Restoration of a Duo-Art Expression system for Skyp Harmon
121) 
122) 
123) 
121) The finished accordions.
122) Believe it or not, restoring this mounting assembly is the hardest part of the
project of rebuilding the Duo-Art expression. The bushings must all be perfectly
tight but not too tight, and polishing the metal is very hard due to its shape.
What's more, these things have an alarming number of parts to them.
123) Here you see the accordion's mounting carriage blown apart.
124) 
125) 
126) 
124) The restored carriage. All bushings have been
replaced with graphite impregnated felt, all metal polished and all wood repainted.
The assembly has been put back together. This project took the better part of a day
to complete. In fact, most of this could be skipped. Even if the bushings are
a bit loose, the thing will still work but it won't work quite as perfectly. In
addition, the cosmetic work is not needed at all to make the unit work. However, I
have a belief about these player mechanisms. They are not just gizmos. These
are beautiful, artistic, sculptural pieces that also serve a mechanical and musical
function. All of the arts come together as one. The decorative arts, the
mechanical arts and musical arts are all brought together into one device. The
cosmetic beauty of these devices (when they are properly restored) is breathtaking and
well worth doing right.
125) Another angle of the finished accordion assembly.
126) The finished expression.
127) 
128) 
129) 
127) From another angle. As I mentioned before, I got
those expression knife valve control arms too tight. I will be taking this partially
apart again to correct the mistake. (Nothing like "truth in advertising."
I make mistakes just like everybody else. But I do try to correct all of them
before they ever return to their owners.)
128) This is a photo of the pneumatic stack before work began.
129) This is a photo from the other side of the pneumatic stack. It shows the
old leather nuts and condition of the pneumatic cloth.
130) 
131) 
132) 
130) This is a photo of the pneumatic stack after it has
been largely disassembled.
131) In this photo, all of the pneumatics have been removed and set aside. The
pouch boards and the pneumatic rails have been cleaned and dressed. Careful marks
were made to ensure that the pneumatics would be put back in the original positions before
they were removed.
132) All three tiers have received fresh coats of sealant on the outside and inside,
including the pouch boards.
133) 
134) 
135) 
133) This photo shows the three tiers of the pneumatic stack
in various states of completion. At the top, all of the restoration to the boards
has been completed; this includes resealing all of the air channels and the boards outside
and inside, this also includes removal of all the old burnt shellac and the cleaning up of
all valve plates.
134) This is a photo of the bleed rail. The old seal cloth has been removed
and the wood dressed. Then the air channels were resealed and new seal cloth was cut
to size. All of the nipples were removed and polished free from all shellac residue
as well.
135) Now the new seal cloth has been glued into place. All of the nipples are
now being inserted into the rail with fresh shellac.
136) 
137) 
138) 
136) Replacing the leather pouches with new ones.
After the glue has set up, the pouch leather will be sealed with Dow 111. The
sealant will be allowed to get completely dry and then it will be dusted off with talc.
Note that all the pouch wells have been resealed as well.
137) The valves have been releathered. The outside valve facings do not need
to be glued down. However, the inside valve facings do need to be glued down.
Because of the way that player piano valves work, many systems do not have any fixators
for the outside valve facings. However, Aeolian valves have little collars which
hold the outside facings snugly in place. Often times these collars can be
reused. However, in the case of this unit, the old callers were shot. Since I
didn't have new callers in stock and didn't want to lose time waiting for new ones to come
in, I made new ones of my own. The design I used for making the new collars is more
than adequate to compensate for the fiber collars originally used.
138) Installing the inside valve plates and valve guides. These plates must be
set into the wood using burnt shellac. Many of the screws that hold these plates in
place were stripped. New, larger screws had to be used to replace them.
139) 
140) 
141) 
139) Now I am turning my attention to the pneumatics.
The first order of business is to seal the insides of the pneumatics The original
design did not call for the sealing of the insides of the pneumatics. However, over
time the wood dries and can develop tiny cracking that isn't apparent until you try to
test them for air tightness. It becomes important to make sure that they are truly
sealed. To accomplish this, I seal the insides of the pneumatics as well as the
outsides. After the sealant has dried, I pull off the old hinge material and clean
the wood where they were glued. Then new hinge material is glued into place and the
pneumatics clamped tightly. At this time, I also install felt bumpers to prevent the
pneumatics from closing down too tightly on the cloth. If the pneumatic cloth is
creased too sharply, it won't last as long.
140) Recovering the pneumatics with new rubberized cloth. The first three
sides are glued down first. Then the cloth is trimmed flush with the wood.
After the glue has dried the hinge ends are glued down.
141) As each pneumatic has its hinge end glued down, it must be closed. Then
the pneumatics must be weighed down to keep them closed until the glue sets up. This
ensures that the hinges will not be glued down in such a way as to bind up the motion of
the pneumatic.
142) 
143) 
144) 
142) The next day, after all the glue has dried, I cover all
of the air holes into the pneumatics with tape. This enables me to test each
pneumatic to make sure it is airtight. At the right of this photo, you can see one
pneumatic that did not pass muster. I was not able to get it to seal properly.
To save time, I took off all the new pneumatic cloth and recovered it a second
time. This got the pneumatic airtight.
143) Installing new construction paper gaskets. The original design called for
pouch leather gaskets. However, pouch leather was very inexpensive in those days but
it is no longer. In addition, construction paper will not rot like the pouch leather
did. This solution for gaskets is a much better one than the pouch leather.
The paper will really never get weak over time.
144) Hear, you see me making new blotter paper gaskets for the outside valve seats.
Some rebuilders insist that you should run a bead of PVC-E glue around the outside
of these valves and blotter paper after screwing them down. Others will say to use a
thin coat of shellac on the paper. All that these things do is to make future
adjustments to the valves very difficult. The blotter paper is excellent at sealing
the system all on its own. Since the later design of Aeolian system calls for valves
which are very small in diameter, it becomes critical to get the valve motion correct.
There must be as much air permitted to pass by the valve as possible while keeping
the motion of the valve to a minimum to ensure fast repetition. If this delicate
balance of function is thrown off the least little bit, the function of that note can
become compromised. Because this balance is so delicate, I think it is foolish to
make it hard to get at the valves for future adjustments.
145) 
146) 
147) 
145) Installing the outside valve plates with their new
blotter paper gaskets. As I install these, I must put the valves inside of the valve
wells and check to make sure that they have the correct amount of motion. Various
sizes of shims are placed under the outside valve seats until the travel is correct.
Then the retaining rings are driven over the valve stem to hold the outside valve
leather snugly against the valve body.
146) The air channels inside of each tier are sealed on the outside with pneumatic
cloth. To ensure that the corners seal properly, dots of pneumatic cloth are glued
over them.
147) In this picture, you can see that the valves with their valve stems and lifter
discs have been assembled. Using a probe I adjust the gap between the valves and the
pouches until there is just the correct amount of lost motion.
148) 
149) 
150) 
148) My reference marks have been translated into layout
lines. These lines will guide me as I glue the pneumatics into place.
149) This photo shows the pneumatic cloth after it has been glued down over the
outside of the tier.
150) Gluing the pneumatics back into place. You can see, in this photo, that I
have totally removed all the sealant from the outsides of the pneumatics. This gives
me a much better glue joint between the lifter fingers and the movable boards of the
pneumatics. This also leaves a much stronger, cleaner and more attractive finished
product.
151) 
152) 
153) 
151) Gluing the lifter fingers onto the pneumatics. Later, the outsides of the
pneumatics will be freshly sealed.
152) Cleaning up the lifter rods. I also have replaced all the leather nuts on
these and made sure all residue of rotten leather and tarnished brass have been removed
from the threads on the rods. The felt up-stop punchings were in good shape so I
kept those. The guide rail bushings were in good shape but a few of them had a
problem with the glue holding them in place having lost its strength. All of those
which were compromised were reglued.
153) After the tiers were checked to ensure they were air tight, they were
reassembled. I did make one minor mistake during the reassembly. I picked up
the wrong batch of screws to attach the cast iron suction nipples. The screws I
ended up using were longer than the originals. This simply means that the nipples
will be held in place stronger than ever. I don't want to pull them out and replace
them with the shorter ones at this point because I suspect this mistake will make the
shorter screws less able to hold tightly than before. Anyway, the original screws
were showing signs of stripping out so this mistake could be categorized more like a
"happy accident."
154) 
154) The lifter fingers have been reattached and their regulation roughed into place. It will be necessary to fine tune the positions of all the leather nuts with the stack in the piano. The next and final operation to be done is to tube the bleed rail up to the stack. However, when I went to pull my new supply of tracker bar tubing from the box, I discovered that they had sent me the wrong stuff. I had to reorder the tubing and the project will have to go on hold until it comes. After the tubing is in place, I can proceed with the testing and fine tuning of the expression device. Then the system will be ready to be picked up by its owner.
© Copyright 2010 {David Rodgers' Piano Rebuilding}. All Rights Reserved.