While this interesting discussion continues, it seems to me that
adopting a standard terminology of identifying the stack split point
would help clarify the issue.   The root of this discussion is where
the stack split occurs, not how many notes are on each side of the
split.  In the case of the Duo-Art, there are in the USA two stack
sizes: 80 notes and 88 notes.  This split always occurs in the same
spot: note D# is on the bass side, note E is on the treble side.  The
corresponding numbers of those notes are 43 and 44, respectively.

The Duo-Art Test Roll #3 verifies this configuration.  In setting the
theme minimum intensity, several arpeggios are played in different
keys.  One can observe the "snake bite" for each note played.  When
note 43, D# is played, the snake bite is on the bass.  When note 44,
E is played, the snake bite is on the treble side.

If the stack does not conform to this standard division, the errant
note will sound softer or not play at all during this test.

Unfortunately, the common Duo Art tubing schematic which is in wide
spread use, details the stack division in the wrong place.  I noticed
that when I rebuilt my first Duo Art thirty years ago.  In thirty years
of rebuilding American made grand pianos, I have never found a Duo-Art
stack that didn't have the division at 43/44, notes D#/E.  I have never
heard from any American rebuilder that they found any other configuration.

The test roll is coded that way, the pianos are built that way, the
split, note 43/44, appears to be the standard.

Bob Taylor