The mystery deepens...

Several folks have offered opinions about the 80-note stack "relic"
I spoke of in MMD 17.03.08, including Robbie, Larry Hollenberg, Dave
Saul, and Bob Taylor (in a private message).  Most have concluded that
it was probably in a very late Duo-Art installation; however, upon
further analysis I am discounting that probability and returning to
an Ampico variant.

Dave Saul rightly suggests that any attached expression components
should be helpful if clearing up the mystery; if only it were that
easy!

The first attached photo is a mosaic of four different areas of the
stack, taken from the rear, top-side and underneath.  Maddeningly,
whatever expression components were originally affixed to the bottom
of the stack are missing, along with all of the unit valves and poppet
wires.  What's left are 14 holes, 7 per side, that would have conveyed
the vacuum from the expression units into the stack.  This looks to me
like the "normal" way "B" style expression components are glued to the
stack in an Ampico design, but there is other evidence that discounts
the Duo-Art theory.

Dave mentions the stack "division point" as supporting the Duo-Art
premise, but I think he mis-read my analysis.  I started with the notion
that the stack split would have followed the Ampico practice of being
between note 44 (middle E) and note 45 (middle F), and worked backwards
towards each end, calculating that the first playing note would have
been #5 (low C#) and #84 (high G#), which nicely meshes with the
Duo-Art's 80 note compass.

But then the Duo-Art "split", which is between note 43 (middle D#) and
note 44 (middle E) becomes problematic.  So, if this stack was intended
for a Duo-Art installation, one of two equally unlikely situations
would have been present: If the split is "accepted" as being between
D# and E, then working toward the ends, the first playing note would
have been low C, (the 40th note below the split) but the installed
first striker pneumatic would have never played, since no Duo-Art roll
is capable of playing that note.

Likewise, the highest playing note would have been high G (40th note
above the split), and any Duo-Art roll having a legitimate perforation
for high G# would be silent on that note since there is no pneumatic or
valve installed to play it.

The other situation would have been to match the stack's 80 playing
notes with the Duo-Art compass, whereupon the split would occur between
E and F, rather than D# and E, which means that middle E, _if themed,_
would not sound at the (treble) themed vacuum level, it would sound at
whatever accompaniment level was then present and the treble theme
snakebite would be ineffective.

I suppose it's possible the editors would have accepted that
infrequently occurring result, but I'd like to think they would not
have allowed either of those compromises.  I therefore come down on the
side of an Ampico installation with a shortened compass of 80 notes,
C# to G# rather than 83 notes, B to A.  This leaves the split in the
correct Ampico position.

The other photos clearly show that there was room to have had the
missing three poppet guides and valve blocks installed in the same
overall stack width of 47-1/4" (Ampico assumed).  If we accept that C#
to G# comprised the 80 notes, that means the actual piano action breaks
would have been (bass to treble) 26-28-17-17.  Does this match any known
scale?

John Grant

 [ http://www.mmdigest.com/Attachments/17/03/11/170311_214654_80-Note%20Stack%201.jpg 
 [ http://www.mmdigest.com/Attachments/17/03/11/170311_214654_80-Note%20Stack%202.jpg 
 [ http://www.mmdigest.com/Attachments/17/03/11/170311_214654_80-Note%20Stack%203.jpg 
 [ http://www.mmdigest.com/Attachments/17/03/11/170311_214654_80-Note%20Stack%204.jpg