Regarding Adam Aceto's questions about Ampico A regulation [121011
MMDigest]:

> Because of the split stack, most instructions say to test Brilliant
> Medium - Subdue pressures on the bass, then repeat and test the treble.
> All say that "results should be the same at both bass and treble,"
> however I have not found one article with instructions on what to do
> if they are _not_ the same.  I assume either the bass or treble spring
> pneumatic needs to be adjusted, however, I'm not sure whether it's up
> or down.  Perhaps, too, it's the collar on the spring pneumatic?

First, if the bass and treble readings are not the same in the
split stack, I would strongly suggest one should check for leaks.
Perhaps one or more bad unit valves?  Often the unit valves are
tightened too much and they will develop a crack on the sides of the
laminated edges and begin leaking.  This is why I strongly suggest
that all model A Ampico unit valve "springs" (clamps) be replaced
with model B unit valve springs that apply pressure, but allow for
expansion and contraction of the wood.  In addition, unit valve
seats can loose adhesion and leak.

Also check the ends of the stack where a long screw rod holds the
shelves together.  These are notorious for leaks, especially where the
round black rubber washer meets the head of the long screw.  I would
not suggest making any adjustments in the spring pneumatics until all
possibilities of leaks are ruled out.  Even one valve that is not
seating properly could cause a big difference.

> The 1919-20 manual lists a small "set screw" for adjusting the
> crescendo speed.  However this "set screw" seems to have disappeared
> on later Ampico A.  There are, of course, the crescendo springs which
> adjust the first intensity but if these are for the first intensity,
> and there is no set screw anymore, how does one adjust the speed of
> slow and fast crescendo to match the test roll protocol?

In answer to his second question regarding the speed of the crescendo,
Adam is using the 1919-20 service manual but his piano is a late 1928
model A, which would indicate that he has the non-adjustable crescendo
mechanisms of the later model.

Bruce Clark