John G. DeMajo had an interesting writeup on his experience in making
his own pipe chest [000514 MMDigest].  I would like to add a few
comments to the discussion.

I learned the ins and outs of electro-pneumatic chests while working
at American Organ Supply Co. in Milwaukee for many years.  (I'm also
the answer to the really trivial question, "Who made the high pressure
Posthorn chest in the National Cathedral?")

> Another [modification] is the elimination of Wurlitzer's cumbersome
> bottom board gasket arrangement wherein wind channels are passed
> through detaching wood components.

Viewing the drawing, the only problem I can anticipate is difficulty
when it comes to re-leathering.  Well designed chests are made to be
cleaned up quickly with machines.  Typically, pneumatics like these are
glued on with a gasket which should be sanded off during re-leathering.
Leaking gaskets can be virtually eliminated by using leather gaskets
and compression springs under the screws.  When screwing boards with
action channels together, springs are essential, and screws should be
8" or less on center.

> Finally, a radical departure was a decision to eliminate primary
> pneumatics from the design.

To primary or not to primary...  On very large pneumatics, a primary 
is necessary because a magnet alone cannot exhaust enough air pressure
fast enough (except on model railroad switch machines).  When several
notes are played with one magnet, as in a pitman chest, again primaries
are essential.  On smaller pipes, the primary actually slows the action
a little.  Rule of thumb is that a 1/2" hole is at the break-even
point, but that varies with pressure and valve travel.

The advantage of a primary on small holes is that the valve cycles from
completely closed to completely open, and back to closed.  Without a
primary, the valve can partially open, and then close.  This can cause
defective pipe speech when a note is repeated very rapidly.  A reed
pipe can hum or a flue pipe can sputter.

The key to better actions, with or without primaries, is to limit the
travel of each component to the minimum necessary.  Primary valves
usually operate best when they move only about 1/16"-3/32".  Valves
over pipe holes deliver maximum air flow when they move about 1/4 the
diameter of the hole.  Individual pipes may not need maximum air flow,
or the holes may be too small for good speech.

> After studying many catalog cuts and manufacturer spec sheets,
> we decided to employ Reisner chest magnets (OSI part # 5501.32)
> because they are well suited for higher pressures, feature an
> adjustable orifice, and operate with 150 ohm coil resistance at
> 12 volts DC.

The #5501.12 has a 90 ohm coil which gives a little more power that
may be useful at higher pressures.

> There were several sizes of pneumatics used by Wurlitzer.  A complete
> listing of these dimensions are available on the web site of the
> S.M.G.C. Chapter of A.T.O.S.  The web address appears at the end
> of this article."

I couldn't find information relating pipe holes and pneumatics --
just specs for primaries.  Our stock for the chest side, where the
pneumatics are glued, measured 1-1/4" x 8".  We would suggest that
if the reader chooses to place the magnets on the same surface as the
pneumatics, a 1-1/2" x 10" x 8' poplar board would be appropriate for
this member.

To save you a little cash on the next chest, can I suggest that you
can use much thinner sides?  The 1/2' exhaust hole presumes a primary
exhausting through a similar size hole.  When pneumatics are exhausted
through a magnet only, the holes are between 3/16" and 5/16" depending
on how long the run is.  4/4 poplar planed to 13/16"-7/8" would be very
adequate, and you can use a 1/4" hole.

> Looking at the construction drawing available on the web site, you
> will note that the air channels exit to the outside of the chest at
> one side of the bottom corner.  In placing the magnets on the side
> board, you will instead end up with drill hole openings at the top
> of the chest side member.  After the drilling is completed, these
> holes will need to be plugged.  We addressed this problem by
> inserting 1/2" wood dowels, cut to 5/8" lengths, into each hole and
> gluing them into place to seal against air leaks."

Commercially, we usually seal holes by gluing a strip of wood the
length of the whole piece.  Whenever possible, two boards that need
to be sealed are glued to opposite sides of a single piece, which is
then sawed apart after gluing.

> Ref 1.  Adjustment of primary-less Wurlitzer chests", by Doug Powers
> available through the Knowledge Base of the S.M.G.C. Chapter of
> A.T.O.S web site at http://atos.stirlingprop.com

This method seems rather tedious, since it requires going into the
chest for each adjustment.  If each note is made with an individual
bumper attached to a machine screw threaded through the side of the
chest, the adjustment can be made from the outside of the chest, and
there is no risk of breaking parts.

Jim Crank wrote about advantages of Kilgen's design: "There is also a
guide rail between the pouch and the valve under the pipe hole.  When
the bottom board of the chest is removed, the entire action comes with
it, guide rail and valves.  It is very easy to measure the height of
all the valve faces when the bottom board is down, making adjustment a
snap."

There is a trick to getting everything aligned with this kind of
action.  The pipe hole and the valve hole are drilled from opposite
sides of the toeboard.  Both the toe board and the bottom board are
located on the frame with dowels.  The valve holes are not drilled until
the wires are attached to the pouches.  Then a heavy ink or paint
pigment is dabbed on the ends of the wires and the top is lowered on
long guide dowels into place to mark the valve hole centers.  Valve
holes are generally larger than the pipe holes.

Bumpers on machine screws through the center of the pouch hole can be
externally adjusted to limit valve travel from the bottom of the chest.
This style is good when several ranks are on the same chest frame,
since no side access is necessary or possible.

> The Kilgen chest is actually faster than the Wurlitzer chest with
> primary valves.  This has been tested by various organ builders
> many times in the recent past.  It will actually repeat faster than
> any other chest.

Again, on smaller pipes holes, approximately 1/2" or less, the primary
slows the action down, but cycles it completely.  Wurlitzer style
pneumatics, rectangular in shape, let you put pipes closer together.
Kilgen style lets you access everything from the bottom, and simplifies
manufacture -- just drill a hole and punch a pouch.  A pouch board style
chest -- valve attached to pouch, not a wire -- simplifies it even more.
Whatever the style, well done with properly sized units and limited
travel of the components, it will work very well.

Something not previously mentioned is that action channels should be
sealed: shellac is traditional and works as well as anything.  Channels
can be reamed with a smooth rod chucked in a hand drill after sealing.
This is not nearly as critical as vacuum actions on player pianos, but
it does make a noticeable difference.  It's also a good way to get rid
of old shellac that is not longer good for finishing.

If you read this far:

a. You really should spend less time on the Internet.

b. You have an incurable obsession that is probably not dangerous.

c. You might actually be thinking seriously of making your own
   pipe chest.

d. You may want to ask for more technical information.

e. You may be in the market for parts or supplies, in which case
   we would like to hear from you.

John Nolte