In the MMD for 220924, Robbie inquired about a lock and cancel system
using a simple pouch for one of the valves. I don't know if this
configuration was ever used in a production device, but it certainly
looks like it should work.

If valve 4 were located just below and slightly to the right of valve 1,
the top board of valve 4 could be incorporated into the pouch board of
valve 1. The connections between them would simply be drilled passages,
yielding a simple compact construction.

There seem to be two small potential disadvantages with this system:
First, when the lock valve is OFF there are two bleeds in parallel.
To trigger the lock valve, the opening in the roll and tracker bar must
be large enough to overcome these bleeds, which might make the system
a bit less forgiving of imperfect roll tracking.

Second, whenever valve 1 is ON, air flows through its bleed. This is
most likely a completely negligible load on the vacuum supply.

I've never been a fan of the "interconnected unit valves" method of
building a lock and cancel mechanism. I suppose it makes sense if
you have a factory churning out thousands of unit valves. But for
a hobbyist who just needs a few lock and cancel functions, it seems
to require more time and trouble than necessary.

(For register control in my current organ project, I have designed
a shameless copy of the Ampico A expression mechanisms's lock and
cancel valves.)

John Wolbach