[ John Hutchinson wrote in 101123 MMDigest:

> I am having a go at making my own singing bird.  I am using some
> drawings to guide me along but have a question about a certain part.

Wow!  That is a complex task.  I have been working over a decade
learning how to make and repair singing birds.  The best results were
after I acquired some rough movements from Siegfried & Joerg Wendel's
Mechanical Musicboxes Manufactory (MMM) music shop in Ruedesheim,
Germany.

The literature on the subject leaves out a lot of detail, probably
to protect the French, Swiss and German makers of singing birds who
were really competitive until the 1970s.  More recently most of the
manufacture has been merged into either MMM (Germany) or Reuge
(Switzerland).

The few books I have are

(a) "Le Monde Des Automates", by Chapuis and Gellis.  (The follow-up by
Chapuis and Droz is not technical.)  This is a rare book, in French,
with patent-like line drawings.  Some of the drawings are of protected
"intellectual property" and the omit some critical details and trade
secrets.

(b) "Repairing Singing Birds", Geoffrey T. Mason.  No drawings, just
photographs of rare 18th and 19th century movements.  The reader is
expected to be familiar and have examples to work with.   Some chapters
at the end outline processes for covering bellows and making valves.

(c) Auction Catalogs by Christian Bailly and others.  "The Flights of
Fancy", a coffee table catalog, does get into some limited technical
detail, but it's mostly to determine provenance and completeness of
the items to be auctioned.

Most other books quote Chapuis' "Le Monde Des Automates".

After WW2 most manufacture was re-tooled to make use of plastic
injection molding machines as part of a western stimulus to keep the
people working and the costs down.  This plastic is now yellowed and
brittle.  Bellows were made from vulcanized latex and are impossible
to repair.  Even when the new prophylactic shaped parts are available,
it is hard to place them without breaking the frame.

> There is a rod which gets tapped to set off the whistle.  It penetrates
> a small pressure box and moves a valve.  What stops air moving out the
> hole where the rod goes in?  Is it just a good fit?  Is the movement so
> fast that it is neither here nor there?  Or does it use a floppy leather
> seal to allow movement but not penetration of the rod into the chamber?

Usually this is a triangular plate of stamped metal or injection molded
plastic.  Two pins on the short side of the triangle hold it loosely in
place.  A small leaf spring (which can be made from watch spring) holds
the valve closed.

There can be no leaks in the bellows -- the smallest leak will deflate
the system.  The valve must be as flat as possible. I usually polish
this flap valve on a sheet of typing paper.  The Mason book has the
best advice on covering the bellows.

I do not agree with the use of rubber cement.  A lot of the repair work
I have dealt with has been done with this sticky mess.  Only hot glue
is good enough here.  The rubber cement can be used to help make the
leather air tight; I find this makes the stiffeners, when glued to the
outside leather, to pop off.

Zephyr skin or the red "Morton" leather is the only material that
works.  Everything else I have used leaks.  Stiffeners must be used.
Pneumatic cloth should not be used on pressure systems.

Valves are a nightmare.  These are usually made from sheets of paper
(paper from and old bank check works best).  Post-WW2 valves are
usually a brass cage holding a punching of blue spring steel.  These
are riveted and prone to rust.  No leather is ever used on the valves
except as a retainer.

Good luck on your endeavor.  There can not be too many singing birds
in the world.

Julie Porter