Some time ago, when my interest in music boxes was very focused, I had
a real desire to understand what made some music boxes sound as they
do: some are okay, some good, and some are wonderful, with great depth
of tone.  I went through various experiments and exercises, including
a "tuning" study.

As an engineer I figured I could measure, evaluate, quantify, and
characterize various tuning factors relating to the scheme existing
on a really great sounding box, and then apply my now-defined "scheme"
to a lesser sounding box, to the end of upgrading the sound of that
box.

I used an electronic tuner to take the tuning from a really great
sounding box (this was a very nice sounding 18-1/2" Mira).  The tuner
was a strobe type with a contact microphone which gave note pitch to
1/100 of a semitone.  I studied the relationships (discord) between
unisons on each comb, between combs, between octaves, etc., etc.
I graphed, superimposed, looked for patterns and eventually came up
with what I believe was Mermot's "scheme".  I will not elaborate on the
scheme here as it is quite complex and involved as I figured it out.

To evaluate the validity of the "scheme", as to merit, the following
efforts were completed:

a.  The Mira had its tuning slightly "corrected", to the significant
detriment of its sound.

b.  The Mira was tuned a bit closer, overall, to the derived "scheme",
with resultant very rich tone, which was perhaps a bit enhanced over
original.

c.  A couple of Reginas were similarly tuned, with the result, in the
writer's opinion, that they could hardly sound better.

Alan Erb