It was January 1981 and we were living in Centerville, Ohio while I was attending the NCR Technical Center in Miamisburg. Our stay would be for six months and if you are familiar with Ohio weather, much of it inside during the winter months. It was during this time that I built this rolling ball clock in the many hours of “”spare” time on my hands.
I had seen a plastic Rolling Ball Clock a few years prior at a gift shop and the principle inspired me to expound on the concept. I knew it would have to be made of metal, strike the hour and be interesting to watch. This kind of kinetic art should be entertaining as well.
Having built a ball clock ten years before and soldered it together with tin/lead just to watch it slowly fall apart from joints fatigued by vibration, I knew it had to be built with bronze rod and silver solder joints to last. Lessons learned.
The first thing to figure out was how to raise the balls in a way that I could control the speed since a ball needed to be released to the track each minute. I found a synchronous motor at a surplus store (Mendelson’s) in Dayton that turned 3 revolutions per minute; this was just slow enough to work with. The motor is the kind used in time clocks and runs at a very accurate speed because it keys in on the 60-cycle line power we have in our homes. The motor is mounted midway between two larger wheels and has a smaller wheel, which transitions the ball between the two. The large wheels have only six pegs so that with three revolutions of the motor they move a half revolution. The wheel attached directly to the motor shaft has one long eccentric shaped arm that works with the pegs on both large wheels. The shape of the arm is designed to minimize the drag on the motor and provide a smooth movement to the larger wheels. All three wheels have a “U” shaped extension to cup and carry one ball through a half cycle. Balls are picked up from the lower track by the large lower wheel, carried to an upper track at the top of that wheel released in time to be picked up by the intermediate wheel and deposited on a track at the bottom of the upper large wheel. The ball is then picked up by the upper large wheel and carried to the top of the clock where it is released onto the top track. There are always three balls in transition on their way to the top of the clock in a slow progression. The lower wheel also operates an arm that separates and releases one ball at a time from the lower track to be picked up by the wheel. This keeps extra strain off of the motor and a consistent pressure to lift the balls. Extra strain could slow the motor, change the timing and cause unneeded wear on the motor. The motor has been running now for 29 years so it must be working.
The time racking follows the same principles and most rolling ball clocks. There are three racks, one for the minutes, tens of minutes and hours. The current time can be determined by counting the balls in each rack. So for say 3:45 AM or PM you would see three balls in the lowest “Hour” rack, four balls in the middle “Tens” rack and five balls in the “Minutes” rack. Unlike the plastic version, this clock does not have numbers on the racks to tell you the number of balls in each. You just have to count them yourself if you want to know the time. Old school.
The balls in the time racks progressive accumulate until they reach a tipping point on each rack much like a teeter-totter. The last ball that tips the rack is kept aside from the rest and when the rack tips, it follows another track to add to the rack below. The rest of the balls follow an exit path to the bottom to be picked up once again by the lift wheels. The “Big Dump” occurs at 1:00 when the racks are full with 9 balls in the minutes rack, 5 balls in the tens rack and 12 balls in the hour rack. The next ball tips the minutes rack which sends a ball to the tens rack causing it to tip. When the tens rack tips, it sends a ball to the hour rack causing it to tip. There is a cacophony of sounds as the all the balls find their way to the bottom of the clock. The hour’s rack is unique to the others as there is one ball that is always retained in the rack so that the time indication for 1:00 will be shown with one ball. The captive ball at the right side of each rack is used to change the center of gravity for the rack to allow all of the balls to exit the rack before tipping back. Without this only some of the balls would dump leaving others still on the rack.
So in it’s basic form, this is the typical rolling ball clock. The difference here is that this clock also strikes the full hour. That became the real challenge of the project. First, how to track the hours and second, how to deal with all the balls as they are used at various hours of the day. There ended up being 63 balls required supplying the needs of the clock and that was too many for the bottom rail to support at one time. It was necessary to create two areas on the track to hold balls until they were needed. The reservoirs so to speak were filled when the 1 o’clock ball tipped the hour rack and the accumulator section of the striker mechanism emptied ending up with 10 balls in one and 12 in the other. Sensing levers down by the lower wheel feel for balls and when they are not there, more are released from above.
When the clock reaches the full hour besides adding one ball to the hour rack, the other 5 balls from the tens rack roll into a segment that allows only one ball at a time to pass, until they have all gone through. The balls roll down to another rack that tips when the fifth ball reaches it. The fifth ball then goes into the striker accumulator section to add one hour. The four remaining balls will exit and provide power to the selector for the hour. The four balls move a selector arm that is stopped by the balls in the accumulator section and as they continue to travel they cause a wedge to be driven between the same number of balls on the front rack matching the hour. The wedge allows only those balls to be released at the end of the cycle. At the end of the selector arm travel it contacts a small lever that releases the correct number of balls into a lower track feeding the striker segment. The striker segment only allows one ball at a time to go through. The weight of the ball causes the segment to rock and the hammer at the other end of the segment falls on the old cash register bell when the ball is released. This continues until all of the balls have passed through the striker segment.
Ten minutes later when the minutes rack tips, the nine balls while finding their way back to the lower rack will be directed to reset the hour selector. Ten minutes later the nine balls from the 20-minute dump will be directed to refill the front rack of the striker mechanism. Every time after that the dumped balls will keep the striker rack full and any excess will roll to the bottom to be picked up again.
Like with anything of this nature, there were many things to consider. Wear and tear, harmonics of the falling balls causing vibrations, serviceability. The clock has been running for 29 years now and shows no signs of stopping. It does require cleaning and oiling periodically since it is out in the open and can pick up dust. It can also be completely disassembled and fixed is necessary.
Here is a video of the clock on Youtube that Michael helped me make: Rolling Ball Clock by Jon L. Long Sr.
So there you have it. Feel free to ask questions about how things are done. Will I build another? I hope so, but it will not look like this one.
Jon Sr.
Hello Mr. Long,
I know it has been a few years since this was posted, but I do have a question. Right now I am in the process of designing a large clock for the new student center at my university (LeTourneau University in Longview, TX). I have been looking at all sorts of different clocks, but this one really caught my eye. I showed it to my teammates, and we are really interested in making a similar one (a small-scale model of what it would be full size). The only problem is that none of us have the experience or knowledge necessary to create something as intricate as this. My question is simply if you would have any suggestions for our clock design, some tips for how to go about building it, and/or recommended materials for such a project?
Lancellot,
Well, it sounds like you have an ambitious project, but one that could be fun as well. I think that may be the key, to keep it fun. In that way creativity will spring forward and a design will follow.
My clock was built as a project to occupy many hours inside while in-climate weather ruled the days. There isn’t much to do outside when the weather is 6 degrees.
Some of the principles I applied in my design were from my experience with mechanical cash registers of the 50s & 60s. Little of which we see today as electronics has now dominated the task.
Whatever design you choose, be sure that you use quality materials and if you weld parts together, that the joins will withstand vibration.
As far as the design is concerned, it should be fun to watch, especially in a student center. Adding elements that show motion helps create interest.
If your goal is to have the piece keep time, then it is essential that the mechanism to either release balls into the matrix or the lift be precisely controlled. So, you could have a lift mechanism that returned balls to the top reservoir run at one speed, while releasing the balls could be controlled at the proper intervals. The intervals would be determined by your design.
The first ball clock I made struck the quarter hours as well as the full hour. Something you may want to consider.
I have chosen to use Brazing Rod which you can get at a welding supply store. I silver solder these with a jewelers torch. If your design is large then the 36″ rods may be to short and you will have to source longer stock from a metal supply.
I wish to the best on your project and remember to keep it fun.
Jon
Jon,
What a great clock. I am in the process of building one myself out of brass wire. I was hoping you would explain to me how you keep the 1 ball to the side and route it off to the 10 or hours.
I have studied several clocks on the internet but most of them are wood design.
Thank you
Thomas
Thomas,
I have replied directly via email.
Thanks, Jon
Hello, Mr. long. Really like Your Watch. Is there a detailed Assembly instructions. Thank you.
Yuriy,
There was no plan as it was built as it progressed. It was not a kit or anything like that. Happy that you like it.
Hi John,
I really like your clock. Fantastic design. I have a question, ie: I don’t understand how the mechanism selects the correct number of balls, say at 6 o’clock it lets 6 balls through to feed the hour striking mechanism. Naturally at 7 o’clock it passes only 7 balls to the striking mechanism, etc.
Looking forward to your answer,
Best regards,
Harry
Harry,
Sorry, I don’t check comments very often as you may have noticed. What happens with the selector is every hour a ball ends up in the back rack for the selector. On the hour the sixth ball is routed to the hour rack. The remaining 5 balls after a slight delay go into a lever rack that separates another ball a routes it to the back rack of the selector mechanism. The remaining 4 balls are used to power the selector and cause it to sweep down the curved rack. When it is stopped by the balls on the back, a blade cams up and separates the same number of balls on the front. As it continues down it releases the balls that were selected into the hammer. It all happens pretty fast. Hope that helps.