S&S D Duplex

[Ron Nossaman]

>If the tone only lasts 3 to 4 seconds, what kind of piano is this that all 
>other features such as strings, hammers, board, action and tuning are 
>correct? Sounds like something is broke to me.
>
>Terry Farrell

Well now, I'm glad you asked. I happen to have a few highly unsolicited, 
unscientific, and subjective unauthorized measurements of sustain times 
from revered random duplex scale patent products I collected as I 
encountered them early this year. Times in seconds were noted from a firm 
blow, to when the perceived sound  became indistinguishable from background 
noise, as nearly as I could tell.

Dampers down, by the way.

                 C-8     C-7     C-6     C-5
S&S L : 1.5     5       9.5     12
S&S B : 1       3       5       10
S&S D : 2.5     5       8       13

This particular D was, in my experience, extraordinary in it's last octave
sustain. The 1.5 to 2 second range being statistically more likely. This 
obviously isn't much of a sampling of the total number of beneficiaries of 
THE tuned duplex patent, but you (all) can take your own samplings and fill 
out the cross section as you wish. I'd be curious to know the highs, lows, 
and averages if you do. Also in my experience, when strings, hammers, 
board, action, and tuning are anywhere near "correct", these sustain 
figures tend to be considerably higher with no hint whatsoever of a tuned 
duplex. Build a decent soundboard, put a decent compatible string scale on 
it, give it some back scale length so it can move, and you can get good 
tone and sustain without duplex scaling.

Approximate projected sustain times from Dan's figures after tuning 
duplexes are:

C-8             C-7             C-6             C-5
2-12.5  6-25            10-47.5 20-67.5

If the very worst examples of sustain I have here could be improved by 100% 
by duplex tuning, they could be brought into the range of the best examples 
I have here of presumably un-tuned duplexes. After all, if any of these 
were already tuned, no improvements would be expected. If the best examples 
I have here would be improved in sustain by 500%, they would be brought 
into the range of the fountain of Lourdes. So let's try the other way. 
Let's improve the worst by 500%, and the best by 100%, presuming that the 
low performance examples are being held back from their full performance 
potential by poorly tuned tuned duplexes - accounting for their not 
altogether impressive showing. We then get:

C-8     C-7     C-6     C-5
5-5     10-15   19-25   24-50

 From my random sampling, I can apparently be assured of getting at least 
5, 10, 19, and 24 seconds from C-8, 7, 6, and 5 respectively just by tuning 
the duplexes. That's a pretty remarkable claim. Can I presume there is a 
money back guarantee on the purchase of the tools sold specifically for 
this job, and compensation for the time wasted performing the duplex tuning 
correction by the official procedure, should these minimum sustain figures 
not prove to be obtainable by these methods? Still, if any improvement at 
all is detectable, it would be worth doing, since it's already designed in 
and we're stuck with it.

Ron N