When I decide to design a clock I always have a specific objective in mind and for this one it is to see if I can activate a chime movement without having to resort to a “warning wheel” on the chime movement. This clock will not have a counting wheel but will “play” a tune each hour.
I started by counting the revs. needed to complete a cycle on the little “hurdy gurdy” music box. The count was 30 revolutions for one complete cycle. I then worked out a 30:1 gear reduction so that I could use a detent pin on the main wheel to use as a start and stop mechanism.
Drive wheel has 48 teeth driving an 8 leaf pinion fixed to a 30 tooth wheel driving a 5 leaf pinion for a total of 30:1.
The first attempt was a straight gear train driving a lantern pinion on the final pinion. The lantern pinion did not give a smooth drive and I decided to change to a conventional pinion with involute tooth design.
Even with the conventional pinion the little music box needed more energy at some parts of the cycle and less at other parts. A fly wheel was added to the mix in order to balance the load a little better.
The Fly Wheel was driven off the same wheel as music box but provides a leveling effect to the works.
At this time I am driving the second wheel which will change in the future. I wanted to get power to the music box as directly as possible in order to see how much variation there was in the power requirements.
Driving the second wheel requires very little power but I will never get 24 cycles, or one complete day, that I desire by driving from this wheel
A close up of the music box. By fixing it securely to the frame-works the sound resonates very nicely.
I may need the combination of both flywheel and a fan to regulate the works. At present it is still somewhat jerky. Now that I know that the gear train works its on to the next hurdle.
I have installed a little actuator lever that will be actuated by the time train. The level is weighted on the right side so that as soon as the pin frees the detent it returns to a position ready to catch the pin after one revolution
The flywheel had to be enlarged to make up for the varying load put on by the music box. The enlarged flywheel diameter was too large to be placed on the front so he arbor was extended through the plate and the flywheel placed on the back.
I have switched from my “normal” cycloid gear form to this involute form since the music box train uses the backlash for a more even power delivery. The gears mesh very nicely and I am going to use the form throughout the gears on this clock.
The “bench” train shows the music box drive on the front and the flywheel on the back of the plate. The wheels and pinions have been recut to involute form and they seem to be a little better in terms of handling the variable load required.
The music box drive train is very, very simple and does exactly what it is supposed to. It remains to be seen how easily the time train will trigger the release of the music train.
I used a little sketch to help me figure out the time train. I want the escape to turn once per minute and the second wheel to turn once per hour. The main drive wheel will turn one rev. every 6 hours to minimize the revolutions of the drive.
Since I will only have 24 inches maximum weight travel and I want about a 30 hour movement or 5 revolutions that means the drive wheel drum can only be 4.8 inches in circumference or 1.5” in diameter (radius = 0.76”).
In order to have the face in the center of the case and have a really simple gear layout I am going to try and place both weights on the right side. (This idea changed later on as I wanted to make the pulleys simpler and reduce the width of the clock
The pendulum for this clock will be a 1/2 second unit and in order to have the escape wheel turn once per minute I need 60 teeth on the wheel. The pallets are graham dead beat design and give a 2º lift.
The time train will have 3 - 60 tooth wheels driving 3 - 10 leaf pinions. Theoretically the involute tooth design will give me a non-sliding contact at all the wheel teeth and pinion leaves. That’s the theory - we’ll see how it works!
Only one gear left to cut in the running train. The first two sets of pinions and wheels run exceptionally well. I am very pleased with the depthing and meshing of the involute tooth form. I just hope it proves as good when installed between plates.
This shows a sort of X-ray of what I’m trying to do. The third wheel is the minute arbor and the escape wheel with its 60 teeth will provide a second hand that moves on the 1/2 second. A small window in the scene on the lower section of the door will show the pendulum
As I progress with this clock I would like to make it narrower. If I maintain the proportions of the “model” and still have a clock that is 24” tall then it is tight to put the weights on the side. I am allowing the gears to overlap to reduce the width.
The levers to actuate the music box have been sketched in on the plates. The actual construction will be done once both trains can simulate the action
At this point the two plates have been glued together to ensure alignment of the finished arbors. I use my depthing tool to get the arbor spacing and then without changing the tool I tap the end of the sharpened depthing tool arbors to correctly place the arbor holes.
The time train is in the lower part of the picture and while the music box train is above.
The trains are placed on the plate and “strings” have been placed to show the route they will take to the weights that will be behind the plates of the trains.
By allowing the trains to over lap I will reduce the width of the finished clock by about 2 inches.
About this time - we changed residences and while I still have "most" of the parts, this clock has not been finished.