from
Letters to the Editor
"Mechanical Music, Journal of the Musical Box Society International"
Autumn, 1994
by
Charlie Hind
When I was asked to write an article describing how to locate pins on the cylinder of a music box, I was happy to limit the article to that aspect of the construction for two reasons. The first was that the scope of a "How to Build A Music Box" article would be overwhelming to say the least. The second reason was that even though I designed the dimensions and machined all the parts for this musical movement, my only original contribution to the art of music box making was the development of this system of putting the music on the cylinder. Other members of the Society have made most of the other parts of a music box, and I felt this was my contribution to our body of knowledge.
Joe Kubin's inquiry into the obstacles encountered in the making of this music
box is well taken, but I am afraid that a description of all the wrong ways I
found to do things would be better suited to The three Stooges Start A Machine
Shop. Actually I was grateful many times that I worked alone so that no one else
could see the pitiful mistakes I made. Sometimes I tried to laugh at myself so I
wouldn't cry. My only compensation throughout the work was that in the back of
my mind I could see a beautiful, gleaming music box with my name on it. I didn't
know if it played or not, but I knew I could make it pretty! and I wanted it!
At the risk of appearing somewhat foolish, I will try to explain some of the
pitfalls I have encountered in the making of music boxes. However, I can't begin
a discussion of how I overcame these obstacles without giving credit to my good
friends Dave and Carol Beck. They have always been willing to answer questions
(if I could ask the right ones) and offer advice based on their many years of
experience in the restoration of music boxes, and without their help success
would have been far more elusive. Their work, their enthusiasm, and their
sharing have been an inspiration in showing me what is possible in this field.
The first time I tried to make a comb it warped so badly during the heat
treating process that it was unusable. However, I practiced tuning it,
accidentally broke out a tooth, discovered that the thickness of the treble
teeth was all wrong, and generally learned a great deal that I could apply to my
next effort. It was then that Dave suggested screwing the comb to a thick piece
of steel during the heat treating. This seemed to hold the comb in place and
prevented a lot of the warping. Because that first music box comb was to play
just one tune, the tips of the teeth were not narrowed and the teeth maintained
a constant width of .150" for their entire length. This allowed for a slight
amount of sideways warpage that had little effect on the playing of the music.
When I started experimenting with dimensions for the comb of my second music
box, new problems surfaced because the playing of three tunes required far
greater precision along with the narrowing of the tips of the teeth. I decided,
after several failures on small sections of combs, that instead of machining the
tooth thickness before hear treating, I would slit the teeth, heat treat the
comb, and then grind the underneath side of the comb to create the proper tooth
thickness. This meant that I had to buy a surface grinder, but what the heck; I
had wanted one for a long time anyway.
The system seemed to work great on several small practice combs of 20-30 teeth.
The beauty of the grinding on the underside of the combs far surpassed my
wildest expectations. I then spent close to a full day machining what was to be
one of the final combs. Everything went smoothly and I ended up with a beautiful
specimen. Even after hardening and tempering it still looked fairly straight. I
then set it up on the magnetic chuck of the surface grinder and started flooding
it with coolant. I passed it back and forth under the grinding wheel, lowering
the wheel about .001” with each pass. When I was almost to the final tooth
thickness, I reduced the down feed to .0005” on each pass. The work was
proceeding perfectly until all of a sudden I heard a couple of very quick
buzzing sounds. My worst fears were confirmed when I turned off the grinder and
noticed that, sure enough, two teeth were missing from this beautiful comb. I
guess there are no words to fully describe my state of mind at that moment. It
was an empty feeling. Time seemed to stop and I looked around to make sure my
children weren’t listening. After moaning for a while, the realization hit me
that there was only one thing left to do: start over.
This time I had a clearer vision of what needed to be done. A magnetic chuck is
very good at holding thick pieces of steel, but not so good at holding thin
pieces. As I had ground the teeth thinner, there was less metal for the magnet
to attract, and in the blink of an eye, those two teeth had vibrated away from
the magnet and were caught by the grinding wheel. To solve this problem, I made
a clamping system which held the tips of the teeth down onto the magnet during
grinding. Several days later I was relieved to finally be looking at 121
finished teeth on two combs. Unfortunately though, these teeth were not
finished. A closer look revealed a slightly wavy, roller coaster look to the top
of the comb. The discrepancy in the height of the tooth tips approached
.050"-.060", and a careful measurement showed that nearly every tooth was curved
either to the left or the right by .010"-.025". I knew that for this music box
to play properly I had to achieve a “perfect” alignment between the .025" wide
tooth tips and the .022" diameter pins on the cylinder.
The distance between the centers of the tooth tips was supposed to be exactly
.100", making the track width of each of the three tunes .0333". This meant that
if a tooth tip was out of alignment by more than about .008", it would be
touched by pins from adjacent tunes. To make matters worse, because I had
decided to make the tooth tips slightly wider (.003") than the pin diameter to
allow for a slight error in pin placement, I now realized that what I had gained
in greater tolerance for pin placement error, I had to make up for in accuracy
of tooth straightness. The tips should have been cut to the same width as the
pin diameter.
Well, I knew these combs were not going to work in their present condition; but
since they looked so nice, I placed them in an important location on my work
bench and went around for the next few months pretending that everything was
going to be all right. (Not only do I suffer from denial and repression, but
I’ve been told that I have a Pollyanna complex as well.)
I then worked on other parts of the music box as time permitted and tried not to
think too much about the comb problems. When I finally reached the point where
nothing else could be done without having the combs operational, I sat down with
them again. I could either try to straighten each tooth individually, or I could
completely remake both combs. It didn’t make sense to remake the combs knowing
that there was no part of the heat treating process that I could control
sufficiently to guarantee an improvement in tooth straightness; and besides, if
I tried to straighten the teeth and broke one, I could then remake the entire
comb. After rereading the literature on comb manufacture and adjustment, it was
very hard for me to swallow the idea that I was going to have to hit these
crooked teeth with a hammer. The thought was repugnant and slightly nauseating
to me. Also, the fact that I couldn’t even tell which teeth, if any, were close
enough to the correct position to be used as a reference point didn’t help my
confidence either. To make a long story short, I spent the next week hammering
the teeth left and right and bending them up and own. It was some of the most
tedious, blinding, and frustrating work of the entire project. Fortunately I got
through it with out breaking a tooth, and I ended up with comb teeth that lined
up satisfactorily with the cylinder pins.
I think that the hammering and extreme bending of the teeth was successful
because the spring temper of the roots of the teeth was correct. My heat
treating process was as follows: after drilling the screw holes, then slitting
the tips and then the teeth in a piece of 0-1 oil hardening tool steel, I
screwed the comb onto a piece of ¼" thick flat steel plate. After wrapping this
in stainless steel foil to prevent oxidation on the comb, it was then placed in
the kiln and brought to 1200 degrees. After about 15 minutes, the temperature
was raised to 1525 degrees, at which time the comb was removed, and the edge of
the foil quickly cut off with scissors and the entire piece plunged end first
into quenching oil. I then removed the rest of the foil; and, leaving the comb
screwed to the plate, placed it in a toaster over (real high tech). Slowly I
raised the temperature to 500 degrees and left it there for one hour. At this
point the entire comb had a hardness of probably around Rc 55-58. This is very
nice for the tooth tips, which have to withstand the constant abrasion of the
pins; but it is much too hard for the bending part of the tooth which should act
like a spring. To achieve this spring temper at the roots of the teeth while
leaving the tips at a higher level of hardness, I slowly heated the back (the
part with holes in it) of the comb with a propane torch. As the comb reached the
right temperature, a deep blue color began to spread out onto the teeth. I
allowed this color to reach the lead platforms, but not the tips. This color
indicates a temperature that would place the hardness of the bending part of the
tooth at around Rc 48. Next I ground the tips and teeth to the proper thickness.
The drilling of the holes for the pins was one of the biggest obstacles I had to
overcome. The largest pinning wire (already notched) commercially available to
me was .015" diameter. I was never able to get good consistency and reliability
in drilling holes this small in the hard brass cylinders. In other words, the
drill bits broke. Since I had over 1500 holes to drill the method of drilling
had to be reliable. This is why my pins are so big. I drilled .021" holes and
pressed .022" wire into them. Of course, this meant that I had to make all the
pins as well from a long roll of music wire.
The endless screw was an interesting problem. I thought about different methods
of making it for a couple of years. However, before I ever tried, Ted Leverett,
who was then chairman of our Piedmont Chapter of the Musical Box Society
International, told me how to do it. He had learned the method in a workshop
given by the late Lloyd Kelley. It uses a simple grinding fixture whereby a
piece of suitable drill rod is advanced past a small cutoff wheel by means of a
standard machine screw. The drill rod is connected to the end of the screw so
that as the screw is turned, the rod rotates and advances at the same rate as
the screw. This causes the worm teeth to be ground with the same teeth per inch
as the machine screw. The depth of cut of the grinding wheel is increased with
each pass until the final depth is reached. The worm is them hardened and
polished.
For me, the making of the tiny worm wheel was much more difficult. I made it
four times. The first time I simply botched up the indexing. The second time I
made the teeth too radial. The third time, the tooth angles were right and it
ran part of the way around; but it always stopped at the same place. I later
figured out that my rotary table had not been set exactly perpendicular and
concentric to the lathe spindle (for indexing), and the teeth on one side of the
worm wheel were a couple of thousandths off.
My first tune-change snail cam had three teeth and three steps. I never could
make it work, and I don’t remember how many I made. Finally, I decided to make
one with six teeth and two identical sets of three steps. The SECOND one of
those looked beautiful; but when I put it on the cylinder, I realized the steps
were going in the WRONG direction. The third time was the charm, but it still
didn’t work, because the adjusting rod that engages it was mislocated by about
1/32" on the cylinder’s end cap. I couldn't just move a 1/8" threaded hole by
1/32", so the rod had to be moved to the opposite side of the snail cam by
redrilling its screw hole. Then the end cap’s position in the end of the
cylinder had to be rotated to make the governor stop the cylinder at the end of
the tune. After all that, the finger on the tune change lever had to be filed to
the right height to make it all work. When I finished adjusting it, I was truly
amazed to see the cylinder shift after each tune and stop as directed.
A thorough description of the rest of this work would take up far too much
space; so for the sake of brevity, and to warn other’s of what lurks in the
shadows, I will merely list some of the more memorable moments.
I made the bedplate before the combs (big mistake), and I cut out so much space
for the cylinder that the front of the comb base had nothing to rest on. I had
to remake the entire bedplate.
I misdrilled the hole for the screw which holds the governor to the bedplate.
You only get one chance!
After all the excitement of winding up the spring and getting it in the spring
barrel, I found that it was in BACKWARDS. After taking it out and putting it
back in the right way, I discovered that the spring was too long, and it
couldn’t be wound more than two turns. When I pulled it out again, it flew off
the arbor and into a bucket of dirt. (I had just swept the shop so that
everything would stay nice and clean) Give me a break!
When using a buffing wheel to put the final polish on the spring barrel, it was
suddenly torn from my grip and slammed down onto the worktable. A cursory
examination revealed no major damage; and just as I resumed work and was
thinking “Thank God it didn’t hit the floor,” it was yanked from my hands again
with even greater force and this time crashed to the concrete. I didn’t even
pick it up or look at it for at least five minutes; I knew what had been done.
The list goes on and on, and gradually it digresses to the little mistakes we
have all made; but what it confirms in my mind is the idea that determination
and perseverance can compensate for the lack of skill and experience. During the
four and one-half years it took to complete this music box, I questioned my
skill and competence many times, but there was never any doubt about my desire;
and it was that desire along with careful workmanship, not exceptional skill,
that ultimately got me what I wanted.
