I often wonder if we're preaching to the choir when we discuss these
things in this forum.

My friend Stephen Kent Goodman made a valid point (though possibly
expressed negatively); I almost hear his frustration through the
MMDigest.  I must admit that I'm ashamed today of the mess I made when
I first tried to "restore" a player piano.  Yes, it played after we
were finished with it, but we left so much un-done and that which we
did was glued with carpenter's glue so that if anyone would think of
attempting to do it correctly they would face a situation similar to
what he described above, and for an old Autopiano with chipped yellow
ivory, a mahogany case with missing/peeling veneer, the cost far
outweighs the value of the restored piano.

Art Reblitz published his work on restoring player pianos just
shortly after that.  I learned not only the importance but also of
the existence of hide glue.  I read about techniques and methods --
differences between systems and how to restore valves, dipping pouches,
and too many other items to mention.  All this was before the internet
and MMD.  I restored a couple more players, this time not doing damage
but building skills and learning, following procedure and noting
everything.  At that time I was sketching everything and how it was
connected on a pad of paper.

After working on six regular players I attempted to restore my Ampico.
The first time I put it back together and turned it on there was
nothing.  I had to learn the hard way about selecting leather for
valves.  (How do you describe in words what makes for good leather?)

The hard part was to decide what was the cause of the problem.  Once
I knew what caused the problem the solution was easy (but here I had
rebuilt the entire mechanism, it all seemed to work alone, just when
I put it all together there was a problem).  After learning thusly,
I restored my Duo-Art, expecting fabulous results immediately.  Not
the case -- there were new obstacles to overcome -- but after attacking
each one individually the results are spectacular.

In short, I would add one suggestion to Stephen Kent Goodman's work.
When you are proficient enough to work on a reproducing piano or other
"important instrument", tackle it one bite at a time.  Restore each part
until it's _perfect_ and go through each individual valve until it is
completely tight and shows zero leakage, otherwise you won't know where
the problem lies.  Test everything before moving to the next stage, and
bench test the whole system together before installing it into the
instrument.

Eli Shahar
Tel Aviv, Israel