S.K. Goodman asked, in regard to adding a xylophone to an 88 note pedal
player:

>1) How is "25% vacuum loss" arrived at?

Well, 24-30 notes of a xylophone represent about 30% of the total
number of playing notes in an 88 note player (actually only 82 notes
were usually cut).  On the other hand, the percentage of treble notes
which play may account for over 50% of the notes playing at any one
time. But keeping it simple, and not knowing where you intend to tube
it into the trackerbar (you might also want to octave couple as well),
valve losses will be almost identical to the player stack, so just add
that extra number of valves. (If you octave couple, however, add twelve
more).

>   2) Wasn't the Ceclian a foot pumped piano player of 58 notes that
>      operated both pnuematics that played the piano as well as
>      contained a reed organ?

I've heard of it, but never seen one. Maybe somebody on this page knows
more about them.  On the other hand, the reed organ only requires an
inch or two of vacuum pressure, maybe a little more or less depending
on the reed, so the power required to play reeds is *about* 1/5 to
1/10th that required to play piano notes. By bleeding a very small
percentage of your player vacuum into another reservoir controlling the
reed pressure, you could play the reeds by vacuum, or, conversely, you
could also build pressure using a double pump system the way
orchestrions play their pipes. I'd be interested to know how this was
actually done. power-wise, however, you have no problem.

> 3) Did the 58 note range of that push up compensate for the vacuum
>      loss of having both a piano stack and reed organ, or did it not
>      use a vaccum motor to turn the rolls?

I doubt that the 58 note range compensated too much, since those early
pianos used a pneumatic twice the size of later models, and the valves
used were also twice as big. Everything was lower pressure and bigger.
But I suspect it still used a vacuum motor to drive a roll. Some of
those vacuum motors were humongous, too. I have one that's about 15"
long, and the bellows  look big enough to power a go cart. (And
speaking of "go-carts," someone already thought of that. The Link
trainer actually had their own version of one; very slow, powered by an
air motor, to trace the path of an imaginary instrument flight by
rolling around on an air chart below the trainer. Very precision. I
don't think the pilot had to pump any pedals on that thing, though.)

Craig B.