Complete tellurion assembly. Complete sidereal dual dials - February 2016

This month we finish the tellurion and complete the dual sidereal time chapter ring drives.

In the first photo one can see a thin dished washer inserted between the wheel and a screw-down collet. This acts as a slip clutch for the the seasons indication. One of scores of clutches used throughout the machine for safety as well as setting functions for various components and indications.  

The last part needed to complete the tellurion is the quick-release lever which will allow the operator to easily remove and reinstate the entire assembly within the rest of the machine. We use the same design philosophy for the calendar complication. The modularity aids both in servicing as well as safety in the transportation of the clock when it becomes necessary. The first photo shows a paper inserted into the area where the lever will be located. Here areas where the lever cannot go are outlined, thereby giving Buchanan an idea of the shape and scale of the lever that will be allowed within this area. As the overall movement becomes more and more crowded, this exercise becomes ever more critical and difficult. Next the areas are transferred to the brass blank and last a rough outline for the lever is drawn.

The first photo shows parts for the spring loaded locking pin. The lever is tested for fit on the tellurian test stand.

The completed locking knob and pin as well as it mounted to the lever are seen in the first two photos. The steel arbor is where the tellurian assembly is mounted. Next the quick release lever is in place on the strike train frame pillar, circled area. 

Now Buchanan takes the entire tellurian apart to complete the jewelling of the pivots and the semi final finishing of the frames. This process will be repeated at least once again in the final frame finish, screw bluing and polishing stages.

After all the pivot holes are planted the jewel holes are prepared. First Buchanan centers on the pivot hole with a microscope, then drills the hole out undersize, then bores out the hole with a boring head as seen in the first photo, then reams the hole to size in a staking tool. Repeat 29 times for all jewels. Then the frames are finish filed. Everything loses a little more weight. With so many components, the slimmer each part can be, from a visual perspective, the better.

The first photo shows the 13 pillars used for the tellurion triple frame assembly. Next the two main frames ready for jewelling.

The nineteen jewels are now in place within the four frames and one cock. Next the triple frame design is readily seen from the side view.

The organic ivy design found throughout the movement is continued here within this subassembly.

The bare frame shown in the prior photo is now filled out with the compliment of components for the entire complication. The celestial bodies are still mock ups and will later be replaced by Mammoth ivory and semi-precious stones.

The tellurion is now complete and installed within the context of the rest of the movement. Is all this this looking a bit complex yet? Total parts count on this complication comes to 395.

Each turn of the demonstration crank equals one day. Total turns are ten. One can just see the movement of the tellurion armature counterclockwise as well as the rotation of the Earth and Moon's orbit. Although it is not clear in this clip, the calendar is also advancing at the same time.

We now move on to complete the sidereal time dual rotating chapter ring drives 

We now turn to the completion of the sidereal time indication. Here I will recap the process by which we created this function. The original design envisioned in July 2006 was to have a separate, small dial for this. It would have been delineated in the conventional twenty four hour format. In September of 2013 I attended a symposium where a Thomas Tompion clock was displayed that showed sidereal time on a stationary dial with mean solar time as a clockwise rotating chapter ring concentric around the stationary sidereal dial ring. This allowed one to read both times simultaneously with only subtracting twelve hours from the sidereal time reading after midyear of June 30 to convert to the twenty four hour format since the sidereal dial was delineated in a twelve hour format. This is necessary as the hour and minute hands are geared to rotate in a conventional twelve hour format. Since we had already created the dial and mechanism for the mean solar time, and this is a more important function in our design than sidereal time, we chose to have the rotating dial for the sidereal time which would be rotating in a counterclockwise direction. We had assumed that we would be able to use both the hour minute and second hands to read off both dials simultaneously and created the inner rotating dial and all seemed well. However, after studying this I came to the realization that this arrangement would only work with the hour hand, not the minute or seconds. To enable a more accurate reading there must be a separate counter rotating rings for the minutes and seconds.

So a decision had to be made. Do we go with the arrangement as made allowing a reading to the nearest five minutes or so using just the hour hand, or do we go with additional rotating rings or dial hands geared to sidereal time? It immediately became apparent that three additional hands would be prohibitively expensive. We would have to create an additional three cannon pinions within what already was a set of four nested cannon pinions (hour, minute, second and equation). Since this was not anticipated when these were made, the seconds would be the diameter of medical hypodermic needle. Not only that, but the dial would have been cluttered with seven hands. However, one additional ring for the minutes was practical. An additional seconds ring was ruled out since it would have made the entire dial structure with four nested chapter rings too thick; nearly obliterating the open center section and throwing the entire sense of visual symmetry off balance between the left and right main dials. The following is Buchanan's beautiful solution to the problem, a classic case of ‘making lemonade out of lemons’.

Buchanan’s solution was to create a rotating cage driven by the current sidereal hour and within this cage are three rollers which support a central hub. That hub is driven separately to display sidereal minutes.

The parts begin to be made. Notice the number 240 and 256 in red on each toothed disc. These are the number of teeth cut into each rim. Next the rollers and roller cage begin to be fabricated.

These photos show how the cage rollers are mounted. Each of the three roller’s arbors are attached at one end with a countersunk foot that is then secured by a screw, red arrow.

The carriage is beginning to take shape. Notice the jewelling on the rollers supporting the center hub.

Now begins the spoking out process. The first photo shows the jig Buchanan uses to accurately scribe out spokes. The center steel rule has an adjustment screw on both ends so as to accurately have the edge perfectly bisect the center pin. The pins arranged around the outer perimeter allow Buchanan to select any number of spoke combinations from three to six and more. Next is a large and amazingly delicate 256 toothed wheel showing a completely missing center section. This will later hold the rear roller cage frame.

Now the artistic aspect begins; the fun part! Buchanan's talents are put to work making beautiful the center rotating cage controlling the sidereal hour chapter ring. The first design was an attempt to make the cage pattern look similar to that used to support the anti-friction wheels in the pendulum assemblies. The geometry between them was too different to make this visually acceptable. The next three designs became progressively more elaborate, and guess what? We chose the most elaborate. Is this a surprise?

The final design is shown within the context of the wheel that will support it as its center; the same wheel as was shown in Buchanan's hand six photos prior. Next that design is transferred onto the cage blank and later will be scribed onto the surface of the metal.

The first photo shows one of the cage plates being milled to shape. The next two photos show the rear and front cage plates being decoratively cut out on the jeweler’s fret saw.

The decorative curvilinear shapes now have to be hand filed to final profile after being cut out from the brass blank on the fret saw. Buchanan makes extensive use of a fleet of files, first photo. An array of filing buttons is needed to help shape each curve. In the second photo a steel cylinder is affixed to each side of a desired, rounded area on the rear roller cage frame, and front frame in the third photo. Buchanan will file around this cylinder to achieve a near perfectly round profile for each area that requires this particular shape. In this case it is the three pivot points for the cage rollers. I stress that the contours are near perfect with the slight imperfections that are always present with hand work. These variances are highly desirable as they lend to the movement being recognizable as being made by a master craftsman and not a computer-aided machine. To my eye one cannot ever be confused with the other.

Next the completed decorative front cage frame and its setting within the center of the wheel for the hour chapter ring. It is this type of work that sets Buchanan apart from the rest.

There are 68 parts that compose the sidereal roller cage and twin rotating, concentric chapter ring platters. There are additional wheels that derive the sidereal minute from the original sidereal hour.

The following six photos show the build out of the roller cage driving the sidereal hours and center hub for the sidereal minutes.

First the cage drive gear has the rear decorative cage plate assembly mounted to it. Then the part is flipped over and the three cage rollers are installed. Next the center hub is inserted.  

Next the upper decorative cage plate is installed and next a three quarter view. Finally the part is again flipped over to the rear view and the center hub’s drive wheel is attached.

The following photos show associated gearing and platters which attach to the roller cage and center hub assembly.

The large wheel shown here is driven by the decorative front roller cage mounted to its center. Notice that this wheel has a smooth rim. Upon this wheel is mounted the sidereal hour chapter ring support platter. Next the rear decorative roller cage is attached to the slightly smaller, 256 toothed, wheel.

Next the same perspective from the rear is kept and the roller cage assembly is installed. The sidereal hour and minute drive wheels are seen in the center foreground. The part in the prior photo is seen just below the smooth, larger wheel described above. Next a larger wheel which had existed before in connection with the mean solar time is added along with its drive wheel. Next the assembly is flipped over to the front view. The inner minute platter is attached to the center hub.

The hour chapter ring platter is attached to the outer smooth wheel in the prior photo. Next the entire assembly is flipped over to reveal the attachment points of the two rotating platters. Next a close up of roller cage and center hub.

The cage assembly is now installed within the machine. Next is the 240 toothed wheel and it is used to derive the sidereal minutes from the sidereal hour. This is the opposite of conventional motion work where the hours are derived from the minute gearing. Conventional engineering would expect this where the faster rotating gear with more torque (minutes) would drive the slower wheel. However the sidereal hour was already constructed, and fortunately there was plenty of power available to allow for this reversal.

The first photo shows the careful planting of the derivative minute wheel. Next the wheel installed, rear view.

The sidereal hour and minute counterclockwise rotating platters are shown. Next dressed up with dials. Only the mean solar dial is real enamel. The sidereal chapter rings are mocked up in paper and still have to be made in fired enamel. All of the dial bezels are also wooden mockups, yet to be fabricated in a decorative metal design.

This video more clearly shows the operation of the roller cage supporting the sidereal minute and hour time dials. These should be rotating counterclockwise (retrograde) but the operator was turning the demonstration crank in reverse. The video was made in October of 2021 after the dials, bezels and hands were competed; the entire project was finished in November 2021

I remember at the outset of this project that I commented to Buchanan how I was often disappointed with the side view of skeleton clocks. One usually saw just two plates edge-on with a few skinny looking arbors between them, a rather uninteresting sight. I think we have managed to avoid that problem here!

This is where we stand at this time. The left side of the clock’s mechanicals is complete. The right hand side has its main and lower subsidiary dial complications complete. We still need to create the functions for the small separate dial, a thermometer, as well as the upper subsidiary dial above the tellurian, the sun and moon rise and set functions. And finally the central area is to be crowned with an orrery. Afterward is the fabrication of the stand and movement crowned with an orrery.