Eric Shoemaker writes in 030216 MMDigest:

> Turning the adjustment screws on the spring mounted on these units
> will only change the _time_ between up (loud) and down (soft), and
> not the level of loudness.  This adjustment was placed there to
> determine the rate of slow/fast crescendo/decrescendo.

This is incorrect!  The screws are to adjust the first intensity, and
have nothing to do with the time of up or down.

No one has suggested to check the pump pressure, or the possibility of
leaking pump bellows.

To properly set the expression on an Ampico A system, without the use
of a test roll, one starts by covering the tracker bar with tape.  Set
the lever on repeat, or put tape over the take-up roll port so that the
lever stays in the play position and the motor stays on.  Set the tempo
to 75, and move the modify lever to medium.  The reason for setting the
tempo to 75 is to compensate for the air used by the air motor.  This
will effect the readings.

Next, nudge all three of the intensity buttons, 2,4, and 6, for both
treble and bass so that they lock in the upward position.  Using an
accurate vacuum gauge connected to the stack via a nipple on the
primary chest, test the bass and the treble separately.  Both readings
should be identical and should read 20" of vacuum.  If not, the spring
on the pump (Front assembly, Illustration 9, fig. 7 of 1923 Service
Manual) should be rotated very slightly to increase or decrease the
pump pressure to 20".

This is the most important setting in the Ampico A.  If the pressure
will not go higher it shows there are many leaks that need attention
before proceeding.

If the pressure reads 20", the next step is to move the modify lever
to brilliant and the reading should increase to between 27" and 30",
according to room acoustics.  (Do not be tempted to adjust this above
30".)  The adjustment to attain this reading is the screw in the top of
the amplifier pneumatic (Front assembly, Illustration 9, fig. 8 of 1923
Service Manual).

Once this is set, the first intensity or lowest should be set by
playing many rolls and turning the screws on the crescendo pneumatics
until soft notes are very soft but not dropping out.

In addition, one should carefully check to be certain the Amplifier
control box is working properly.  I have found several with faulty
valves that do not seat properly (Illustration 10).  The valve stem,
fig. 11, will seat much better if the stem is not attached solidly to
the valve button.  A piece of felt glued between the valve stem and
valve button solves this problem and will allow the valve to seat
properly each time it is called for.

One other problem that has been overlooked the use of rubberized cork
by some restorers as gasket material.  While it sounds like a fine
idea, the rubber material mixed in the cork hardens and will not allow
a good seal.  I have had better results with plain cork gasketing or
leather gaskets.

Bruce Clark