[CAUT] pre-stretching new string Wet vs Dry Soundboard

[Fred Sturm]

On 6/10/07 7:21 AM, "Don" <pianotuna at accesscomm.ca> wrote:

> Dean Reyburn's first efforts to generate tunings relied on measurements of
length and
> diameter of strings. The generated tunings were then uploaded via a midi
> cable to a SATII.  They didn't always “fit” the piano.

    Were the differences between calculated and real life tunings due to
differences in humidity (equilibrium moisture content) and its effect on the
board? I guess that's one speculation. I don't know if it's true. Seems like
there are a plethora of factors that could and would contribute.
    The RCT doesn't use calculated figures from length and diameter, but
rather empirically gathered figures: partial ladders of the As, read
directly. Then each A is set according to the "pre-sets" or the "Custom EQ"
(based on the partial ladders) and notes between are entered according to a
mathematical curve. There are a couple additional algorithms for the tenor).
There are often occasions where a partial or more in those ladders appears
somewhere it shouldn't be according to an even curve (predicted value). I
thought _that_ was what Dean called para-inharmonicity.
> 
> It is also why “stored tunings” don't work well in areas where humidity
> changes by more than a few percent.

    I've read statements to that effect. I'm not sure it's true. The
assertion is that real life inharmonicity readings (on which the FAC and RCT
calculations are based, using slightly different means) vary with humidity.
I have looked for a pattern and haven't found one. Have you? Has anyone? If
so, what is the direction and size of the change? From, say 20% RH to 60%
RH, does the A number increase or decrease? By how much on average? (On RCT,
compare the figure in the tuning spreadsheet for G#4 to get a rough
equivalent of the A number difference in FAC for SAT).
    I've certainly seen variance in readings, but no pattern with respect to
RH. The largest differences I've seen are before and after tuning (same
day), occasionally a change of 2 cents and more for that A number (or G#4 on
RCT), always downward after tuning. I speculate it is due to output of the
board being influenced by the vibrational modes of the strings: more
synchronicity, leading to more coupling of partials. Just speculation, and
if anyone else has a different explanation, I'd like to hear it.
    In my own experience, stored tunings work just fine in an environment
varying from under 10% to over 70% RH. Never had a complaint, never noticed
a problem. Has anyone else? In which case, what was the problem? More
precisely, in what direction was the problem? Dry to wet, which "wants
narrower octaves" and which "wants wider?" (If you stored the tuning when it
was dry, did it produce octaves that were too narrow or too wide when wet,
and vice versa?)
> 
> I believe that some SAT users have documented inharmonicity tends to lower
> over time when a piano is kept “at pitch” as the strings conform to the
> bends in their length at the capo bar and the bridge pins.
    Somewhat similar to what I describe above. I'm not sure what the
mechanism would be from "conforming to their bends." Do you read higher
inharmonicity on a newly installed string which has had no "massaging"
compared to one that has? I can't say that I have measured that, myself.
Regards,
Fred Sturm
University of New Mexico