Mulligan Bespoke Clock No. 5 (2008)

After reading the article Dual Oscillators in the Horological Journal, I thought about the idea of making a genuinely unique clock. 

My experience in horology started in the early 1980s when I started repairing clocks as a hobby. After finding myself unemployed, I took the step to repair clocks as a full-time occupation. I gradually built up a repair business, and antique clock and watch sales shop with hundreds of clocks and watches in the CBD of Launceston, Tasmania.  

Please forgive me for my lack of technical input in this article as I build clocks from the heart, not the head. Back to the clock in question, after reading the two-part article on dual oscillators, I was fascinated by the whole concept of it. It had been two years since I had built my last clock, so it was about time that I started on another project. I searched through the cupboards of my workplace for a starting point and found an 1850s Fusee movement with a long-drop pendulum. It would suit my idea perfectly; all I needed was the wheelwork. I had not advanced yet to cutting wheels, although I am more than capable. It was more timesaving to use what I had accumulated over the years. Ideas in my head for different and interesting clocks to build might see me cutting my own wheelwork in the future. 

My first step in creating a timepiece is to sit down with pen and paper and draw what is in my head. When I was happy with the design, I drew it to scale, purchased all the brass I needed, and made a start. Initially, I planned to put the clock onto a French marble clock base but then decided against it because it may look like I had made it out of an old French clock. I wanted more of myself in the clock and less of a replica of someone else's design. I then visited the local stonemasons, where I purchased a solid polished block of black Tasmanian granite. I drilled four holes for my columns with an altered masonry drill bit and plenty of water. Two hours later, I'd finished the holes. The first part of building the actual clock was constructing two identical pendulums and a suspension block to hold them. I wanted to do the pendulums first to experiment on how they worked when one was impulsed.

 

When I had all that built, I fixed it in the bench vice and pushed it gently with my finger; it seemed to do what I wanted. Then, I packed it away and started to build the clock movement. In the past, I had drawn the plate design on paper, then glued it onto the brass sheet, gang drilled around the plates, and cut it out with the fret saw—this time, I did things differently. I took the design and brass sheets to a local engineering firm, where they loaded my design into the computer and laser-cut the two clock parts out. At the cost of $120.00, it saved me a lot of time and hard work. Unfortunately, the firm cannot do it anymore, which could mean back to the dark ages with my next clock. This part of the clock was relatively straightforward - layout the gear work in the new plate design. All went well; I had a new movement up and running in no time. Now came the exciting part of getting it running by fitting the suspension block and the two pendulums. I had not made the base at this stage, so I clamped the movement to the bench and hung the pendulums over the bench edge. I gave each pendulum a gentle push in opposite directions, and away it went for about thirty seconds, then came to a slow stop. I checked the beat; that seemed fine. I removed the back pendulum and set it going. 

 

Leaving it for a few hours, it had an excellent strong tick, and everything appeared to be working, so I added the second pendulum and set it going again. Sure enough, about thirty seconds later, it stopped! The suspension block I had made up was two halves of brass at the top bolted together, and two halves at the bottom bolted together, joined by a solid strip of suspension spring. The pendulums hooked onto the same block at the bottom. What I was trying to achieve was for the movement to impulse one pendulum, and in turn, then impulse the other through vibrations. Alas, it would not run, so I decided to try a swivel link in my pendulum hooks, achieved by using a hardened pin and balance wheel cones. That kept it running for a little longer, but it still stopped. 

 

The next step was the actual suspension spring. I punched a series of holes along it to give it a little more flex; this worked for a while but then stopped. I then started to thin the spring in the middle. All these variations seemed to help but still no luck. I attempted to cut the whole centre out of the spring and leaving two strips at either end like a traditional suspension spring. It seemed to improve, but again it stopped. I experimented with different thicknesses, lengths, and styles over the next month, to no avail. It was becoming frustrating. It almost came to the point of deciding to cut the second pendulum off and have a simple Fusee skeleton clock. I decided to try one more combination to cut the bottom block of the suspension spring into two pieces and link them with another piece of spring steel so that the suspension spring had the freedom to move on two different axes. I set it all up, attached the two pendulums, set them in motion, and to my amazement, they kept going. The clock ran for the rest of the week. I had solved my problem and could now make the clock I had wanted. Not a lot has been written about twin pendulum clock making, not that I could find, so it was all coming from trial and error. Please note that I was not building the clock because it would be superior in timekeeping or efficiency. Unlike all twin pendulum clocks produced in the past, I sought the novelty of twin pendulums off a single impulse. Twin pendulum clocks in the past used either two movements side by side to impulse each pendulum or one movement with twin escape wheels impulsing each pendulum.  

Now I had the clock running; the movement had to be lifted high enough for the pendulums to swing. To achieve this, I used four columns from an old broken French marble clock case and four solid brass blocks to make up the difference. Now it was starting to take shape and look like a clock. For the next step, I screwed a piece of brass onto the faceplate of the old trusty Myford ML7. I then indexed to mark out the dial, engraved the chapter ring, minute markers, and all tops and bottoms of the Roman numerals in the lathe. The rest was done with the steel ruler and scriber and finished with a dental burr in a Dremel tool, black waxed and silvered.

 

With the clock just about finished, it was time to work on a case. I was a bit unsure of what I wanted. I thought about the idea of a wood and glass case and again wished to make it from scratch. After much thought, I decided to make it out of brass and incorporated the clock's base as part of the clock. That meant discarding the granite block and getting a bigger one that had a step in it so the case could sit on it. So here we go again, another four holes into the solid granite base, thinking about all those wasted hours on the last base. The top of the case was solid brass with square brass sides, and a solid brass bottom with the centre milled out to go over the granite base and sit on the shoulder. Next, slots two millimetres in size were milled into the four corner pieces for the glass. It was starting to look good with that all finished, but I discovered the clock's height created a problem. When it sat on a mantel, it was just about impossible to lift the glass top to wind it up. To fix that, I drilled a hole in the glass, made up a brass grommet with a swivel cover, and made two more grommets on either side of the case. Now I could push each pendulum in opposite directions to start the clock.

 

Next came the part I regard as the worst job - polishing it all. I was not too fussy with the polishing part, not because of laziness, but because I wanted the clock to have an older look, not looking freshly made. I purposely left minor dings and scratches in all the brass work. I was delighted with the finished product, despite it not being an easy clock to regulate.

© Graham Mulligan, updated 2022