Restoration of a Mason & Hamlin Ampico for David Lucas Graves & John Herrmann
37) 
38) 
39) 
37)
Rebushing the keys themselves. This shows the bushings installed and the clamping
cauls in place. These cauls are precision constructed to enable me to create exactly
even sized openings for the key's guide pins. Since there will be many months before
we actually complete this action, I will store the keys on the pins themselves to ensure
that any expansion and contraction occurring due to moisture swings will be under the
limiting force of the key pins.
38) The front key bushings are installed in much the same way. These mortises
must be bushed to a large specification because of the larger size of the front key pins.
It is interesting to note about these keys that someone really didn't know what
they were doing somewhere along the line. A couple of the original bushings had
apparently fallen completely out. The bushings that were installed to replace them
were at least four times deeper into the mortise than is correct. This would have
very likely created a sluggish key. Also, the front key guide pins are oval in shape
and had been turned in the past to take up the slop in the keys as the bushings had worn.
This is a fine expedient as a quick and dirty repair on a lower grade of instrument
which doesn't warrant quality work. However, to have turned the key pins on a piano
of this quality is very dubious at best. Once the key pins are turned, you take a
chance that they will dig there way through the rest of the bushing and start to egg shape
the mortise as well. Once the mortises are damaged, it becomes harder to rebush the
keys in a clean, even and stable manner.
39) The original pin block was used as a pattern to rough cut a new piece of
Delignit. Now the rough cut piece of new pin block is being fitted to the plate.
Mason and Hamlin coated their plates with a red material which stuck and lingered
very well. Most of the fitting work of the new block was able to be done using the
original marking medium. However, after a while enough of the medium was used up
that I had to substitute my own. I spray the plate with contact adhesive and then
coat the adhesive with powdered graphite or chalk. This leaves very easy to read
marks on the pin block so that I know readily where to remove material. I do have an
eye condition which leaves my sight bad on some days. If I am doing this task on a
day when my eyes are working perfect, I will use blue chalk. If my eyes are less
than perfect, I use the black graphite. No harm will come of using this graphite
because all the residue on the block will later be either removed or buried in epoxy.
No graphite will ever make it into the holes for the tuning pins.
40) 
41) 
42) 
40) The pin block is set into place on the plate and hit
with a rubber mallet to drive it into the metal. The marking medium leaves itself on
all the high spots which are making contact. Any areas which don't touch yet are
left unmarked. I remove the marked wood and repeat the marking process. I know
that I'm close enough when the entire leading edge of the pin block has areas which mate
with the metal of the plate's lip and when the face of the block is making good enough
contact that no rocking occurs and any tiny gaps are slight enough that I can pull them
closed with my fingers. Once this is accomplished, then I use PC-7 Epoxy to dress up
the fit. Using a releasing material to prevent the epoxy gluing the pin block to the
plate, the block is coated with a thin layer of the epoxy. Then the block is clamped
heavily to the face and lip of the plate. This is left over night to cure.
When the block is removed, the fit to the plate is 100% perfection. The more perfect
the fit of the block to the plate the more stable the tuning and the better the tone will
be. This photo shows where the thin layer of epoxy is on the edge of the
block. Any areas of blonde represent wood which did not retain any epoxy during
clamping. This was wood that fit the plate perfectly without the need of any epoxy
to dress it up. Some rebuilders lay claim to the idea that they prefer to work the
pin block until they have a fit just like this one that is done all in the wood itself
without any epoxy. First, I find this unlikely. Steinway uses no epoxy and
they have a special machine to do a good fit on the blocks. Yet, my blocks always
fit the plates better than Steinway blocks which I have removed. This wood to metal
fit without the use of any filler is seldom achievable to the level that the rebuilders
lay claim to. Many lesser companies make mistakes when using fillers to fit a block.
The biggest offense is failure to fit the block well to the plate before
introducing a filler. The blocks fit terribly and they use filler to compensate.
The filler used should be kept to the thinnest layer possible. The next
mistake many companies make is to use filler which is not hard or durable enough.
I've had blocks come out of pianos where this filler was falling off it was so
brittle. PC-7 or West System Epoxy are the two best choices for this job because
they set as hard or harder than the wood but retain enough flexibility to remain intact
permanently. As a final proof of the value of this method, I believe the original
block in the Mason and Hamlin made use of the technique to good advantage. The fit
was better than any Steinway block I've ever seen.
41) This is what the block looks like after it has received a very good fit to the
plate. When you pound on the block with the butt of your hand you feel as if you are
pounding on the plate itself because there is no gap between the two. Now the block
will be reclamped to the plate to simulate string tension and the holes in the plate
marked onto the block along with the index holes drilled earlier to locate and size the
finished block which is still oversized at this point. The marks I make on the block
are done with a brad point bit. The bit is just small enough to fit into the holes
on the plate. The brad point makes a perfectly centered mark on the wood and also
makes this mark rather deep. By using a fairly long bit at the drill press, any
slight misalignment between the mark and the bit tip during drilling is compensated for.
The mark either pulls the bit over slightly into the correct position or else the
bit digs into the mark and pulls the pin block into better alignment.
42) Using the original block, I adjusted the angle of the drill press to match the
original angle of the original block. I prefer to do this than to set the press at
any predetermined angle. I usually assume that the original builder knew their
plates, blocks and the thickness of their stringing felt. All of these go into
calculating the correct angle for the tuning pins. On lesser brands, I will
sometimes change this bore angle from the original but on a Mason and Hamlin, I feel
confident copying the original angles exactly. A vacuum hose is kept close to the
bit at all times to maintain the coolness of the bit. Even with this, however, the
bit will get warm. As it gets warm, it will grow in size. Therefore, I start
my drilling (while the bit is cold) in the bass. This will leave the smallest holes
in the area of the block where we need the tightest pins.
43) 
44) 
45) 
43) Aligning the index templates onto the holes drilled
through the plate and transferred to the new block. Bridge pins are driven into
these holes and the templates placed over those. Then the three templates are
aligned and a straight line marked across the block with a chalk line. Then the
edges of the block are transferred onto the wood off the templates. All this is cut
out at the table saw. While sawing, I leave the lines behind so that I can dress the
long edge on the jointer to ensure a perfect joint. I also leave enough wood to make
minor adjustments to the edges so that they will fit the piano well.
44) The main case of the piano has now had a final cleaning with alcohol. This
melts the original aniline dye and redistributes the color for an even look. You can
see the legs are also stripped as are all the rest of the boards. The next big job
is to stain, seal, fill and lacquer the wood. The customer will be coming to see the
piano soon so I will leave this job off until they come and we can discuss color.
45) The next stage of refinishing the case involves doing the wood repairs.
The first repair was to replace a missing piece of trim on the tail leg. The trim is
half round and 5/16". This is not a stock item I can just buy.
Therefore, I'm using a wooden dowel rod which is of the correct diameter. The dowel
rod is cut to a longer length than needed and then affixed into this planing guide or
cradle. A hand plane is used to turn the round rod into a half round piece of trim.
46) 
47) 
48) 
46) I'm sorry for the fuzzy focus on some of the pictures
shown in today's updates. I was having some camera troubles which are now sorted
out. After the trim is formed, it got an initial staining to guard against burn
marks and then drawn slowly through my burn in oven. As I drew it through the oven
and warmed it, I applied light pressure to cause it to bend. I was not able to get
it completely arched to match the shape of the leg but I did get it close enough that I
could get it to form around the leg the rest of the way with minimal force. It was
necessary (due to the fact that the original glue joint had been contaminated) to
supplement the glue joint on this trim with small nails which were predrilled. After
the glue was dry, I made sure the nails were set below the grade of the wood and then I
burned in little repairs over each one to make them invisible.
47) Here you see clamps on a number of pieces which needed reglue type repairs done
to them. The one end of the trim piece was just stubborn enough that I needed to
supplement the nail with a clamp until the glue could set. The next day, I found
that even that hadn't been enough and I had to take sterner measures to make sure the
piece stayed cleanly in place and looked correct. That is one of the secrets to
being a good piano rebuilder. Never settle. If you think you have something
all fixed and you come back the next day and realize it isn't perfect, don't hesitate to
dedicate whatever additional time you have to in order to get it right. Even if that
means that you loose days and do work that gets you no remuneration. Don't stop
until it's right.
48) The main case of the piano after it had been sealed, filled and resealed.
It is now ready to receive the lacquer finish that awaits it.
49) 
50) 
51) 
49) The rest of the loose boards after they were prepared
for lacquer.
50) The lid also is ready for lacquer. Mason and Hamlin found a very special
piece of veneer for this lid. It is of the same species as the rest of the piano but
has special mottling in the grain which is outstanding. It was a joy working on this
so far and I look forward to seeing how it will play out in the final product.
During the process of refinishing I spend a great many days doing tasks which are all the
same and don't really warrant photographing. It is during this part of the job that
the work seems to be moving the slowest as you watch it over the Internet. Rest
assured, the work is progressing at a good speed but it is a lot of the same tasks which
just don't need to be listed here. I can show the piano being rubbed out, for
instance, with one photo. However, the job may take a week to two weeks or more to
complete...
51) Sorry for the blurry image on this one. Some day I'll get used to this new
camera! As I was doing a final check on the rim of this piano before passing it on
for lacquering, I discovered an area of veneer which was loose. Rather than reglue
veneer which is already cracking and loose, I cut all of it off and prepared a fresh piece
of quarter sawn mahogany veneer to put in its place. The technique being used to
apply it involves coating both surfaces with aliphatic resin glue. The glue is
allowed to dry just until it turns clear. Then the veneer is literally ironed onto
the piano. The drying yet still soft glue is able to melt with the heat and the two
surfaces bond. The next day the glue will be set hard enough that heat application
will have little effect on it. Also, this glue is totally waterproof.
52) 
53) 
54) 
52) Now it is time to start rebuilding the Ampico player
mechanism. This is a photo of the pneumatic stack before work began on it.
53) Here is the stack turned upside down to show the rest. You can see that
somebody recovered the pneumatics not all that long ago. However, as I was pulling
these pneumatics off the stack, I discovered that they were sloppily aligned, and the
original cloth had not been completely sanded from the edges during that last restoration.
If, when this previous work was done, the work had been done properly and
completely, the mechanism would still be working today. The cloth was in excellent
condition but the pneumatics were badly aligned and the cloth prone to leakage due to
contaminated glue joints. Also, the valves and gaskets had never been restored
previously although it is obvious they were toast long ago...
54) The movable boards have been removed from the first tier and the lifter fingers
taken off. Even though I was gentle with the fingers and heated the wood to soften
the glue, the fingers came off with a lot of damage to the pneumatics which dates back to
the previous restoration. Between the damage done to the fingers and the fact that
these fingers are very brittle, I have decided to replace them all. In addition, the
screws that reinforce the glue joint holding the fingers in place have all been badly
mangled. I was barely able to get them out of the wood. I will not reuse them,
new ones will be installed instead. You can also see in this photo that the unit
valve blocks have already been removed from the trunk.
55) 
56) 
57) 
55) The pneumatic boards have been warmed and then pulled
free from the trunk. Now the area of wood where they were glued must be sanded
smooth and free of old glue and the cheesecloth gaskets. The heating of the
pneumatics helps preserve the trunk from damage during disassembly. However, this
heating also tends to cause the pneumatics to warp. To ensure that they go back to
totally flat, I dampen the side of the pneumatic which was heated directly and then clamp
the entire set tightly together. This allows them to draw back to a flat condition
before I start any further restoration work on them. In order to ensure that the
pneumatics go back onto the stack in the correct orientation, I have made witness marks in
a couple places on the trunk. Then, I created a scale stick which had marks for each
tier and where the pneumatics were set and were their corresponding fingers were set.
Since the previous job was sloppy, the scale stick will only serve as a guide.
Much of the setting of the pneumatics and fingers will have to be realigned to the
poppet rails the hard way.
56) Every hole in the trunk is now carefully taped shut with masking tape.
Then, when I'm sure the tape is thoroughly sealed, I fill the internal void of the trunk
with shellac. Then the shellac is drained back out. This ensures that the
trunks are totally air tight. They must not leak any air through the wood and
especially not through the divider that separates the inside of each tier. You see,
each tier consists of two separate internal air chambers; one for bass and one for treble.
If the bass is to play very softly while the treble is set very loud, any leaks in
the dividers will cause the high tension side to raise the suction levels on the low
tension side causing a musical imbalance. The previous rebuilder had clearly never
done this. As I taped off each tier, I tested to see if any were showing signs of
severe leakage anywhere. They were all pretty tight. The difference to the
finished product resulting from this step is minor at this point in time and the job takes
a long time because there is so much to tape off. That is why most rebuilders skip
doing this. However, taking this step will give us added safety against future
leakage development and it will make the stack just that little bit tighter. The
ideal is to have the finished product absolutely tight as a drum. If this is
accomplished, the mechanism will work surprisingly better than it would if the seals were
only fairly tight.
57) This tier has been internally sealed and is now getting marked to show where the
pneumatics are to be placed during reassembly. These marks will probably be altered
a bit as I get ready to actually glue things up. I will want to make sure that the
finished glue up is as good or better than factory original; unlike the previous job...
58) 
59) 
60) 
58) The outside, metal, valve seats are being heated on an
iron to soften the shellac which holds them secure. Then the seats are gently
levered out of the block. The valves are also removed at this time. The
outside metal seats have old burnt shellac sealant on them that must be removed. So
they will get a soak in alcohol to dissolve the old shellac. This shellac is so hard
and dry after all these years that it doesn't usually dissolve completely. Some
flakes of shellac will have to be wiped off each seat as I remove them from the alcohol
bath. Each seat is then checked to be sure it is flat. They are actually made
of nickel plated brass. If I have any damage to the nickel plating, the seats will
have to be flattened and then a coat of lacquer applied. Nickel and leather get
along fine. However, brass and leather do not. The tanning agents in the
leather will corrode the brass and the brass would rot the leather.
59) The old leather has been removed from the valves themselves. As is usual,
the leather which was exposed to outside air the most was in the worst condition.
Those facings were totally rotten. The inside facings weren't bad and could have
been reused. However, I never do this with valves. Once you have them this far
apart, the small extra amount of work and materials needed to replace the inside valve
facings is not significant enough to warrent skipping that step. Here, the wooden
valve heads have been cleaned of old leather and glue by sanding them on fine sandpaper
which is backed with a piece of glass. These wooden discs are now perfectly
flat. The inside valve facings are actually tacked to the valve stems. This
tack connection is the only connection between the valve head and the stem. This
allows the valve to wobble slightly. This wobble ensures that the valve will seal
completely should the inside or outside seats be slightly less than a perfect 90 degree
angle to the stems. Amphion designed these wooden valve heads to have a dimple on
one side to accomodate the head of the tack. They were, however, very lazy about
taking advantage of their own design. I have always found numerous valves whose
heads were assembled upside down at the factory. In the case of this piano, some of
the older style heads with the dimples were used (some correctly oriented and some upside
down) and also another design was used which had a smaller dimple combined with a dot of
felt. The felt dot was intended to keep the tack head from making a tapping noise
during valve operation. This puzzles me. The original design with the dimple
in the head works perfectly. So, I removed all the felt dots (these dots only served
to make the valve operation spongy anyway) and made the dimples on the second style valve
deeper. When reassembling the valves, I will be able to set the tack heads into the
dimples so that no noise will occur and the valve will work perfectly. I will never
understand Amphion's haphazard way of treating these valve heads. I can only assume
that all new staff who were untrained and coming into the factory for the first time got
stuck with the job of gluing up valves and this is why the work was done so badly...
Now the sides of all the valve blocks have been sanded free of
original orange burnt shellac. This enables me to see the glue joints between the
layers of wood which make them up. The bottom layer (called the pouch block) must be
carefully split along these glue joints and removed. Then the glue joints will be
dressed and the old pouches removed along with their lifter discs. All old glue will
be scraped away and then we will be ready to start putting in new leather. I seldom
find Amphion mechanisms which have restored valve blocks. These are some of the
hardest valve blocks to restore properly that were ever made in quantity. This means
that most amateur and average professional rebuilders usually try to reuse these valves
without rebuilding them. The results of that lazy approach are usually pretty awful.
60) The valve block to the left has been split apart and the old pouch pulled free.
The valve to the right has been restored and is ready for a new pouch. The
glue joints have been sanded on a known flat surface to ensure a good joint. Care
has to be taken at this stage not to remove any wood if possible. If some wood must
come off, it must be very minimal. On a couple of the blocks, there were, in fact,
some problems with the wood here. I had to add layers of veneer to replace wood that
had to be sanded away. In the case of one pouch block which had cracked long ago and
fell apart in my hand, I placed cross grained veneer over the bottom of the block to
ensure that it remained air tight and structurally stable permanently. The black
valve seats (made of Bakelite) are usually still solidly in place. A little extra
shellac around their edge is all that is needed just to be sure they're snug.
However, on this piano, most of these valve seats were loose. I removed them all and
dressed them on fine sandpaper backed with a piece of glass just to be sure they were
clean and flat. Normally, these valves have very little internal shellac sealant in
them. The sealant around the valve seat is about all there is except for a little
light sealer in the wood. The company relied upon the heavy layer of burnt shellac
the valve body was to be encased in to keep it air tight. In addition, these valves
are made of maple which is quite resistant to air seepage. However, due to the age
of the wood, I preferred to add a good amount of orange shellac as a sealant inside the
valve bodies. This will ensure against any seepage due to hairline flaws developed
over the years. It will also preserve the wood from moisture infiltration and leave
the overall finished product even more air tight than the factory left it. (And
that is very air tight indeed!) In addition, I sealed the recesses of the pouch
blocks. Normally, the exterior sealant is all that is used to seal these pouch
blocks. I have rebuilt many Amphions with that method with fine results.
However, because the sealant doesn't go on until the block is totally restored and
reglued, it is impossible to properly test the air tight nature of the pouch blocks until
after assembly and layering with shellac is complete. By sealing the interiors of
the pouch wells, I make them double air tight and also make it possible to bench test them
for air tightness before I glue everything back together.
61) 
62) 
63) 
61) The valve on the left is split apart but otherwise
untouched. The middle valve has had its glue joints trued up and the valve seat
popped out and cleaned up. The slight difference in color of the middle valve around
the center hole shows just how little sealant there actually is in these valves. The
nature of this design means that a small amount of seepage through the wood is inevitable
by original factory design. The valve on the right has been given a proper amount of
sealant to ensure a totally air tight product. The inside valve seat has been sealed
into position and the pouch block is ready for a new pouch.
62) This is the poppet guide rail. The felt of this rail had been eaten by
moths and one or two of the bushings had been badly replaced in the past. All these
bushings have been removed in this photo and the rail is ready for new felt.
63) The new felt is being installed. The wooden dowels are just the right size
to make the felt dry in a dimension which will hold the poppets securely without allowing
excess wobble or sticking; either one.
© Copyright 2010 {David Rodgers' Piano Rebuilding}. All Rights Reserved.