> John:  I wonder if you would answer a question for me.  I am
> restoring a player piano that has a Simplex action.  I have
> recovered the bellows and repaired the valves but it still pumps
> hard.  (That is to say, I have to pump very fast to get it to play.)
>
> The bottom pumps check out okay.  I connected a vacuum motor to the
> player action on the work bench and then I connected a vacuum gauge
> to the hose going to the last bellows in the stack.  I get the same
> reading there as I get out of the vacuum motor (that is 15 to 45)
> where ever I set the speed control.
>
> Does this indicate that the player action has no leaks?
>
> Thank you for your help.

Hi,  First, let's presume that you connect a vacuum pump to your stack,
and place a vacuum guage in line so that you can monitor the vacuum
level.  It only stands to reason that the vacuum level you measure at
any particular note will be identical to the level you measure at the
pump.

Why?  Because you didn't start the test fairly!  In order to determine
the amount of leakage in the stack, you must first measure the vacuum
level of the pump _before_ it is connected to the stack.  Then, and only
then, can you determine how much vacuum is being lost in the stack.

Let's say that the vacuum level of the pump is set to 15 inches before
it is connected to the stack (as though the stack is 100% airtight).
Then, after it is connected to the stack, the vacuum level drops to
seven inches.  It should be obvious that eight inches are being lost
within the stack.  Let's say that the pump is set to 20 inches before
it is connected to the stack, but the level drops to 15 inches after it
is connected to the stack.  Then, you have five inches of loss within
the stack.

Another way to determine the amount of loss within the stack is to
'feel' (with your hand) how much air is flowing 'out' of the exhaust
port of the vacuum pump.  If the stack is 'air-tight', _no_ air will
be flowing out of the pump.  Also, bear in mind that the exhaust port
on most vacuum pumps is rather large (over two square inches in most
cases).

So, the only way to really 'feel' how much air is passing through the
pump is to reduce the size of the exhaust port.  This is easily done
by putting a piece of duct tape over the exhaust port such that the
exhaust port measures just 1/2" by 1/4".  Doing this will also give
you a clearer picture of the severity of the leakage within the stack.

If the amount of air flowing out of the exhaust port is minimal, then
you know that the leakage is fairly minor.  However, if the air flowing
out of the pump feels like a 'strong breeze', then the problems are
more severe.  If the wind blowing out of the pump is like a hurricane
(in excess of 60 mph), you have major problems.

A properly restored Simplex action requires less than seven inches of
water vacuum to play the piano.  At fifteen inches, the volume of the
piano (music) should be fairly loud.  At 25 inches, the music should be
very, very loud.  At 45 inches, parts in the piano action might start
breaking!

Here's another test for you.  Close off all of the valves as if there
are no notes playing.  Then, by mouth, suck on the vacuum supply hose
that is attached to the stack.  As soon as you start sucking on that
hose, you should 'feel' all of the valves seat.  In a perfect world,
it should feel like sucking on a glass coke bottle.  In other words,
there should be no airflow.  If there is airflow as you suck on the
hose, you have leakage in the stack.

The beauty of sucking on the hose by mouth as opposed to using a vacuum
pump is that you can usually hear where the air is leaking "in".  It
generally makes a 'hissing' sound that is relatively easy to localize
using a stethoscope or a 'listening tube' (a piece of trackerbar tubing
that you stick in your ear).  However, if the leakage is a combination
of 80+ leaky intake valves, or leaky gaskets, or 'internal leakage',
you might not be able to hear anything.

The Simplex valve-pnuematic blocks can be tricky.  One of the problems
I've encountered, that is hard to detect, is air leaking 'around' the
intake valve seat -- through the wood.  Even though the valve might be
seating solidly, the vacuum on the underside of the intake face leaks
out (or, more correctly stated, the atmosphere leaks 'in') through the
hole in the top of the block (the exhaust valve plate).

This happens because the wood around the brass intake valve seat is not
100% airtight.  As the wood ages, and looses its moisture, it becomes
more porous, thus allowing air to pass through.  Also, wood shrinks a
tiny amount as it ages.  So although the valve seat might appear to be
solidly connected to the wood, air can pass around the seat.  Although
the amount of air passing through (or around) each valve seat might be
very small, it becomes quite a leak when multiplied by 80+ devices.

The only way to eliminate what I refer to as 'internal leakage' is to
seal all of the internal wooden surfaces with a good sealer.  I use
Phenoseal, but shellac or lacquer also work well.  They just take more
time to dry, and multiple applications might be needed, depending on
the severity of the leakage.

In your case, since you have already recovered the pneumatics, there
is no way to seal the internal portions of the blocks without undoing
all of the work that has already been done.

This is where the true test of a man (or woman) comes into play.
When you find that everything you have already done has to be taken
apart and done over the right way, you can fairly judge your own
capabilities and limitations.  The faint-of-heart will be satisfied
with less-than-optimum performance.  The courageous will do the right
thing!

In closing, a copy of this email has been sent to the Mechanical Music
Digest (http://mmd.foxtail.com).  Your name has been deleted from the
response, and you were sent a 'Blind cc' so that no one will know who
you are.  I feel that the information I have given you is significant
enough to warrant being placed in the Archives at the MMD, and that is
why this response was also sent to them.

Musically,

John A. Tuttle
Brick, NJ, USA
http://www.player-care.com/

 [ John, thanks very much for patiently replying to requests for help
 [ like this one, and thanks again for sharing your knowledge with MMD
 [ readers.  May your Player Care business prosper!  :-)  -- Robbie