Restoration of a Steinway X Upright for Richard
Davis
1) 
2) 
3) 
1) The piano has entered the shop and work can begin.
The action has already been removed and the front fret-cut cover boards are out because I
needed to trace them for another piano earlier. The timing of the commencement of
this project deserves a little explanation. I found myself short on tubing to
complete Rev. Menaugh's project. In order to keep the ball rolling for everyone, I
took the delay in waiting for parts to commence work on something else. The tubing
was a long time in coming which worked out fine since the efforts described between photos
#1 to #114 took up a considerable chunk of time as well. It is important to all my
customers to know that my priorities are two-fold. One: Never compromise
quality for expediency. Two: Get the projects done as quickly as I am able.
With that said, let's get started...
2) The design of this model of X (even considering its era) is rather
unusual. It is possible to literally look right through the back of the piano on to
the front because the plate and sound board assemblies don't completely block your
view. I'll show several photos of this to help you get a better feel for what I
mean.
3) Another shot.
4) 
5) 
6) 
4) One last shot.
5) As is not uncommon with pianos of this vintage done as ebony instruments,
they used Rosewoods which are now rare or embargoed. This photo and the next two
show an area of the side from which I have removed the black paint and sanded and cleaned
until I was down to the raw wood. The first two photos are the same although the
second is a little fuzzy. The colors come out slightly different in the photos but
the real color is a beautiful, clear medium red as is common for unstained Brazilian
Rosewood which is the species I have determined we have here. A vastly more
beautiful wood than Macassar Ebony which is the other wood they used for doing black
pianos. It only remains to strip the piano completely and find out if they did, in
fact, use Brazilian Rosewood on everything; and, if they did, was their piecing and
matching adequate for a final presentation? If so, the owner has indicated that the
Rosewood will be far preferable to them than the ebony.
6) Another shot of the Rosewood. Although Brazilian Rosewood is
embargoed from Brazil, supplies stockpiled around the world still leave us able to acquire
material. In addition, I do have a supply of Bolivian Rosewood (Pau Ferro) which is,
in essence the same tree.
7) 
8) 
9) 
7) It is not uncommon for the veneer at the back bottom
of the side of upright pianos to come loose. (It's damp down near the floor and was
especially so in the early century.) I removed this wood chip from just such an area
to help us see the black paint and the Rosewood underneath and just what a difference they
make. To enhance the effect, I scanned the piece of wood rather than just photographing
it. The scan, perhaps, gives a little more detail than you would see in real life. Also,
the pores turned out white in the scan when, in fact, they are deep purple/brown. But this
shot gives you a good look at the veneer. Rosewood of this color and with this interlaced
grain is certainly from Brazil. Indian Rosewood is much more gray in color and the
grain is almost universally straight.
8) In the early and mid century, piano tuners often made a habit of
writing their names and tuning dates onto upright piano plates. It helped them
remember when they saw the piano last, what they did and what may be expected. It
also helped the owner to remember who they called before for tuning since they could look
inside and see who it was.
9) The upper cabinetry has been removed. Also, the board which covers
the top of the piano's wooden frame has been taken off. Many rebuilders (myself
included) often leave this board in place when they find (as I have in this piano) that
the board is glued to the back frame. However, in this case, I wanted to be able to
made quality visual checks of the condition of the glue joints under that board and also
have the board easy to strip and easy to veneer if we find the Rosewood underneath
anything short of what we hope for. Doing this does create nearly a half day's extra
labor. The board is maple inside but veneered outside in Rosewood and on the
opposite side in mahogany. The veneer on both sides is needed to prevent any
tendency for the board to warp between the time of production and when it is glued down.
Boards which never get glued down must have veneer on both sides just to insure
that the boards won't warp. In the case of this situation, the soft mahogany helped
serve as a gasket. It's low specific gravity makes it easy to split. The maple
is anywhere from ten to twenty points more dense than the mahogany so I could split the
board off and only mess up the mahogany. The maple of the board and of the back
support frame remains solid and the mahogany separates. After removal, I put the
loose board through the planer to remove the mahogany residue and I used a hand plane to
remove the residue off the main frame of the piano. More pictures of that process
will follow.
10) 
11) 
12) 
10) Now the key sticks have been removed. Cursory examination shows
the ivories to be in good condition with good glue joints. The bleaching process
will either go smoothly because of good glue or it will reveal hidden glue joint problems
that will plague the owner in years to come. We can then repair those up and coming
problem spots while the piano is in the shop.
11) Both screens have been removed from the back of the piano.
These screens are often found in better quality pianos from the late 1800's.
Remember the lifestyle, technology and homes of those days. They needed means which
they could use to keep moths from entering the piano and also to keep mice, chipmunks,
small squirrels and other bugs and vermin out of the piano. Later in history, houses
were better sealed against such intruders and the screens became largely
superfluous. In any event, we can now see the inside of the back of the piano better
so I took a picture to let you see it.
12) With the knee board out, you can see inside the bottom of the piano
also. The movers broke the right pedal on this piano when it was delivered to my
shop. Since I have replacements and seriously considered that I might want to
replace the pedals altogether, I didn't give the movers a rough time or make them pay for
the broken pedal. There's too many people out there already that make it their goal
to confront, hurt and abuse their fellow man without me adding to their numbers.
13) 
14) 
15) 
13) The piano is on its back on my piano tilter. This
will help me remove the legs, bottom boards, plate, keybed, keyframe, castors and other
various parts. I have blacked out edges. You will find me doing this
occasionally when their is a picture that would become confusing to see if all the
surrounding items in the shop were visible in the photo.
14) The tilted piano taken from the other side.
15) This is a photo of the bottom board which the pedals operate from.
Most piano bottom boards cover the entire base of the instrument. Older
pianos like this one often have very narrow boards like this one on the bottom.
These narrow boards make the pianos very hard to stabilize onto a moving dolly.
This piano had no such trouble because the unusual and "over-killed"
construction of the plate meshed with the bottom board under the piano and left the entire
under surface flat. This board is clearly filthy (normal) and in need of much
repair. But it is basically solid so that an entirely new bottom board will not be
needed although I have made such many times before.
16) 
17) 
18) 
16) This photo shows how the plate was cast with an enormous
undulating flange that fits under the piano and then turns up to make the rest of the
plate (all in one casting.) Honestly, I have never seen an upright piano with such a
heavy duty and beautifully thought out construction before. Not even on other
Steinway uprights which are for all intents and purposes exactly the same as this
instrument. This plate's bottom flange waves out underneath each back post so as to
carry and share string load with each member. Then it circles back a bit between
posts to save on weight. Because the screws that hold this part of the plate into
the posts are under the piano near the floor, moisture had gently wicked up the wood and
rusted the threads of the screws. They were locked into the piano so tightly that
they had to be heated (see photo) for a considerable period of time each to get them to
free up enough to come out without breaking.
17) Time for truth in advertising. My memory failed me and I
messed up. The cheek blocks on this piano (blocks of wood that create ending points
at each side of the keyboard) are screwed to the keyframe (that which holds all the guide
pins and green/white colored felts on which the keys operate) from underneath. I
knew this but plum forgot. When I saw no means to unscrew them, I concluded that
they were glued in place. It is so very seldom necessary to remove the key frame
from the key bed (a component which we will talk more about in a moment) that I jumped to
this conclusion; forgetting my knowledge to the contrary concerning Steinway uprights from
the 1880's and 1890's. I tried to remove the blocks with a chisel. I
remembered my lapse and my error in time to prevent any damage to the cheek blocks.
But the key frame got split. My fault. I removed all the screws that hold that
keyframe in place, removed it and got the cheek blocks off the proper way. Then I
glued the keyframe back together again. I will probably take time to retrofit
cross-grained wood over both sides of the damaged areas to hide the breaks and strengthen
the spots. That will leave the keyframe stronger than it was when it was new so I
suppose all is well despite my being a dunce for a moment.
18) The main body of the pianos structural elements have been completely
stripped of all their attached components. The arrow at the top shows the key sticks
all held together in unison by a pair of boards and a series of clamps; which is how I
transport all keyboards around the shop.
19) 
20) 
21) 
19) Now that everything is out of the way and the plate,
sound board and bridges are all in their original positions, it is time to measure the
scale. Being a Steinway, I didn't have to use a micrometer on the plain treble wire
since the scale (the wire sizes) is stamped right on the plate. This is common
practice at Steinway although I have seen original Steinways that lacked this upon rare
occasions. I still had to measure the speaking length of all the strings in the
piano. The speaking length is that segment of the string that is free to vibrate in
such a way as to send the appropriate note's vibrations into the sound board. Other
sections are separated for other functions like hitching the string to the plate and
attaching it to a tuning pin. It is vital that the design of the bridges and plate
be such that the speaking segment of the string can move readily while its vibrations are
efficiently blocked from passing into the other sections which have no effect on tone.
Now, since this piano was built around 1890, it has to be remembered that the
desired esthetic of the day was one that was simple, soft, not bright, sustained in an
average way but nothing extreme was wanted. They wanted the instrument to sound
pleasing but not monstrous. And they wanted pianos to sound warm, not jazz player,
Yamaha bright. In the remaining photos on this page, I am going to present to you
some confusing graphs. It's impossible for me to convey to you all of the
information I can glean from these charts but I hope to show the work I am doing to ensure
that the finished piano meets modern expectations while continuing to honor the esthetic
that the piano was built under. Later, if the customer expresses a desire to match
scales precisely so that the piano sounds no better than it did when it left the factory,
we can cut it back. But, if my hands are left free, I can up the ante on any piano
(especially a good piano like this one) considerably.
20) This chart shows the original string tensions and
"inharmonicity" that the piano was built with. The smooth black lines on
all the graphs show a default point where a piano of this size would generally fall.
The lines show a lower and upper limit that would be considered average for fine
pianos. This graph shows that our Steinway X was built with very low tension (which
coincides with the eras desire for gentle, soft yet rich sound since low tension does help
produce this.) The graph shows the inharmonicity stretching up above the normal
lines in the lowest bass and the highest treble. On the treble side, this just means
that the tuner will be setting the string tensions a little different to get a correct
sound but it could produce an improved sustain and a louder, more balanced, result in
comparison to the rest of the piano. The bass notes (to the left) that show high
inharmonicity do so because Steinway built this upright piano with bass strings that only
had one layer of copper wrapped onto them. This is unheard of in a piano of this
size. Usually, single wraps around the steel core of a bass string don't come into
play until you get to grand pianos that are seven feet long or longer. Considering
the single wrap solution their scale designers used, I think this scale was brilliant for
its day. All the various factors that effect the sound of the strings are incredibly
even and very appropriate for the goal we now know those men were aiming for. (Sorry
for using "men" but it would have been males alone doing the job back in those
days.) Considering that the only tools at their disposal was previous experience,
research and development and calculations performed with pen and paper, what they
accomplished is nothing short of miraculous. I will continue to gloat over these
fine gentlemen as we proceed. One other comment, you will notice at the left end of
the "trichords" there is a sudden dropping off of tension accompanied by a steep
increase in inharmonicity. (FYI: Inharmonicity is that feature which makes a
string tune cleanly with its neighbors.) The reason for this "flaw" is
that these strings were at the bottom of the treble bridge. Steinway often solved
this by putting wound bass strings onto a tenor (or third) bridge placed in between.
The easiest solution is to change out the trouble notes with wound notes placed
directly on the treble bridge. But they put plain wire on that part of the bridge.
I assume their ears told them that the results would suffice and went no further
since a third bridge and/or additional wound strings had cost concerns that had to be
weighed against profit expectations. Now that good tone is the goal and all else be
passed, we will present a solution to this problem shortly.
21) This is a chart of the original "Loudness/Sustain" of the
piano's stringing scale. This chart is difficult to completely follow because it
represents a relationship between two opposing objects. The louder a piano gets, the
less sustain it produces. The more sustain you give it, the softer it becomes.
Now, that was a glaring generality and in the fullness of my scaling knowledge, the
statement is only partly true. Many other factors come into play that are beyond the
scope of this explanation. What can be clearly discover that means something to us?
Well, first, the bass never gets above 3000 on the scale. This isn't all that
bad. Later, you will see comparisons to larger Steinway grands and discover that
this little upright is holding its own with vastly superior Steinway products. But
my rescaling of the piano is capable of pushing that up dramatically. Much louder.
We needn't worry about effects of sustain in the low bass since the weight of the
strings along with their length produces a fine sustain no matter how high the number gets
so the fight between loudness and sustain is moot in the bass in most cases. The
chart shows that the far left notes (notes which used a single string for each key) ends
up lower on the chart than the next group starts. The next group up the scale
consists of notes which use two strings for each key. The jump in loudness/sustain
is caused by this and is, in fact, not bad in this chart but stands scrutiny.
Finally, we see that the low treble notes have a nasty drop off just before we change to
the bass bridge. This is a flaw in the scale. No matter how much I admire the
scaling team who made this piano's scale, they dropped the ball in this area. In
fairness, most builders did back then... Oh yes, also note how low the plain wire
section is on the chart compared to the bass. This low read out demonstrates the
desire for more sustain and less volume which was the public's preference in piano tone in
that era.
22) 
23) 
24) 
22) This chart is the "Break Percent" of the
original scale. This chart is of little real aural value. But it reveals
something remarkable. With no computers, these men not only came up with a nice
scale that met the public's expectations for piano tone and did so in a remarkable way we
haven't even begun to explore but even the break percent is very smooth. What is
break percent? When a piano string is tightened up until it reached 100% of its
break percent, it will break. If a piano wire is taken above about 66% of the
tension needed to break it, the string will begin to stretch unnaturally. This will
reduce its longevity and could lead to breaking strings. Even amongst the great
maker's scales that look great everywhere, the break percent is usually OK in general but
is a bit "scatter-drilled" when you look at the number. This piano's %'s
are lovely.
23) This chart shows the actual speaking lengths of the strings.
An arrow points to a mistake made by Steinway for which they have no good excuse.
They messed up here. There is a bar on each piano's plate every so many notes
to carry the tension from the pin block above down to the bottom where it is transferred
and/or shared with the back frame. When the bridge passes under this bar on cheaper
pianos, it has to be cut out with a deep notch to make room for the iron bar.
Steinway was smart enough to make the bars beefy enough that they could raise them up in
such a way that the entire bridge can remain its full thickness without any need for
cutting into it. This cutting into the bridge for clearances causes a reduction in
the transmission of vibrations up and down the bridge yielding a lesser voice, less
sustain and less richness. But these bars make it necessary for you to slightly
alter the curve of the bridge so that each note on either side of the bar is correct in
its length to make an even scale. They did the highest one correctly but they got
the bridge made wrong on the point with the arrow. This yielded a scale that has a
serious "glitch" in it at that point. It will have to be corrected when
the new bridge is made. At the very top of that curve for the treble bridge, you
will note how the markers take a curve backwards at the far left. This is because
they moved the bridge down as far as they could to produce the longest possible speaking
lengths through the most important areas of the piano. This backwards portion is
forced upon them because they are nearing the edge of the board where it is glued down.
The bridge won't make any sound if it is placed right over the area where the sound
board is glued down. So, they have to swing it backwards to keep it out in the
movable area of the sound board. Frankly, I'm a huge advocate of the "third
bridge" or "tenor bridge" since it is perfect at solving this problem.
FYI: I have noted that the model A, L, O, S, and X all suffered from this
error in their construction. Only the concert grands seemed to be important enough
to Steinway to step the bridges properly to allow for the problem of making allowances for
the plate struts.
24) I'm sorry but this chart is tough to read. You'll have to use some
patience and spend some time with it and the following charts. Please do so,
however, because there is much to be learned and wonders to behold. This chart shows
the original scale of this 1880 upright piano whose tensions and inharmonicities are
displayed in the form of circles. There is another scale overlaid onto this which is
made up of squares. The instrument being compared is the original scale to a 1920's
Steinway A (about 6' 10" long!) The Steinway A is by far the best piano they
built besides the B and D concert grands (and possibly the C concert grand.) They
also made grands model's S and O which were very small (S) or fairly long but scaled
terribly and generally disappointing. The instrument they made the most of was the
model M which is almost 6' and was the most popular to buy because of its tone/size/price
combination. So, we're comparing this old X upright with the best home model grand
Steinway made. If you will examine the charts, you will note that this piano does a
proud job of holding its own against the "A" grand. Yes, the A is big
enough to solve some of the "X's" problems but, in general, this simple upright
is rocking like a serious grand.
25) 
26) 
27) 
25) This is the same comparison now showing loudness/sustain.
In the upper treble (far right) is a demonstration of reasonably identical
solutions. Since those are short notes, almost all pianos in the world solve those
upper notes the same way. The speaking lengths of the highest note on the piano only
varies (in general) from 48mm up to 52mm from one brand/design to another with only a very
few going outside those parameters. But if you follow the scale the rest of the way
along, you will find that this little Steinway X has a bigger loudness/sustain than does
the high end Studio/Parlor Grand!
26) A comparison of the speaking lengths. Steinway was able
to cram a batch of speaking lengths into this upright that aren't all that far off from
the high end model A.
27) This comparison of break percents shows how even the best grands by
Steinway were left to a bit of a "scatter-drill" in this area. Yet, the X
was neat as a pin.
28) 
29) 
30) 
28) All we see from this comparison is that our little X
matched the big A in tension and inharmonicity except in that the A was able to solve
certain "X" problems due to its length of strings in certain areas. And
that the A was philosophically intended to be a bit bigger sounding than was the middle of
the road design preferred in those days. Of course, remember, the A was built at a
time when preferences for big/bright sound were moving upwards towards what we have today.
29) This is your first glimpse at my initial ideas on rescaling
this "X". Comparing my tentative redesigned scale on the "X" to
the "A" shows that the loudness/sustain of this little upright will blow any
original "A" out of the water.
30) This shows a comparison to our piano's original tension to my new plans
for tension. String tension is one of the primary determining factors on what we can
expect to hear from the piano. It is good to have the tension smooth from note to
note and also to have it as high as can reasonably be accomplished without impinging on
break percent restrictions. In other words, without damaging the wire.
31) 
31) This compares the original scale to the new scale. You will see that the original scale was quite tidy and even while the new scale is a little more "scatter-drilled" as I mentioned earlier. This factor will not effect the tone so long as it is generally placed together and doesn't exceed the 66% rule. It is impossible to include the dozens upon dozens of charts I generate seeking solutions to each instrument's problems and exploiting their unrealized potential. The most recent scale I am work with right now, for instance, places the power of this upright Steinway with power levels that are 60% higher than what you get out of a factory original German Steinway 9' Concert Grand!
© Copyright 2010 {David Rodgers' Piano Rebuilding}. All Rights Reserved.