Continue Sun / Moon rise set, dial output drive assemblies - June 2017
This month Buchanan continues with the output drive section of the sun/ moon
rise-set module responsible for the sun revolution, sun horizon shutters and moon
sphere revolution and rotation.
Here one sees the
‘technical’ workbench where the design and mathematics are studied and
worked out in an “as you go” basis. One can see these drawings and
calculations on the desk below the machine components. No detail design
drawings for the entirety of each subassembly were made at the beginning of
the project. But certain of the more complex subsections were carefully
rendered as Buchanan began those systems. Indeed for this module there could
not have been complete drawings since the functionality of this complication
using Janvier variable differentials was not contemplated at that time.
The first photo shows a
jewel being ground down to a smaller diameter to fit in the space between
the main master gear and the small gear drive above the bearing. A standard
watch jewel would be the right size but too thin since there needs to be a
greater cross section to suit the arbor pivot.
Here we see the hole
within the brass bearing ring where that jewel will reside. The diameter of
the hole is just barely narrower than the width of the ring. Next one can
see the contrate wheel that will mount to that bearing. This wheel is part
of the drive that gives rotation to the moon.
These photos show the
degree of effort made to spoke out even the smallest of wheels. Many other
makers would have skipped this laborious step and left it as a solid wheel.
This video shows the initial demo of the revolving platform containing
the rotating moon assembly. The complex gearing shown behind this in the
second clip are the pair of variable differentials that account for the
first and second order orbital anomalies of the moon's orbit.
This drawing reveals
the design of the moon drive and is an example of a carefully rendered
complex subsystem. The angled drive is needed to correctly
position the moon within the dial. An arbor equipped with a universal joint
was first considered, but it was discovered that the output needed to be
reversed or the moon would spin in the wrong direction. This arrangement
allows for a set of idler wheels to accomplish this.
The entire assembly
rotates around the central axis at a variable speed controlled by the pair
of variable differentials that account for the main part of the moon’s
orbital anomalies allowing the moon to rise and set within a fixed horizon.
In contrast the sun’s position rotates at a constant rate with a set of
movable horizon shutters to account for the seasonal length of the day and
Here begins the
fabrication of the bracket that will hold the two drive wheels that are
angled to each other for the moon drive.
The first photo shows
the setting clutch of the moon mechanism; next some of the other parts
within the assembly.
The first photo shows an initial setting of the potence jewels for the
pair of angled moon drive gears. The
following three photos show sections of the bracket being cut and machined
away to make it as small and delicate as possible.
A bevel wheel is cut
off from the tapered, toothed blank. Next the bracket is now placed within
the context of the rotating platform.
Further refinements to
the complex bracket shape are performed on the EDM, (electrical discharge
machine). The negative mold shown in the first photo is represented by the
large copper anode just above the surface of the part. Next the anode is
raised to reveal the bracket surface.
The bracket is
test-mounted to the platform and next further turning and refinements are
The finished bracket
now mounted to the platform and the first moon drive arbor is mounted between
the bracket and its opposite support potence. Next the arbor is finished off
and has the drive and contrate wheels attached to either end and the
contrate wheel is now mounted upon the potence jewel.
The drive train for the moon’s rotation is complete. The photo
accentuates the diminutive size of this component. Notice the thicker two
spokes on the right, these add structural rigidity where needed for the
support of the bracket and attached moon.
The rotating platform
is now mounted to the sun/moon module front plate. Notice in the first photo
how Buchanan has added some decorative turning to the bracket.
Front view of the
platform behind the inner twenty four hour dial ring. The square brass blank
representing what will later be the module main support frame in seen in the
The completed drive
platform is now ready to join the rest of the sun/moon module. The solid
disks are what will later become the sector gears to drive the sun rise and
set shutters. Can you put this back together?
These photos show the
rear and front of the retainer cap which will be part of the ‘quick release’
system to allow for the complete removal of the sun/moon module from the
rest of the movement; the same as the tellurian, perpetual calendar and will
be for the planisphere and orrery.
Now begins the
decorative process for the frame plates and other components. Here
Buchanan’s mother contributes her design talents. The full plates are shown
on the table (the round and square plates), Buchanan says “This is the most
rewarding part, a little like when a statue is unveiled”. I am indeed
fortunate that he feels this way. Many fabricators view this as the most tedious
part as in “the devil is in the details” and try to minimize this most
important step to creating a truly spectacular piece of mechanical art.
The first drawing shows
the round disk seen in the photo above. This is a sub frame located behind
the moon rotation gearing in the center of the dial. It will largely be
hidden. Due to other component locations the three legs could not be
perfectly made equidistant. The next drawing shows the proposed skeleton
design for the shutter sector gears. We retain the nautilus design employed
in the snails within the strike work.
The five scans above
show the various orientation options for the sector gears controlling the
sun horizon shutters. The last example was chosen. The last illustration is
a drawing of the rear sub frame for the year drive which powers the sun
complication; the horizon shutters as well as the rotating sun indicator.
The main rear frame
design, left, shows many similar characteristics to that used for the
perpetual calendar, right illustration, which was drawn exactly two years
ago. This is another demonstration of Buchanan’s design consistency
throughout the project. This makes the entire machine a harmonious whole
despite its complexity and diverse subassemblies. There are fewer jeweled
pivots in this drawing because of the unique frame design. This is the first
assembly that largely does away with a conventional plate and spacer frame
where the majority of wheels are suspended between plates. Here there will
be one single plate that will have the rotating platform output system
suspended from the front and the slant wheel variable differentials from the
rear. The jewelling is spread throughout those systems rather than between
two conventional plates.
The drawing of the rear
sub frame for the year drive is placed over a brass blank. Next a
second sub frame is cut.
After all of the holes
are drilled into the plate, the hand drawn design is attached to the brass
blank, and then the design is scribed into the surface with the razor knife.
the blank is cut into a disk and the hand fretting process begins on
the jeweler’s saw.
The process continues
as more metal is cut away to reveal the curvilinear ivy design.
This photo shows
the completed main module frame as well as the year drive sub frame.
The first photo shows a
set of eight filing buttons used to guide the hand file while shaping curved
surfaces. These were all custom-made for this part of the project. Buchanan
could have used a standard set, but customizing a variety gives each
component a unique, handmade look. Compare how open the look of this dial
has become compared to the same view before the main plate was skeletonized.
The sub frame now has a
geared center hub. Next the frame hub is inserted into the central hole of
the moon drive.
A second outer drive
ring for the sun is added over the sub frame.
The first photo shows a
completed horizon shutter sector gear using the nautilus design. Next that
sector gear is mounted to the drive. There will be a pair of these for the
sun horizon shutters.
A side view of the
completed sun, moon and horizon drives. Notice the two shutter sector
control gears at the bottom of the assembly in the first photo. Next is the
This video shows the drive assembly. The actual dial
hands and other numerical readouts have yet to be added. The sun hand will
be attached to the center arbor.
The main module frame
now supports the sun, moon; horizon shutter drive mounted to the front and
from the rear the Janvier variable differentials controlling the moon's
orbital speed. Everything is attached to this single frame; there is no
conventional pair of frames separated by spacers where the wheel work is
suspended. This the only module within the machine designed this way.
Buchanan now turns to the development of the
year drive for the sun mechanism.
The input to the wheel
train is one revolution per day with the output one per year.
The year drive sub
frame with all wheels completed is now mounted to the main module frame.
Next is a side view.
The first photo shows
the quick release catch arm being fabricated. Shown is the clamp ring being
split with the slitting saw. Next the flat stock arm is cut out to a
decorative ivy design, red arrow.
The sun horizon shutter
cam followers are now installed. The jeweled cam rollers
2 are attached to the sector
gears 1’ and
2’. The shutter sector gear
3 is attached to one of a pair
of horizon shutters and meshes with cam follower sector gear
1’.The second shutter sector gear is hidden behind it and meshes with
The jeweled horizon
shutter cam follower wheels are seen within the circle in the first photo.
Notice how these are near the perimeter of the outer dial bezel. When the
cams are fitted they will extend beyond the dial bezel. This same extension
of the wheel work beyond the dial was replicated with the time train dual
remontoire, the equation differential work and the perpetual calendar
overdrive safety clutch. This technique adds to visual impact by emphasizing
complexity through the spilling of wheel works beyond the dial perimeter. Next
one can see the tight space between the sweep of the moon and the tellurian
counterweight. The weight had to be contoured
to accommodate a change in the position of the moon sphere.
In the first photo one
sees the moon positioned too closely to the inner bezel. The redesign of the
counterweight allowed the arbor upon which the moon is attached to be
lengthened; positioning it correctly at an equidistant position between the
inner and outer bezels.
Buchanan now begins the fabrication of the manual setting controls for
the pair of variable differentials.
The first two photos show the machine setup for
scribing the inner wheel hub. The scribe is spring-loaded and the wheel is
turned to create the circle. The thickness of the rim is determined by the
load that wheel is expected to carry and to a lesser extent the diameter of
the wheel. A very tiny wheel will need a thicker rim because of machining
The thickness of the rim is determined by taking
the distance from the tip of the wheel tooth to the root and using a
multiple for the rim thickness. The rim thickness is the distance from the
tip of the wheel tooth to the base of the rim. For lightly loaded wheels the
multiple can be as low as 1.1:1. This wheel is lightly loaded but very small
so it is closer to 1.2:1. Most wheels are about 1.2:1 and heavy loaded
wheels can be up to 2:1. The maximum used in this project are found in the
main weight drive
wheels at 2.5:1, made in September 2007.
The next four show the
spoking out of that very small wheel.
In the first photo drilling the pivot bearing in the large boss,
1, was a little tricky as it is
behind the outboard bearing, 2,
and a long drill would wander off center. In the next photo Buchanan made a
guide bush that fitted into the wheel and a special drill bit. The red arrow shows
spade bit. He held the drill in the lathe and kept the wheel in
the correct depth and then drilled the pivot hole into the boss.
The long drill is now
replaced and the bevel wheel fitted with a proper decoratively machined
stainless steel arbor. Notice the use of what I call a compound gear. The
inner tooth ring is a proper bevel design and its surface is machined at a
900 angle to the normal teeth around the circumference of the
wheel. This allows power to change direction and drive a subsequent set of
wheels using only one wheel. We first encountered this in
October of 2013 in the two speed transmission
for the orrery demonstration drive. There two wheels were mated. Then in
October of 2014 when making the drives to the
equation and sidereal time a single wheel was used but there the bevel was
located halfway between the center of the wheel and its outer edge.
The variable differentials are now equipped with
their manual setting control topped with a knurl knob.
These photos show the
differential pair installed within the sun, moon rise-set module. The pair
of setting knobs is visible, red arrows.
The first photo
highlights a pair of jeweled cocks where one is twisted by the same degree
as the slant wheel differential. The two together offer an interesting
mechanical conversation. Next one can see the numerous knurl knobs (nine red
arrows in this photo) representing a subset of the many manual adjustments
that the operator can access within the machine.
The first photo shows
an initial mockup for the readout dial associated with the manual
adjustments for the variable differentials. Each differential will have a
dial delineated with the period governing that anomaly.The next photo shows the artwork design for what will eventually be a
pair of enamel sector dials depicting the name of each anomaly and attached
to the counterweight of each differential set.
This photo shows how
the left and right sectors of the clock have come into a nice balance.
The right side for some time looked to be a bit sparse with only the bell
set, but with the addition of the variable differential set the right side
has filled out well. The sun-moon, rise/ set dial on the right has a
decorative outer bezel to balance with that on the perpetual calendar to the
left. That bezel is purely decorative and is the only bezel in the clock
that serves that purpose.
This rear elevation
clearly shows the gap nestled in the middle between the two upper dials.
Here is where the machine will be crowned with the orrery.
This and the next photo
further show the balance between the two sides of the clock. In this view
one can see on the left side a density of machine work just below the
horizontal, silver main upper frame pillar behind the perpetual calendar
dial. Just above that it is empty. The variable differential work fills out
that same empty space on the other side balancing out the less dense
appearing bell set just below where the left side shows density.