Restoration of a Knabe Ampico Reproducing Upright for Carol Drummond
100) 
101) 
102) 
100) These are the lifter fingers, connecting rods and flange fingers that transfer
the motion of the pneumatics to the piano's action. The bushings on these were in
good order but they were very dirty and some of the felt had come unglued. Finally,
the old glue residue on the fingers themselves needed to be removed so that they will have
a clean glue joint to the pneumatics.
101) Now the lifter assemblies have been glued back onto the pneumatics. New
clear shellac is used to seal the wood of the pneumatics. Originally, this sealant
was needed to make the pneumatics more air tight. Because I have thoroughly sealed
them from the inside, this coat of shellac is strictly meant to help keep moisture and
dust from contaminating the wood.
102) These are the devices that hold the three tiers of the stack together.
Each component has to be sealed. The original cork gaskets on the air manifolds have
been scraped off and new leather gaskets installed. There are also a series of
leather washers that attach at the top and bottom of the finished assembly that seal the
bolts themselves so that no air can sneak past them. The fronts of the air manifolds
were recovered with heavy bellows cloth and the insides of them resealed with burnt
shellac. I would have preferred to replace the bellows cloth coverings at the fronts
of these manifolds with wood. If I could have used wood, then there would never be a
future need to replace these fronts. However, I wasn't positive as to how much
clearance there is between these manifolds and the key cover. I used the cloth to
ensure that I wouldn't end up with a clearance problem later.
103) 
104) 
105) 
103) The three tiers are now clamped together using the parts shown in the previous
photo. All new leather gaskets were made for them as well. The top board in
this photo is a flange rail that holds all the lifter fingers in place. All these
lifters have been screwed back onto the rail and are ready to be regulated once the action
of the piano is in place.
104) This shows the stack from the other side. All these holes will be covered
by the valves that operate the pneumatics.
105) Now all the valves are in place. Sorry for the fuzzy picture, I am still
doing battle with my new digital camera...
106) 
107) 
108) 
106) The primary valve chest is now back on top of the
assembly.
107) The "head" is the thing which sits atop the entire assembly.
This must be taken completely apart and cleaned, refinished, polished and any moving parts
restored.
108) This shows the head torn apart. One aspect of the design of this upright
unit is that all the indicator plates are made of black celluloid with white lettering.
The lettering on many of these has become faint, one of them has a broken corner
and most of the extremely tiny and hard to replace tacks that hold them down are missing.
Except for the tempo indicator, none of these plates are replaceable so the
originals must be used at all costs.
109) 
110) 
111) 
109) This photo shows the transmission on the right side of
the spool box. It was removed and thoroughly cleaned. All metal components
were cleaned and the frame repainted. All residue of previous lubrication attempts
must be removed from the transmission or it will not operate freely.
110) These are the pot metal control switches. The standard price of
rebuilding this type of mechanism does not include replacing these valves. That is
extra. However, in the case of these valves, the original pot metal is in good
shape. Only one of the valves showed any signs of checking and that was not
interfering with the function of the valve. The replacement valves are always made
of brass which will never check like pot metal can. However, they seldom arrive in
working order. I always have to resurface the new valves before they will operate
properly. To restore these valves, they are ground on fine sand paper backed with a
piece of glass to get them absolutely flat. When both mating surfaces are trued up,
the valve will operate very well.
111) There is a lot to say about this photo. On the left you see the left side
spool retainer for the take up spool. Next to it is the board which backs the spool
box at the bottom near the take up spool. On this board is a mechanical device which
the paper roll engages. The action of this device is meant to tell the system when
the roll is done rewinding. This, then, triggers the system to shut itself off or
replay the roll depending on the setting which has been selected.
As we move to the right, we see the problem. On the right is a left side take up spool retainer and take up spool from a grand system. On the upright Ampicos, the take up spool is solid and the valve already described is used. On grand systems, the spool is hollow as is the retainer. The spool has a hole that is surrounded by a seat that the paper of the roll seals against. When the paper rewinds far enough to uncover the hole in the take up spool, the system is triggered to either shut off or replay. This design is far superior to what this piano came with originally. However, the signal given out by this system is backward to the design of the original system. A reversing valve will need to be added as well if this superior take up spool and retainer are used. The valve is simple enough and presents no real problem to making this improvement.
The upright mechanical design is hard on rolls as the device must ride on the paper at all times. The grand system is much easier on the paper. Normally, I would just restore the piano using its original design. However, we are faced with a problem in this piano. The original take up spool is missing. I didn't become conscious of this fact until I went to take it off for restoration and found it wasn't even there. After consulting earlier pictures of the piano, and searching amongst the parts for this instrument, I realized that the piano was never in possession of its original take up spool. Unless the customer can find what happened to this spool and get it to me, we will have to replace it.
There are several ways that replacement take up spools can be acquired. My supplier does have cheap take up spools for replacement on standard pumper players. I'm not sure they will fit this piano and they are made of clear plastic which would look awful on this piano. The supplier also has a replacement spool for Ampico pianos which looks like the originals. However, the original design has the spool riding on a ball bearing which is very smooth in its operation. The replacements have a Teflon bushing instead so they will not run as smoothly as the originals nor will they seal as completely as the originals. Happily, I do have a spare original take up spool and retainer that I can use on this piano. Because these are from a grand system, we can eliminate the poorly designed rewind valve with the better design. However, the cost of these replacement parts will run to $800 by the time we pay for restoration of the necessary parts and build a proper reversing valve to make the new spool interface properly with the system, so I have to wait to see what the customer wishes to do before I proceed to install any particular design.
112) 
113) 
114) 
112) This photo shows four of the peripheral devices which
are attached to the top action in the upright Ampico design. Starting at 12 o'clock
and going clockwise, they are the wind motor, play/rewind mechanism, automatic shut-off
device and the tracker (which keeps the roll centered on the tracker bar during
operation.)
113) The first device I took apart was the automatic shut off and the on/off switch.
All the original wire was still in the piano. The coatings on this wire are
not to be trusted so all wire must be replaced. Happily, all the fixtures in Ampico
units are attached to the wires via screws allowing me to keep all of them by installing
new wire. In this photo, I am using a multi-meter tester to check the on/off switch
to make sure it is in good working order.
114) This photo shows the valve assembly and pneumatic associated with the automatic
shut off. Also, you can see the on/off switch after it was restored in this photo.
The valve assembly is still apart so that you can see the new pouches, gaskets and
valve leather that has been installed. Everything has been resealed and refinished
as well.
115) 
116) 
117) 
115) The completed on/off assembly and the tracker after
restoration. The tracker is usually covered with cloth of the same weight as the
other medium sized pneumatics (like that of the on/off pneumatic.) However, I ran
short on this cloth the day I was doing this project. I had cloth which was very
slightly thicker that I chose to use on the tracker. The tracker is made to function
by a system which requires that it be absolutely air tight in every way. The action
of this device requires it to exert quite a bit of force but it doesn't move all that
quickly. This heavier cloth will actually serve very well on this device since speed
is not of the essence. The thicker cloth will last longer and will hold up to the
torque better.
116) The restored head has been put onto the stack along with a few of the
peripheral components.
117) Now the play/rewind device is to be rebuilt.
118) 
119) 
120) 
118) This photo shows the play/rewind after it was taken
apart. At the bottom, towards the left, you will see two pieces of pouch leather
attached to scotch tape! Apparently, this mechanism began to leak pretty badly at
some point in the past. Rather than fixing the device, somebody taped pouch leather
over the outsides of the valve seats. This would have prevented the mechanism from
working at all but may have stopped up or at least quieted the leak.
119) The pneumatics and other components have been resealed. The valve box is
completely resealed and releathered. The pneumatics have been rehinged in the
Amphion fashion (using three pieces of hinge that are set in at crossed angles as seen in
the photo.)
120) The pneumatics have now been recovered and tested to be sure they are totally
air tight. The original design of this involved cork gaskets once again.
Because of this, the gaskets were glued on to the pneumatics and on to the valve
box. This meant that the pneumatics could never be removed for servicing. I
replaced this with leather gaskets. The leather seals well enough that the two
screws that are inside the pneumatics, holding them to the valve box, are sufficient.
In the future, if one or both of these pneumatics need to be serviced, they can be
removed quite easily by removing two screws each. You may notice a brass plate on
the face of the right pneumatic in this photo. Both pneumatics have these plates.
Underneath them, you will find a piece of leather which is held close to a hole in
the board. When the pneumatic is placed under suction to collapse it, these flap
valves are pulled against the wood and make the pneumatic air tight. However, you
need to be able to move this back and forth manually as well when you are manually putting
the piano into play or rewind. For this reason, the pneumatics must be free to move
in this way. If these flap valves were not in place, the pneumatics and the valves
inside would prevent you from moving the play/rewind lever. Normally, these flap
valves are simply some medium thickness leather. However, this type of leather will
seep a little bit of air when it isn't glued to something. So that this seepage will
not occur, I rubberized the leather flaps just like I do with pouches. This makes
them totally air tight and also prevents them from rotting over time.
121) 
122) 
123) 
121) A close up photo of the auto shut off.
122) Now is the time to rebuild the wind motor. Amphion grand wind motors are
some of the best ever made. They run very quietly. However, Amphion used a
different motor for their uprights. This motor has some points which make it
superior to all other motors. For instance, the metal plates that the slide valves
run against are heavy and true and never wear or go out of adjustment. Most motors
have no plates but rather run the slide valves directly against the wood of the trunk.
Also, the weight and design of the slide valves themselves is far superior to most
other designs. On the flip side, there are some features about these motors which
are not as good. The connecting wires that come from the crank shaft to the
pneumatics and the valves are rather delicate compared to other designs. These wires
attach to the crank shaft in an odd way. They are wrapped around a delicate circle
of wood which has a bushing in its middle. These wood pieces are not heavy enough to
stand up to much strain, especially the strain of putting them on and off the shaft during
restoration. In addition, the way the wires attach to the wood circles is not very
snug. It is prone to lost motion and "clicking" noises. The
restoration of these went well and everything came out beautifully. However, the
design flaw that enables the connecting wires to click against the wood circles will have
to be corrected by sealing the wires to the circles with burnt shellac.
Also, the guides that keep the slide valves in place are
very delicate in their design. When properly restored, these operate perfectly but
they cannot hold up to any abuse. Once the motor is in place in the piano, it can
run beautifully for decades. However, if you must work with the motor or around the
motor, you must use care not to bump, break or bend anything on the motor. Finally,
the worst aspect of this motor design is that the pneumatics which drive the motor are
glued to a thick layer of cardboard. This is a ridiculously bad design. The
cardboard is very weak compared to wood. Also, the design of the trunks make it
necessary for the cardboard to be partially responsible for the passage of air through the
device. You will see more of this in future pictures.
123) This photo shows the beautiful and heavily made valve plates of this design.
I love these plates! However, the gaskets that must be made for them are
complex and tedious to make. This photo also shows the guides that keep the slide
valves in position. You can see that one of these guides has already been rebushed.
This reveals how delicate these guides are...
124) 
125) 
126) 
124) Now that the pneumatics have been removed from the
trunk of the motor and the old cardboard sanded away, you can see the inner design of the
motor. The six slots take the air from the pneumatics and channel it to the
locations that line up with the valves on the other side. The pneumatics themselves
cannot cover up these slots sufficiently so some sort of cover must be made which has
holes for feeding the pneumatics while it covers up the air slots. The cardboard is
such a bad solution that I didn't even have to remove the pneumatics. They fell
off!! Because of this, I didn't get the chance to really mark the pneumatic
locations well either so I had to rework their locations from scratch.
125) To solve this design flaw, I am gluing a piece of poplar to the trunk where the
cardboard used to go. The poplar is far thicker than the cardboard was and will
never let the motor run properly because the linkages won't reach the extra distance.
After the glue is dry, I will remove the clamps and plane the motor down until the
poplar is the same thickness as the thick cardboard.
126) The poplar has been planed down and locating lines for the pneumatics have been
set up and placed on the wood. The holes that are drilled into the poplar will align
perfectly with the holes in the pneumatics themselves.
127) 
128) 
129) 
127) The pneumatics have been cleaned up and are now being
rehinged. The sixth pneumatic had a crack in one of the boards which had to be
repaired before it was rehinged.
128) After the pneumatics were rehinged, I placed a felt dot inside them to prevent
the cloth from taking a hard crease during the gluing up process. As mentioned
before, if the cloth is creased sharply, it will not last nearly as long. Also, I
sealed each pneumatic board with burnt shellac from the inside to ensure that they are
totally air tight. All the components of the wind motor are shown in this photo in
order to give you some idea of just how many parts in this wind motor must be releathered,
recovered, refinished, replaced, polished, cleaned...
129) The slide valve seat plates are getting their new gaskets made and attached in
this photo. This photo also shows that the slack adjuster wheel has been rebushed
inside and out and the air nipple that mounts to the side of the motor has received a new
leather gasket.
130) 
131) 
132) 
130) Back to the wind motor trunk for a moment. The
next time this motor is taken apart, the pneumatics will not just "fall off"
like they did this time. A restorer has two choices, he/she can heat the boards
until the glue softens enough to remove the pneumatics or they can use a thin bladed knife
to "chisel" them off. Either of these actions could seriously damage the
thin poplar that I used to replace the cardboard. To prevent future damage, I am
covering the poplar with a layer of cheesecloth. This will weaken the joint just
enough to ensure that, no matter which method the future rebuilder uses, the pneumatics
will come off cleanly with no damage to the trunk.
131) The recovered pneumatics have now been glued to the trunk.
132) A photo from the side to show the thickness of the poplar layer that I
installed to replace the cardboard.
133) 
134) 
135) 
133) This is the device which makes the player mechanism
compatible with standard 88 note rolls. Without this device, an 88 note roll would
use its lower and upper notes to signal the expressions in a way that was random, making
the roll sound odd or possibly making it not work at all. This was taken before
restoration began.
134) The valve block and pneumatic have been removed from the main body of the
expression cut out and the two halves of the device have been separated. On the
upper board is the layer of pouch leather which covers the nipples inside. Depending
on whether the lower board is under suction or not will determine whether wind signals
will be allowed to pass through the device or will be blocked.
135) The pouch leather and the cork gasket material have been removed. Also,
all the outer layers of sealing cloth and black paint have been removed.
136) 
137) 
138) 
136) The new pouch leather covers have been installed using
chrome tanned pouch leather. This leather is a little thicker and a good deal
cheaper than the ultra thin brown pouch leather. Since the very thin, blown pouch
leather is not needed in this device and also since the thicker leather will last better
than the thinner type, I used this type leather for this task. In this photo, you
can see that the other board has been chalked so that the pattern of holes in the board
could be transferred to a piece of leather. The leather will form the new gasket
instead of cork.
137) This is the valve block and small pneumatic which were attached to the
expression cut off block. The valve block is simply used to alternately block or
pass on an air signal. The pneumatic is the smallest pneumatic in the instrument.
Its sole function is to cover two nipples which must be blocked when the expression
cut off is activated. Due to its shape and size, this is the hardest pneumatic in
the piano to recover. Because of the challenge of properly restoring this pneumatic,
I have seen many botched repair jobs done to these. Some of the more common mistakes
are to try to patch the cloth rather than replacing it; trying to patch the leather that
seals onto the nipples rather than replacing it; and sloppy attempts at replacing them.
Amphion actually covered this particular pneumatic's sealing finger with rubber
cloth instead of leather. This is not the way they usually approached this.
The leather will seal the nipples better than cloth could so I upgraded the way this was
made as I restored it.
138) The outside of the expression cut off has been refinished and the nipples have
been cleaned and polished. Each nipple was tipped with a layer of burnt shellac
before it was inserted back into the block. When this is done, a bead of shellac
forms neatly at the base of the nipple; ensuring that it is air tight and solidly locked
in place.
139) 
140)
141) 
139) The original label which warned against using oil and
grease to lubricate the motor was made of cardboard and then was glued across the
pneumatics. It was impossible to get it off in one piece because of this. For
this reason, I made a new label to match the old one. This is the photo of the new
label which will be trimmed to size and glued back onto the motor in the same way that the
original was done.
140) I left this photo a little larger in size because I was so delighted
with how it came out. The finished product is better than it was the day it left the
factory. This wind motor runs extremely smooth and quite. With the
improvements I made, it turned out very well. And cosmetically, it is truly
beautiful to look at.
141) This is the back of the stack after all the restored components have been
installed. The next step is to run all the signal tubing.
142) 
143) 
144) 
142) Now all the new tubing has been run in the back,
including the expression lines.
143) This is the front of the mechanism after the tubing runs were installed between
the primary valve chest and the secondary valves.
144) This photo shows the cover board back in place on the mechanism. When I
get ready to test and install the unit, this board will have to come off again, but for
now, the best way to store it is on the mechanism where it belongs. The color of the
board appears a bit blotchy in the photo because the picture was taken before the fresh
paint had finished drying completely.
© Copyright 2010 {David Rodgers' Piano Rebuilding}. All Rights Reserved.