Restoration of a Mason & Hamlin Ampico for David Lucas Graves & John Herrmann
64) 
65) 
66) 
64) I have added more information about the valves to those
photos on page two on and around #59. Here, you see some completed valves and in the
distance some half finished valves. In order to get the valve heads the right
thickness, Amphion used two different thicknesses of leather on the valve heads.
Since I, too, did not have leather that would be just the right thickness to be able to
create the final valve size desired, I had to use two different types as well. the
white leather is glued to the outer surface of the valve body. I used the white on
the outside because it is thinner. Since the inner layer of leather must support the
tack which holds the valve stem to the body, I used the heavier leather there.
65) All new pouches have been glued to the pouch blocks. Although the valve
bodies look assembled, they are not glued together yet. They are only stacked
together to keep the parts that belong together in order. The valves are
interchangeable but you really shouldn't swap pouch blocks from their original valve to
another. Slight variations in valve body shape and size would mean sanding away wood
we want to keep if that were done. A dot of masking tape is attached to the centers
of these pouches to block that area from absorbing sealant. Then the pouches are
triple sealed with thinned rubber cement. Usually, three coats is enough to ensure
an air tight pouch.
66) Now the masking tape dots have been removed and the lifter disks glued onto the
pouches. These disks help the pouch move the valve by acting as a support that the
pouch can push on. Once the glue and sealant were totally dried, the pouch blocks
were tested to be sure that they were totally air tight. Any blocks which didn't
pass with perfection, got additional coats of sealant until it did pass. Once every
pouch was totally air tight, they were dusted with talc to keep them from being sticky.
67) 
68) 
69) 
67) Now it is time to glue the valve bodies back together.
When using spring clamps for this, I like to use at least six clamps for each valve
body. It takes considerable pressure to do a really good job of closing the glue
joints well. The reason for this is the pouch. The pouches are partially
positioned in the glue joint area. Enough pressure needs to be used in order to
compress the leather and leave the wood to wood glue joint totally tight. Since I
don't own 498 spring clamps, I let these sit until the hide glue has started to gel enough
that the boards won't slide. Then I take the spring clamps off and replace them with
one heavy duty ratchet clamp. This gives me the pressure I need. The ratchet
clamps are almost impossible to use when the glue is still wet. They tend to make
the parts slide against each other. The spring clamps are very easy to use.
You can add the pressure slowly, ensuring that the alignments stay perfect. The
heavier clamps won't misalign the parts after the glue has taken enough of a set to
prevent it.
68) The valve blocks are left in the clamps over night so that the glue will harden
thoroughly before I remove the pressure. In this photo, many of the valves have been
taken out of the clamps. Then I very lightly dress the edges to remove glue squeeze
out and true up any imperfections in the alignment. Then the valve blocks are sealed
all around five sides with burnt orange shellac. This is done in two stages so that
I don't have to worry about layers of shellac on the sides of a valve body sagging over
time as it dries. Once the shellac is dry, the new gaskets are installed. Cork
is the gasket of choice for Amphion from the factory. They used cork very
successfully. It is possible to get a terrific seal with cork. However, cork
requires a great deal of extra clamping pressure from fixators and screws to keep the
pressure tight enough to prevent seepage. Many primary chests are layered with a
strip of sealant tape around their joints because of this tendency to develop seepage.
I have also seen Amphion stacks develop stripped screw holes because the needed
clamping pressure was more than the wood could take. For this reason, I always
replace all the cork gaskets with leather. Premade cork gaskets can be bought quite
cheaply while the leather is much more expensive and takes up a great deal of time since
it must be hand made. However, the leather stays air tight with much less clamping
pressure. I feel that over all cork tends to be slightly less air tight over time
than leather and the leather is much easier on the screw holes. That is why the
valve bodies pictured in this photo are all gasketed with leather.
69) This shows valve bodies in varying stages of work. The ones on the left
are shellaced and drying. The ones in the middle have their gaskets installed and
some of the ones on the right are blanks.
70) 
71) 
72) 
70) Once the shellac is totally air tight, it is time to
install the restored valves. This is done in four steps. First, the valve is
dropped into the valve block and its seal tested by placing suction on the lower hole.
If the lower hole is totally air tight with just the valve dropped into place, I
know that the inside seat is sealing perfectly and that all the redone glue joints are air
tight. Second, the outside metal valve seat is pressed into place. A tool is
used to set the position of this outside seat. The important thing is to get the
travel of the valve just right. Too much travel and the valve will work slowly; too
little travel and the air won't pass quickly through the valve so the pneumatic will
operate slowly. Once I have the travel set with the tool, I check the valve for air
flow and also eyeball the valve travel to confirm that my experience and tests confirm
that the valve travel is correct. I make this travel just a little snug. It is
not so snug that the pneumatics will operate sluggishly, but it is snug enough that as the
leather compresses over time, the valve will not become sluggish due to too much travel
developing. In fact, although the mechanism will work very well when it is newly
finished, the settling in of the valves and other factors to do with the strings and the
hammers will cause the piano to get better in its operation over the first year. The
final step on this photo is to line the outside edge of the metal valve seats with shellac
to seal them. These seats also have slits in them that were necessary to shape the
metal at the factory. These slits must also get a drop of shellac on them (there are
six in all for each seat) so that no air can leak through these either. The next
day, a final test is performed. The valve is allowed to seat onto the outside valve
seat and the block tested for air tightness again. This ensures that the sealing of
the metal seat was totally successful.
71) The valves are done. They have been bench tested and proven to work
perfectly and are ready to be put back into service.
72) Now it is time to restore the pneumatics we removed from the stack earlier.
The heat and moisture associated with removing the boards safely and of removing
the lifter fingers causes the boards to want to warp. They are, therefore, set in
stacks and clamped tightly to keep them totally flat. Now, in the case of this
piano, these pneumatics were recovered once before. The job was sloppy and the
result was that the lifter fingers were mostly impossible to remove without destroying
them and much damage was incurred to the pneumatics in the process of getting the fingers
off them. Also, the previous restoration had left the visible portions of the
pneumatics cosmetically very messy.
73) 
74) 
75) 
73) Because the boards were such a mess, I chose to
resurface not only the sides (as is usual) but all faces as well. The original
hinges had been reused before but were not worth salvaging this time. (I never
salvage hinges anyway.) In addition, the areas of damage to the pneumatics where the
lifter fingers had been so badly butchered by the previous restoration have been repaired
by using a special filling epoxy that simulates the characteristics of maple very well.
74) The excess epoxy has now been sanded even with the surface of the pneumatics.
75) The new hinges have been made, ironed in half so that they are the correct shape
and then glued into place. Thin strips of plastic are laid into the folds of the
hinges to prevent the glue from seeping through the cloth and gluing the halves of the
hinges to each other. The pneumatics also received felt dots near the other ends
that will keep the pneumatics from closing completely. These dots will keep the
pneumatics slightly open even when they are clamped to the rails again. I do this to
prevent the new cloth from developing a crease. Once the cloth is sharply creased,
its life expectancy drops considerably. The glue is allowed to set up over night
before the clamps are removed.
76) 
77) 
78) 
76) Now that the clamps are off, I resealed the outsides of
the pneumatics. I also sealed the insides as seen in the next photo. The
insides of the pneumatics are hardly ever sealed and never on Amphions. However, due
to the age of the pneumatics and the terrible work done on them previously, I wanted to
take out insurance against any leakage I can't readily see. The shellac coat that is
on the visible boards in this photo also cover the areas where the lifter fingers will be
placed. When it is time to place the new lifter fingers into place, I will scrape
the shellac under the finger away so that it can be glued in place properly. The
underside of the pneumatics where they glue to the rail did not get any shellac on the
outsides. This will be added after the pneumatics are glued back on the rails.
This will be done carefully to prevent any shellac from getting on the new rubber
cloth.
77) The freshly hinged pneumatics set up on their sides to the shellac painted onto
the interiors can dry better.
78) The first step in recovering the pneumatics is to cover three sides. This
photo shows that having been done.
79) 
80) 
81) 
79) Next, the excess rubber cloth is cut cleanly away
leaving a nice edge on the pneumatic. The pneumatics are then closed so that the
cloth folds into place properly and the hinge edges glued into place. The excess
cloth is cut away and the pneumatics stacked in the closed position. This method
ensures that the rubber cloth being glued to the hinge edge will dry in such a way that it
will still permit the pneumatic to close easily.
80) The primary valve chest before work began on it.
81) The chest is now opened to show the interior.
82) 
83) 
84) 
82) The primary valves have been removed. They are
interchangeable but the heads and stems should be kept as matched pairs.
83) A close up view of the primary valves. As usual, the outside facings were
completely rotten but the insides looked brand new. Since it is not a huge task to
replace the inside facings, I always put on new ones despite the fact that the originals
look good.
84) One pouch was tested to failure to determine the condition of the pouches.
These pouches were in good enough shape that they could have been resealed and put
back into service. They probably would have given good service for another decade or
so. However, I don't work that way. All pouches get replaced with the
exception of the expression lock and cancels. More on those later. Also, the
primary chest is gasketed with cork. I will be replacing all that with leather as
described earlier.
85) 
86) 
87) 
85) The pouch wells were nicely sealed by Amphion.
However, to ensure that no hairline cracks might leak a bit, the wells all got a fresh
coat of sealant. The brass bleed cups were removed, and all the pouches scraped away
and the edges of the pouch wells sanded to ensure they were free form glue squeeze
out. Actually, installing the primary chest pouches is one of the more challenging
tasks. I usually ruin one or two in the process because I don't get them lined up
properly when I place them into the glue with my pouch setter.
86) The idea of setting pouches is that they must have a sag or dish in them.
This enables them to pop in and out thus moving the valve. The leather is flat and
must be cupped before it is laid into place. If the leather were laid flatly onto
the glue and then the dish or cup worked into it, glue would be dragged into the moving
part of the pouch. This would make the pouch operate poorly and would eventually
cause the pouch to fail as the dried glue becomes sharp and saws through the
leather. This photo shows the completely restored pouch board with the metal center
divider reinstalled with a sealant layer of burnt shellac and the valve board after it was
resurfaced using fine sand paper and a piece of glass to ensure a flat surface for the
valves to mate with.
87) Releathering the primary valves. The inside valve facings are not glued
down; ever. In fact, there was nothing wrong with the old ones but they are too easy
to replace to bother with keeping the old leather. The outside faces are glued to
the valve head. A piece of glass is used to press the leather into the glue to
ensure a flat surface to the leather. Then the valves are placed into the valve
board and a special spacer is used to set the outside heads just right on the stems to
give the valves the exact amount of motion that they require. Too much motion will
interfere with repetition and too little motion will make the secondary valves and the
pneumatics act sluggishly.
88) 
89) 
90) 
88) After all the primary valves are releathered and in
place, a dot of glue is applied to the top of the valve stems. This dot of glue is
all that is needed to hold them secure. In the future, the glue can be snapped off
and the stems knocked back out of the outside wooden valve heads and releathering done
again with no harm to the valves.
89) The pneumatics are all recovered. Now each one is tested to ensure it is
air tight. The air hole is covered with tape and the pneumatics is pulled open.
If the pneumatic will not open, then it is air tight. If it tries to open
slowly, a leak needs to be located and corrected.
90) Preparing the trunks to receive the pneumatics. The trunk is first layered
with cheese cloth. A very watered down hide glue is used to attach the cloth.
The purpose of the cheese cloth is to make the job of removing the pneumatics the next
time easier. The joint created by gluing the boards together with a layer of cheese
cloth is a strong joint but not one which is difficult to reverse in the future.
They will hold as long as we need them to (many decades) but the cheese cloth will prevent
damage the next time the pneumatics have to come off.
91) 
92) 
93) 
91) By using a straight edge, a tri-square and the indicator
marks I made when I removed the pneumatics, I can glue each pneumatic neatly into place.
The last person to do this job was not so neat and the pneumatics looked really
sloppy and the alignments of the poppets were off. This time, they should all look
great and align perfectly. The only thing that will look less than perfect is that
some of the fixed boards may not align to their neighbors as cosmetically perfect as the
moveable boards. Sometimes the finished pneumatics are not absolutely parallel to
themselves. In this rare instance, the moveable board is the one that must be
perfectly placed even if the fixed board has to be a 1/16" off.
92) All the nipples are placed back into the primary chest and burnt shellac is run
all along the wood around them to ensure they seal tightly. This is a job that is
usually done before the gaskets are put into place and before the chest is screwed
together. I forgot... So I had to be careful with the shellac to make sure I
didn't get any on the gasket. If I did, it wouldn't hurt anything but would be
cosmetically unattractive.
93) All the pneumatics are glued into place and held with a clamp. When the
glue has set I will take the clamps off. You can actually do this job without
clamping but I like to use the clamps because it gives me more control and ensures a
really solid glue joint. I installed felt bumpers inside the pneumatics to keep
these clamps from pinching the pneumatics closed. This saves the cloth from
developing a hard crease which would shorten its life span.
94) 
95) 
96) 
94) Testing the primary chest to see it is air tight.
For me, the best way to be sure a chest is tight uses an old fashioned device rather than
suction motors and gauges. I seal up all the nipples, attach a hose to the suction
supply nipple and then I suck on the hose good and hard. If the chest pulls up hard
and won't let me pull any air, I know it is absolutely air tight.
95) Now is the time for suction motors and gauges. A suction turbine is placed
onto the chest and set at a minimum intensity. Then each valve is listened to
through a stethoscope. I'm listening for leaks around the valve. Then I take
the signal tube blocker off the nipple to that valve. This activates the valve.
I listen to the valve again to see if the inside seat is leaking. Once I'm
sure both sides of the valve are silent and not leaking, I check repetition on the valve
to ensure that the valve is capable of repeating faster than the piano is capable of
repeating. I also listen to the valve moving from seat to seat. The sound of
the valve will tell me for sure that I got the valve travel right. If the valve
spits, I know I have too much motion, if the valve is too quiet, I know I don't have
enough.
96) The cover board is refinished and placed back onto the chest. I added a
piece of thin felt to the gap in the cover board. This will let air move in and out
of the valves as needed but will hold back the sound of the valve operation and will
filter the air to keep dust out of the chest. This isn't something that was original
to the piano but I like to add this little touch as I think it is of some use.
© Copyright 2010 {David Rodgers' Piano Rebuilding}. All Rights Reserved.