
snip.....
>but according to the CRC Handbook of
> Chemistry and Physics, 60th edition,  which I just happen to have
here on
> my desk, Nickel is "hard, malleable, ductile, somewhat
ferromagnetic, and
> a fair conductor of heat and electricity [page b-15, column 2]. It
is
> described in the table of magnetic susceptibilities on page E-126
as
> having infinite magnetic susceptibility, along with Iron and
Cobalt--only
> these three elements are so described (ferro). All other elements
and
> inorganic compounds have numeric values listed.
>
> Tim Keenan (M.Sc.)
> Noteworthy Piano Service
> Kitchener, ON