[Ghosts of impedance past, and yet to come]

[Ron Nossaman]

>Ron:
>
>This sounds a lot like a phenomenon described in the 1978[?] Scientific
>American piano article (grand piano harp on the cover). The crux of it was
>that the horizontal components of neighboring strings communicate with one
>another directly across the bridge, effectively becoming nonlinearly coupled
>oscillators. This is a situation in which (sometimes chaotic) mode-locking is
>common, with the cycling of one of the oscillators being captured (steered,
>entrained) by the other. It has been studied in some detail in biological
>(small groups of neurons) and optical (laser mode-locking) systems, among
>others. Not simple, not necessarily reproducible in detail, and hard to
>predict.
>
>Marc Damashek
>Hampstead, MD

Hi Mark. Yes, I read the article, and this is in the same ball park, but I
thought it curious that some pianos exhibit a fairly dramatic pitch drop in
the unison as the strings are tuned in, and some don't. Since the coupling
between strings takes place through a medium (soundboard assembly) that
exhibits fairly broad differences in tone production as it's impedance is
changed, it seemed reasonable to me that there was a correlation. I started
with one of my rebuilds because the soundboards I'm installing don't have
the high tenor and treble tone problems characteristic to a lot of pianos,
especially those with compression, or panel crowned boards. The piano in my
shop now no longer has an active front duplex, so the soundboard impedance
and back scale seemed to me to be the most likely players in this case.
Since taping off the back scale made the phenomenon disappear altogether,
I'm momentarily encouraged that doing so took the feedback and cross
coupling effect from the back scale out of the system and what was left was
pretty pure and primarily between the speaking segments and the soundboard
assembly. If the same thing happens in a piano I know to have low impedance
problems, I'm shot out of the water. If I mute all the non speaking
segments and still get a pitch drop with unison tuning, I may have a case
and I keep collecting data.

Incidentally, clock makers have known about coupled oscillators for
centuries. A wall full of pendulum regulated clocks of the same model will
synchronize to the point that they will tick in unison. Rather an
impressive effect, I bet, especially at noon in Switzerland.

Ron N