The two recent posts on this subject, the first by Ben Gottfried and
the second by Jeffrey Woods, bring up a subject that I devoted a huge
amount of time to in the 1980s.  Both of these rebuilders stress the
very valid point that changing the valve seats to round replacements
may be very problematic.

In my research, it soon became apparent that the little understood
"cross valve" possessed some unique qualities that make it perform
in a superb manner.  Yet, it has a serious flaw.  The flaw is not the
valve design, but rather, the fact that the leather facing of the valve
may become imprinted with the cross design.  That by itself is not the
problem.

The problem is that the valve body may rotate slightly during piano
moving.  It might also rotate all by itself during normal use.  When
rotation occurs, it leaks horribly because of the imprinted cross
pattern no longer lines up with the cross of the valve plate.  In that
condition, the valve(s) will not seat properly.  This is a devastating
condition for the Duo-Art system with its non-compensating knife valve
expression -- or almost any other player action, for that matter.

Once the cross imprint on the valve facing no longer lines up with
the cross in the valve seat, it is almost impossible to fix.  That
is why parts makers advocated changing the valve seats to the later
Duo-Art style of round openings.  The trouble was those parts makers
had not noticed the difference in the openings of the top and bottom
valve seats.  As stated earlier, the top opening is 1/2 inch, but the
bottom is smaller at 3/8 inch.  After numerous failures, the parts
makers discovered their error and offered the smaller bottom seat.

Using only 1/2-round valves for both seats meant that it took,
on average, about 6 inches of vacuum to make the valve operate.
Previously, with cross valves, the same stack would lift the valve
at pressures under one inch!  While the piano would not play at one
inch vacuum, the valves _did_ operate at that level.  At one inch
vacuum, the pneumatic is simply not powerful enough to send the
hammer to the string.

This brings up a very important design issue.  That is, under low
vacuum conditions, the valves must not be the first component to fail
-- it must be the striker pneumatic strength that fails first as the
vacuum is reduced.

But all is not bad with the properly sized round valve replacements.
Realizing that many factors are responsible for proper valve performance,
I analyzed the cross valve system converted to round valve plates.
The conversion can be successful if the size of the valve body (and its
mass) is reduced as well.  There is no easy way to do that and retain
the rest of the components of the old cross valve.

I did offer a system of using solid sponge valve bodies that would fit
on the metal stem.  Those systems would work flawlessly if the bottom
seat was 3/8 inches, but the attitude towards sponge valves had been
forever tainted by the Aeolian players of the 1970s.  Some sponge
will last a long time, but the sponge used by Aeolian does not.  The
sponge I used is still supple after over 32 years.

It is my theory that the cross valve reduces the laminar, annular flow
that is found in all round valve seats.  In round valves, the force
necessary to lift the valves slightly increases as the valve first
leaves its seat due to annular flow effects across the full surface
area of the valve, not just open area of the seat.  By reducing that
type of flow as in a cross valve, the lifting force necessary to bring
the valve to full travel, is reduced.  That is why the cross valve can
operate at such low vacuum.

Bob Taylor
Missouri