Here in New England it is not uncommon for later (viz. most) Seeburg
instruments to play weakly or quit altogether during the winter months.
Cold air cannot hold much moisture, and this is the same air that
circulates into buildings and is heated, causing wooden parts to dry
out and shrink.  The return of normal humidity in spring usually brings
affected instruments back to life.  Where two pieces of wood are glued
together with their grain at right angles to each other, some compres-
sion will take place when exposed to extreme summer dampness (as in a
cooler basement where humid air is forced to release more moisture than
it can hold.)  Wooden parts so affected will become compressed and
never fully return to their original dimensions, thus compounding the
problem.  Wood is by nature extremely stable lengthwise (which is what
makes it a better choice than steel rods in uncompensated clock pendu-
lums). Expansion and contraction takes place exclusively across the
grain.

As this relates to most Seeburg stacks, the pneumatics are originally
glued to the pouch boards with their grain at right angles to the pouch
board.  This prevents the outside surface of the pouch board from
shrinking while the inside surface is free to shrink, curling the board
slightly, just enough so that the compressed and no longer resilient
gasket between the pouch and valve boards allows too much air to flow
between it and the valve board -- air that enters via valves open to
atmosphere (as they are, except when notes are playing), and then flows
through the gasketed joint into the channels leading to each pneumatic.
Therefore, because of the pouch board's curling, there is little if any
air entering outside the row of holes in the non-resilient gasket, and
a great deal of air rushing by between the holes and the chest inter-
ior.  Other packed joints throughout the system will also be affected
by dryness, but not to the extent caused by curling of the pouch board,
however imperceptible this curling may be to the naked eye.  I must add
that the front board's thickness is another important consideration
because of the greater than usual length of the steel screws passing
through it.

The simplest way of treating the problems just described is to re-
gasket and tighten up the entire stack.  But there is a rebuilding
technique that can largely prevent these difficulties.  Before gluing
the pneumatics back on (using hot hide glue of course), make a shallow
groove on the outside of the pouch board.  This should be positioned so
that the air channels leading to the pneumatics are halfway between the
edge of the board and the groove.  Use glue sparingly enough that any
excess will flow into the groove but not beyond.  This will allow equal
expansion and contraction of the pouch board on both sides.  Ideally,
in dry weather the pneumatics should be removed and the chests given at
least several weeks without them, to adjust as best they can before
rebuilding.

Another situation that causes poor performance (and not just in winter)
involves the seating of the stack valves.  Original or replacement
valve leather may have had a corrosive effect on the brass valve seats,
resulting in the appearance of a green waxy substance (verdigris) de-
posited unevenly on the brass where it contacts the leather.  Air
rushes between the lumps and bumps which unseat the valves to varying
degrees.  For a quick fix where the leather facings are fairly new, the
valves can twisted in both directions without removing the upper valve
seats by applying pressure with a miniature hole saw or something
similar.  If this method causes old facings to disintegrate, then it's
time for re-leathering anyway.  If you want to prevent the problem from
recurring, take samples of valve leather, clamp them tightly to a sheet
of bright brass or copper, and leave them in a damp place for about six
months. You will then know what effect each sample has on the metal.

Jeffrey Wood