"killer octave:

[Ron Nossaman]

>Thanks to all for making this as clear as mud for me!
>
>Rook


The killer is where it is. It doesn't start at any specific note, nor end
and any specific other. On any given piano, it will shift up or down scale
with humidity swings. It will even seem to disappear altogether in some
pianos during high humidity cycles. The root cause is the soundboard being
too low in impedance in that area. The killer isn't in exactly the same
spot from piano to piano, or even in the same piano from month to month
because no two soundboards are exactly alike, even if the string scales are
(nominally) the same, and no individual soundboard is the same at 30%rh as
it is at 60%rh. The beginning of the killer octave corresponds, generally,
to about where the tuned front duplex starts. It is my opinion that this
isn't an accident, but that the duplex was designed to combat the already
existing soundboard problem in this area. First came the basic soundboard
design, then came the tuned duplex. Why would anyone go to the trouble of
inventing the duplex if they weren't trying to shore up what they
considered to be inadequate sound in that section? This is also the area
that gets the most complaints regarding voicing. Consider all the different
sets of hammers from all the different manufacturers, all nicely graduated
in size from one end to the other. Why would so many of them have voicing
problems in the same area of the scale, but some not? Why does extensive
voicing, and even hammer replacement fail to fix the problem? If the
hammers cause the problem, shouldn't the problem go away when the hammers
are replaced? Why does rebuilding the piano, rescaling, renotching bridges,
replacing the action, and shimming and refinishing the soundboard, too
often result in a rebuilt piano with the same kind of tonal problems, in
the same parts of the scale, as it had when it came in the shop? Why does
no amount of tapping strings down on bridges tame the killer octave? Why
does the soundboard go flat in the killer octave first? Why does that
section go out of tune so quickly, and so far... and why almost always
FLAT? That's because it's a soundboard design problem, not a voicing or
tuning problem. The killer octave does exhibit tuning (as opposed to tuning
stability)  problems though. This is the area where the unison pitch drops
most as you tune the second and third strings to the first. Why is that?
Doesn't that strike anyone but me as an odd coincidence? I think that this
is likely to prove to be a soundboard problem too, though it may be as much
related to resonant frequency as to impedance. I've been trying to get some
folks with SATs and RCTs interested enough to help me chase down some data
on this one. I've got some interested people, but no one has had the time
this fall to chase anything down. I'm working on it though, and I'll
eventually publish the results, whether I get anything conclusive or not.

So there are a whole bunch of problems associated with this area. Every one
of them is a result, not a cause. The cause is the soundboard design, and
no matter what you do about the secondary effects, the only kind of fix
that approaches the root of the problem is to install a more efficient
soundboard. I know that's not a field repair, but the point is that there
IS no field repair that's anything but a band-aid to disguise the symptoms.

More mud, right?

Ron N