Eclipse - an Electromagnetically Impulsed Wooden Gear Clock - DXF Files Included

by Sawdust Willy in Workshop > Laser Cutting

2163 Views, 27 Favorites, 0 Comments

Eclipse - an Electromagnetically Impulsed Wooden Gear Clock - DXF Files Included

IMG_1865.jpeg
May 18, 2024
IMG_1880.jpeg

The pendulum clock was invented on December 25, 1656 by a fellow named Christiaan Huygens and served as the most accurate type of clock for 270 years until it was replaced by the quartz clock. During this time lots of improvements were made to the design to increase accuracy involving pendulum amplitude, escapement design, temperature compensation, etc. Well, we're chucking all that stuff out of the window today in lieu of simply having fun! Yes, we're designing (and building) a pendulum clock but we are violating a couple of really big fundamental rules. First, a pendulum is most accurate as a clock regulator when you keep the amplitude (how much the pendulum swings back and forth) very low... around 5 degrees. And second, you want the pendulum to swing as freely as possible. Meaning that you want it as isolated as you can get it from the clock's drive mechanism. Ha... in our design the pendulum swings through nearly 75 degrees and not only do we not try to isolate the pendulum from the drive mechanism... it IS the drive mechanism! So what we'll end up with won't be the most accurate clock by any stretch of the imagination... but it will be good enough (check out the last pic... I timed 60 beats!), a LOT of fun, and interesting to watch. Take a look at the attached video of the finished clock and you'll see! It's kind of mesmerizing. Ha, I love the tick-tocking but I was surprised by how really loud it is. As an experiment I put little stick on felt pads on the pawls and this completely deadened the sound. So if the loud tick-tocking is an annoyance to you then use some felt!

I've been working on this design off and on for quite a long time (with a lot of emphasis on the off times). But with the recent addition of a laser cutter I was finally able to complete the project. I did publish an Instructable back in 2020 on how to build the most crucial part of the design which is the electromagnetically impulsed pendulum. Before you launch off into your own clock design/build I would HIGHLY recommend completing that Instructable first. You need a functioning coil and pendulum otherwise you're kind of wasting your time. That being said... there still is some good information here on gear design and basic clock design... so maybe not a total waste of time.

My emphasis in this Instructable will be mostly on my thought process as I worked through the design. The intent will be to provide you with enough information that you can design and build your own clock. My thinking here is that once you understand the "why" then you are free as a bird to be creative and do your own thing. You're not stuck duplicating what somebody else has done. Though I get it... not everyone is into designing stuff so... I'm going to include DXF files for many of the parts (all of the gears, main gear frame, and hands) so you can skip a lot of the design process if you want. Just know that in this Instructable... even while you have the DXF files there will still be a number of details for you to work out to complete the clock as the design currently stands.

Now that I have completed the prototype and have it sitting in front of me it is much easier to think about how I want to improve it so that it is a snap to build. I've already started working on the next revision. Once it is complete I'll publish a new Instructable on how to build that version. The emphasis there will be more on how to build the clock rather than design considerations and will include details for ALL of the parts. I'd like for everything to be laser cut. We'll see how close I can get to that. The goal will be to have it ready for Christmas 2024.

I've named the clock "Eclipse". We had an amazing eclipse on April 8 and I was lucky enough to live on the path of totality. We experienced a little over 4 minutes. Quite a memorable experience. If you notice the hands on the clock you will see that the tail of the minute hand is the moon, and on the hour hand you have the sun and a star. Every one hour and five minutes the moon passes in front of the sun for an eclipse!

And with all of that out of the way it's time to get to work...

Supplies

I'm still working my way through the wood scrap table in my garage. My Miniature Wooden Train project put a small dent in it but there's still a pretty big pile left. So all of the hardwood parts for my clock came from the scrap table. I used cherry for most of the frame and there are bits and pieces of walnut for spacers and other small parts. I'm not sure what kind of wood the base is. It came from some old shelves that were in my grandparents' basement. I also recycled the oak base that I had originally intended to use. It turned into the juction box for the wiring. The gears and interior frame section are laser cut from 1/4 inch thick Baltic birch plywood. And the hands are 1/8 inch thick Baltic birch plywood. You can get the plywood from Amazon or your local woodworking supply store (Rockler, Woodcraft, etc.). For the bushings/bearings I used 1/8 inch diameter brass rod and 5/32 diameter brass tubing... again from Amazon. I had a box full of assorted brass screws left over from the Cuckoo Clock project that hold the coil assembly in place. Gorilla Super Glue Gel bonds everything else together and for the most part all of the electrical stuff is soldered. The neodymium disc magnet in the end of the pendulum is 1/2 inch diameter and 1/4 inch thick.

Woodworking tools include table saw, miter saw, scroll saw, drill press, disc sander, random orbital sander, chisel, surface planer, and most importantly a laser cutter. The laser cutter was the key component in being able to finish this project. I had initially worked on cutting the gears by hand on the scroll saw using paper templates over plywood. OH MY GOSH! I know that lots of folks enjoy working with the scroll saw but that effort bored me to tears!!!!! It would take me hours to cut out a single gear... and if you screwed up you had to start over. Too stressful for me. The laser cutter makes short work of it... taking less than 10 minutes to cut the largest gear. I still need to tweak my gear designs to cut down on some of the finish sanding but holy cow the laser cutter beats the heck out of that scroll saw when it comes to gears.

I am a newcomer in the laser cutting world so I don't have the experience needed to recommend one unit over another. I just read a bunch of reviews and watched a lot of YouTube videos on the subject before I selected mine. It is an xTool S1 40 watt laser cutter/engraver. I like it so far. I have AutoCAD loaded on my computer and used it to draw the gears. I saved gear files to a DXF format and imported those into the laser cutter software. It was easy.

Get the Coil Built and Tested Before You Do Anything Else!!!

Once again I will refer you to my previous Instructable on building the coil and pendulum here. It is vital that you are starting with known good components prior to embarking on the clock build. Here's a summary of what is involved in wrapping the coil (from the Instructable)...

1.      Place the coil blank (assembled nail and end pieces) in the coil wrapping jig.

2.      Insert the end of the 35 AWG through the ½” oak piece and into the recessed area. Leave about a 2 inch tag end hanging out to work with during soldering.

3.      Label that end “1”.

4.      It doesn’t matter which way you wrap the coil so long as you go the same direction for both coils.

5.      Apply 4000 wraps (doesn’t have to be exact) and finish at the same end you started.

6.      Cut the wire once again leaving a couple inches to work with.

7.      Insert that end of wire into the recessed area and label it “2”.

8.      Insert the end of the second coil into the oak piece leaving a couple inches of tag end.

9.      Label this end “3”.

10.  Another 4000 wraps and finish at the place where you started. Makes sure that you are wrapping in the same direction as you did on the first coil.

11.  Cut the wire leaving a couple inches for soldering.

12.  Insert that end into the oak piece and label it “4”.

13.  Cover the wire with Kapton tape for protection.

14.  Solder as follows (use fine sandpaper to remove the insulation on the magnet wire): 1 and 4 connect to the emitter.

15.  3 goes to the base.

16.  2 goes to V- (I used a piece of wire for the V- input p in on the coil).

17.  The collector goes to V+ (again, a piece of wire was used for the input pin on the coil).

18.  Test. Apply 9 volts DC across V+ and V- and swing a magnet on a string past the coil. It should keep swinging and not stop.

19.  Pot the recessed area (I used 5 minute epoxy).

20.  Test again.

21.  DONE!

Base Unit

IMG_1709.jpeg
IMG_1711.jpeg
IMG_1710.jpeg
IMG_1723.jpeg
IMG_1719.jpeg
IMG_1725.jpeg
IMG_1727.jpeg
IMG_1728.jpeg
IMG_1729.jpeg
IMG_1730.jpeg
IMG_1731.jpeg
IMG_1739.jpeg
IMG_1736.jpeg
IMG_1737.jpeg
IMG_1740.jpeg
IMG_1734.jpeg
IMG_1743.jpeg
IMG_1741.jpeg

The base piece is made from a piece of wood that functioned as a shelf in my grandparents' basement. It had multiple coats of various colors of paint. I have no idea what it is but it is really pretty. It took a few passes on the suface planer to get down to bare wood. It is 7 x 12 x 11/16. Initially I was going to make this clock battery powered so I installed a little box to hold the batteries. I was having problems with that arrangement so I said the heck with it and used a 9 volt plug in power module. So I turned what was going to be a battery box into a junction box. The top is held in place with little magnets. Also, the bottom side of the back edge is sanded at an angle to make it rockered. Push on the back side and it pops it open. I like it but I still want a battery powered version... maybe later. So for now you plug in a module and flip a switch. Refer back to the Instructable where I originally built the coil for wiring details.

I was really happy with the coil mounting idea I came up with... that I had to totally redo because I was a knucklehead and didn't consider pendulum length... more on that later. But here the coil is installed (albiet temporarily) and I tested it with the prototype pendulum. Worked like a charm! And speaking of that... As you will probably notice going through this Instructable is things will change. Stuff will sometimes look a little different than they did in earlier steps. The way I operate on these designs is that I get about 80 percent or so of the details worked out. The rest get sorted on the fly as I'm building. Sort of a concurrent design/build process. Trying to get it all right at the beginning usually never works for me and I end up changing it anyway. Engineers (like me) have a habit of designing projects to death. This helps me get out of the design phase and start building sooner. It mostly works. Ha... and sometimes it bites me in the backside too!

A Few Thoughts on Clock Design


A clock is basically a device that mechanically performs a division operation. We are building a 12 hour clock so we have the following: Our pendulum will be sized to yield a one second period. Now, there are 60 seconds in a minute, 3600 seconds in an hour, and 43200 seconds in a twelve hour period. Also, there are 60 minutes in an hour and 720 minutes in twelve hours. And finally there are twelve hours in a twelve hour period (duh!). The task is to take those 43200 seconds and figure out how to divide them up in a manner that will indicate how many minutes and hours have transpired during the day. We will be using gears to solve that problem.

When you use one gear to drive another the rate at which the driven gear rotates is proportional to the ratio of the diameters of both gears. Let's say gear X is driving gear Y and gear X makes a full rotation in one minute. If gear X is 1 inch in diameter and gear Y is 10 inches in diameter then gear Y will take 10 minutes to make a full rotation. This is the result of taking the diameter of gear Y, dividing that by the diameter of gear X, and multiplying that by the time it takes for gear X to rotate once. Knowing this allows us to formulate a strategy.

One more note on gears... in the clock world big gears are called wheels and small gears are called pinions. "Big" and "small" are kind of subjective so how do you know? My measurement system was... If I had a gear that was big enough where I could make decorative cutouts then I called that a wheel... if the gear was too small for decorative cutouts then that was a pinion.

There are 9 total gears. 2 will have 60 teeth, 1 gear with 64 teeth, 3 gears with 8 teeth, 1 with 10 teeth, 1 with 30 teeth, and 1 with 32 teeth. All but 1 will be involute gears (more on that later). The exception will be the escape wheel which will have a saw tooth arrangement. The escape wheel will have 60 teeth and have an 8 tooth pinion on the same axis. The 8 tooth pinion will drive another 60 tooth wheel (providing a 7.5 to 1 reduction) that has a coaxial (meaning it will be on the same axis) 8 tooth pinion. That 8 tooth pinion will drive a 64 tooth minute hand wheel (providing an 8 to 1 reduction) that has a coaxial 8 tooth pinion. That 8 tooth pinion will drive a 32 tooth wheel (providing a 4 to 1 reduction) that has a coaxial 10 tooth pinion. And the 10 tooth pinion will drive a 30 tooth hour hand wheel (providing a 3 to 1 reduction). The hour hand wheel will be on the same axis as the minute hand wheel but will be attached to a bushing that rides on the arbor (or axis). Multiplying all of the gear ratios gives you: 7.5 x 8 x 4 x 3 = 720. So the escape wheel will rotate 720 times for the hour hand wheel to complete 1 rotation. Since the escape wheel turns at 1 rotation per minute the hour hand wheel will turn once every 720 minutes (or 12 hours). The minute hand wheel is reduced by 60 to 1 which is the result of multiplying the first two gear ratios (7.5 and 8). So the minute hand rotates once for every 60 rotations of the escape wheel or in other words… once every 60 minutes which is one hour.

Now the are a bunch of different ways to come up with a formula to divide up minutes and hours involving different sizes of gears. But you have to keep in mind how big the resulting gears will be. For example, if you had a gear that was one inch in diameter driving one that was 6 feet in diameter you would get a divide by 720 operation... but a 6 foot diameter gear is kind of unreasonable so better to use a train of 4 smaller gears instead!

Gear Design and Cutting

IMG_3219_Original.jpeg
IMG_1706.jpeg
IMG_3222_Original.jpeg

We have to start with a couple of assumptions. I selected 7 inches for the pitch diameter of the 60 tooth wheel to match the escape wheel dimension. (I selected 7 inches for the escape wheel because it seemed to be aesthetically the correct size and would easily fit in the laser cutter.) 14.5 degrees was selected for the pressure angle since that’s what was used in the example I found on the internet for an involute gear. After that it’s a matter of plug and chug.

Diametral Pitch = Number of teeth/Pitch Diameter

Base Circle Diameter = Pitch Diameter x cosine (Pressure Angle)

Addendum = 1/Diametral Pitch

Dedendum = 1.157/Diametral Pitch

Outside Diameter = Pitch Diameter + (2 x Addendum)

Root Diameter = Pitch Diameter – (2 x Dedendum)

Tooth Spacing = 360/Number of Teeth

The base circle is used to generate the involute curve. Keep in mind that what you’re trying to draw is the curve that would be produced by the following method: 

1.      Wrap a piece of string around a circle.

2.      Attach a pencil to the end of the string.

3.      Draw with the pencil as the string unwinds from the circle.

The process of constructing this curve with a CAD program is fairly simple but I’m sure it’s going to sound really complex but here goes… Draw the four different circles (root, base, pitch, and outside) with a common center using the diameters calculated from the equations previously described above. Draw a line from the center point (of the circles) to the base circle. If we say that 0 degrees is straight up on the y axis then we want the line to be at 90 degrees (or on the x axis). Starting at 90 degrees make several copies of this line but rotate each subsequent copy counterclockwise by 5 degrees. Kind of like spokes on a bicycle wheel. How many times you need to do this depends on how many teeth are in the wheel. Start off with 10 copies. The first copy of the line will be offset from the original by 5 degrees and the second copy will be offset from the first copy by 5 degrees and so on. Let’s go ahead and call these lines “spokes”. The first spoke (at 90 degrees) is fine as it is… leave it alone. Next, draw a line starting at the endpoint of the second spoke that is perpendicular to the spoke. The line will project downward and the length will be calculated to be 5/360 x Pi x Base Diameter. This is how long the string would be if you unwrapped it by an angular rotation of 5 degrees. Then do the same thing for the third spoke except that the length of the perpendicular line will be 10/360 x Pi x Base Diameter (which is the length of string if you had unwrapped it by an angular rotation of 10 degrees. Keep repeating this process until you have completed drawing perpendicular lines on the end of all the spokes. Finally, generate a smooth curve starting at the endpoint of the original spoke through all of the endpoints of the subsequent perpendicular lines at the ends of the spokes. The curve should start at the base diameter and go through the outside diameter. If it doesn’t go through the outside diameter then you’ll need to draw more spokes and perpendicular lines. Once you get the involute curve generated you can delete all of the spokes and perpendicular lines to keep everything from getting too cluttered. Now draw a line from the center to the intersection of the involute line and the pitch diameter. Rotate this line counterclockwise (around the center) by the following angular amount: 360/number of teeth/4. Next mirror the involute curve using the line as your mirror axis. This will generate the other side of your gear tooth. Clip the tops of the involute curve where they project past the outside diameter. Now that you have two sides of the gear tooth delete the mirror axis line. Draw two lines starting from the center to the nearest endpoints of the involute curves (the side that touches the base circle). Trim off the part of these lines that is on the inside of the root circle. Get rid of the outside diameter except for the part that forms the top of the gear by using the two involute curves as cutting edges. Delete the base circle and the pitch circle.  That completes creating the shape of the gear tooth. The tooth needs to be copied and rotated (at equal spacing) around the root circle to give you a gear with the number of teeth this gear was designed for. Draw a circle at the center of the gear for the axis of rotation. This is the basic gear. It is interesting to note how the shape of the teeth change based on the size of the gear (from 64 teeth down to 8 teeth). The smaller gears (pinions) are left as is. The larger gears (wheels) have cutouts in the interior to make them lighter and more visually pleasing. That’s easier done than said so I’m not going to describe that process. Besides, the shape of the cutouts is whatever you want it to be.

The resulting drawings are saved as DXF files and imported into the laser cutter. I think most laser cutters and CNC machines accept this format. I used the xTool machine to cut all of the gears as well as the hands. If you don't want to design your own gears you are welcome to try the ones I came up with. The hands are included too. They're cut from 1/8 inch Baltic birch plywood.

You will need the following:

  1. Three of the 8 toothed pinions. One of them will will require you to enlarge the hole to 5/32 inch.
  2. One 10 toothed pinion.
  3. One 30 toothed wheel.
  4. One 32 toothed wheel.
  5. One 60 toothed wheel.
  6. One 64 toothed wheel.
  7. One 60 toothed escape wheel.
  8. One minute hand.
  9. One hour hand.

Frame

The main consideration for this part of the frame is to get the gear spacings correct. also, the frame is not symetrical with respect to the hole patterns. So make sure you keep track of which is the front and which is the back. The view in the drawing is the front view. As far as gear spacing is concerned... the gear spacing between two gears is the sum of the two pitch diameters.

The core piece is cut from 1/4 ply just like the gears but I also cut some pieces out of 1/4 inch cherry as well to cover the plywood and make it look nice. Legs will be added to ensure there is enough clearance for the fully assembled frame and gear train over the top of the base. There are three 5/32 inch diameter holes and two 1/8 inch diameter holes. The 5/32 holes will have the brass tubing installed in them whereas the 1/8 diameter holes will hold a piece of brass rod. I used an Xacto knife to cut the tubing. You have to roll the tubing under the blade over and over with light pressure until it cuts through. The cut needs to deburred and cleaned. The rod is cut using wire cutters... once again the rod will need to be sanded smooth and cleaned up.

NOTE: The brass rod and tubing worked OK but I'm looking at trying something different on the next design revision.

Downloads

Pendulum Design

IMG_1754.jpeg
IMG_1756.jpeg
IMG_1757.jpeg
IMG_1760.jpeg
IMG_1763.jpeg
IMG_1759.jpeg
IMG_1764.jpeg

Pendulum design is a matter of personal taste. Lots of different ways you can go. One thing to keep in mind is that a pendulum with a one second period is a little less than 10 inches long. I thought that I could make it a bit longer and then adjust the effective length by adding a brass screw with a knurled nut (pics 4-7). Turns out that wasn't very effective since I made the pendulum WAY too long (about 2 inches too long). This miscalculation cost me. I had to totally rebuild the coil mount. Ha... I was trying to make the coil somewhat inconspicuous and it kind of turned into a main feature. Oh well... I'll correct that on the next design.

You do want to keep the bob somewhat light. Otherwise the coil won't be able to drive the pendulum to the necessary amplitude to make the escapement work. I drilled a 3/4 inch hole on one side of the block of wood shown in picture 2 and a 3/8 inch hole on the other before I shaped it round in pic 3.

I encased the neodymium magnet in the little cyclindrical piece. Made it easy to glue it in place on the end of the bob and it looked nice.

Walking Escapement

IMG_1751.jpeg
IMG_1848.jpeg
IMG_1791.jpeg
IMG_1781.jpeg
May 11, 2024

Along with the pendulum... the walking escapement is the star of the show on this clock. There's something fascinating about watching it click back and forth as it pushes the escape wheel around and around. And the pendulum swing is crazy! Almost 75 degrees of amplitude. You'd think that this arrangement would result in very erratic time keeping... but oddly it is quite stable. I didn't laser cut the walking escapement pieces. I had some paper templates already printed out and they were easy to cut on the scroll saw. In picture 2 you see the spacing between the two pawls. This distance will depend on how much pendulum swing your clock makes. Mine was between 70 and 75 degrees so using a little trigonometry this is what I came up with. If your swing is greater then this distance will be less. If it's less then it will be more. The goal is to have the escape wheel advance one tooth at a time. The top pawl is the one that actually advances the escape wheel. The bottom pawl prevents it from recoiling too much as the top pawl is on the return stroke. You DEFINITELY want the wheel to move back just a little bit in order to give yourself a little margin for the bottom pawl.

Getting the escapement mechanism installed and adjusted was the most fiddly part of the clock. It has to be just right. In picture 2 you can see the spacing of where the pawls engage the escape wheel. Picture 3 shows the pawl location when the pendulum is at zero (straight down). The video shows proper operation of the escapement. Take particular note of where the pawls land on the escape wheel throughout the cycle. If yours doesn't match the video you will need to loosen the set screw and rotate the pawl assembly one way or another. With a little trial and error you'll get the feel for it.

Putting It All Together

IMG_1847.jpeg
IMG_1845.jpeg
IMG_1842.jpeg
IMG_1841.jpeg
IMG_1840.jpeg
IMG_1839.jpeg
IMG_1838.jpeg
IMG_1834.jpeg
IMG_1837.jpeg
IMG_1836.jpeg
IMG_1835.jpeg
IMG_1833.jpeg
IMG_1832.jpeg
IMG_1831.jpeg
IMG_1830.jpeg
IMG_1829.jpeg
IMG_1828.jpeg
IMG_1827.jpeg
IMG_1826.jpeg
IMG_1825.jpeg
IMG_1824.jpeg
IMG_1823.jpeg
IMG_1821.jpeg
IMG_1822.jpeg
IMG_1820.jpeg
IMG_1819.jpeg
IMG_1818.jpeg
IMG_1817.jpeg
IMG_1815.jpeg
IMG_1814.jpeg
IMG_1813.jpeg
IMG_1812.jpeg
IMG_1811.jpeg
IMG_1810.jpeg
IMG_1808.jpeg
IMG_1807.jpeg
IMG_1806.jpeg
IMG_1805.jpeg
IMG_1804.jpeg
IMG_1803.jpeg
IMG_1802.jpeg
IMG_1801.jpeg
IMG_1800.jpeg
IMG_1799.jpeg
IMG_1798.jpeg
IMG_1797.jpeg
IMG_1796.jpeg
IMG_1795.jpeg
IMG_1794.jpeg
IMG_1793.jpeg
IMG_1789.jpeg
IMG_1788.jpeg
IMG_1787.jpeg
IMG_1786.jpeg
IMG_1785.jpeg
IMG_1784.jpeg
IMG_1783.jpeg
IMG_1782.jpeg
IMG_1780.jpeg
IMG_1793.jpeg

Once I had the clock built and everything was dialed in just right I let it run for a couple days to verify all was well. Then I took it apart piece by piece and took pictures at each step. Here I present the pictures in reverse order to illustrate the assembly process. For the most part everything is a friction fit for now. It helped me move and adjust things along the way. That will more than likely change on the next version. I would like for it to be a bit more robust and less likely to be knocked out of adjustment.

I don't offer a lot of discription here. Hopefully the phrase "a picture is worth a thousand words" works here. If you do find that something is confusing then post a question below and I'll be happy to respond. To summarize: The 60 toothed wheel goes in first through the left side of the frame. A 8 toothed pinion will ride on the same arbor but is on the backside of the frame. A brass rod is inserted on the right side of the frame. An 8 toothed pinion is first installed on the rod and then the 60 toothed escape wheel. The 64 toothed wheel is on the backside of the frame riding on a brass rod that runs through the middle vertical hole on the frame. An 8 toothed pinion is installed on the front side of the frame on the same rod. The 32 toothed wheel is installed on a brass rod in the bottom hole of the frame. On top of that is the 10 toothed pinion installed on the same rod. The 30 toothed wheel/hour hand assembly goes onto the brass rod in the center hole of the frame. And finally the minute hand is a friction fit over the same brass rod that you just installed the hour hand on. Ugh... I hope I got this right. I've read this section a number of times to be sure. But I'll fix it if it's discovered to be in error. I know the pictures are right! I'm constantly reviewing my Instructables... especially after I first release them to correct any typos or errors that I may have missed.

Other than the escapement assembly there is one other fiddly adjustment. The gap between the neodymium magnet in the end of the pendulum and the nail head in the electromagnetic coil. There is a "just right" sweet zone. Either too close or too far apart won't work. It's another one of things that you simply have to play with for a while to get it working. But rest assured it's not that hard.

Ha... one side benefit to the way I did things here is that it makes me look extremely organized (which I'm not). I bagged all the related parts together as I took the clock apart so I wouldn't lose them. Here it looks like everything is so organized and ready to go for assembly ahead of time. LOL... yeah, I'll pretend that's what happened.

The finished clock is approximately 17 inches high, 12 inches wide, and 7 inches deep. It weighs 3-1/4 pounds. The largest wheel has 64 teeth and is 7-5/8 inches in diameter. The smallest pinion has 8 teeth and is a little over 1-1/8 inches in diameter.

When installing the hands I would recommend that you initially place them in the 12 o'clock position. That's the easiest position to get right. When the clock is completed you can set the time by lifting the escapement pawls off of the escapement wheel and spinning the wheel manually until you reach the desired setting.

Final Thoughts and What's Next...

IMG_1849.jpeg

I am pleased that I was able to get a stable running design but I see a number of areas for improvement. I found some very nice inexpensive bearings on Amazon that I'm going to try next time. They're a little greasy so I'll clean them and use a little clock oil to lubricate them. I wasn't totally satisfied with using the brass rod and tube for the gears to ride on. I also need to do a better job at making the two fiddly parts easier to adjust and less prone to being knocked out of adjustment. I do want to try to have all of the parts laser cut too. That will get rid of a lot of the construction variation. Oh... I do have an idea to make the coil wrapping process a bit easier and hopefully quicker... though it will still probably take a few hours to complete.

Once I get all of this done I'll release a new Instructable. I'd recommend waiting for the new design... but the current design is buildable. It will just take a bit more initiative on your part to complete.

As always... comments and questions are welcome. There is a lot to this Instructable and it's possible I might have missed something or made an error. Feel free to point out any possible mistakes and I'll take a look. You won't hurt my feelings and I'd appreciate the help! Ha... that's kind of the way things work nowadays... release the beta version and fix the bugs later. And if you do build your own version please post pictures! I'd love to see your work.

Thanks for checking out my work!!!

SW

Update

IMG_1850.jpeg
IMG_1852.jpeg
IMG_1851.jpeg
IMG_1853.jpeg
IMG_1854.jpeg
IMG_1856.jpeg
IMG_1855.jpeg
IMG_1857.jpeg
IMG_1858.jpeg
IMG_1861.jpeg
IMG_1863.jpeg
IMG_1864.jpeg
IMG_1865 Copy.jpeg

May 16, 2024: I've taken the clock apart and I'm rebuilding it with the 1/2 inch bearings. So far so good! There is much less wobble in the gears and everything is more solid overall. The first picture shows the first 5 gears that divide seconds into minutes. The next phase is to add the gear train that produces hours. For that I've decided to add a second plate which I believe will result in a more robust arrangement. I should finish that tomorrow and then will move on to the drive system (pendulum, electromagnetic coil, and walking escapement). The goal is to have the clock rebuilt before the Making Time contest deadline expires. More tomorrow...

May 17, 2024: The rest of the clock train has been assembled and tested. All good. In the third pic you can see the secondary plate that I added. There is a screw that holds it in place and is tightened from the back. It allows for small adjustments to be made... and turned out to be very helpful. Time (ha) to repackage the electromagnetic coil.

So the coil is now reinstalled and is adjustable. I can dial in about a 1 inch range. Funny, I set out originally to try to minimize the visual impact but now it's a feature! Another thing I noticed today was that I had some low quality quarter inch dowels that were not straight. These caused a number of problems. Picture 3 shows how the one gear is really skewed! So new dowels were made (this time straight ones) and all is better. Hopefully I can wrap things up tomorrow and have the updated version completed. Once again, I've learned a lot. More ideas to improve the design. I'm going to get this right!

May 18, 2024: I started off by staining the hands. Now they don't get lost in the jumble of gears and are much easier to see. At first I was going to have a time wheel but then I felt as if it would visually detract from the gears and I very much want the gears to stand out.

Woo hoo! She's running again. That wasn't too bad... three days of tinkering. Most everything is still a friction fit so I need to go back and start gluing things in place but I'll hold off on that for now in case I want to do some more tinkering. I do like the bearings more than the brass rod and tube. If you're interested here's the link for the bearings.

So I'm calling it good enough for now. I need to think about everything and kick around the many ideas I have in my head for improvements. I'll pick this back up later in the year... maybe fall.

Once again, thanks for checking out my work.

Until the next Instructable...

SW

Batteries!

IMG_1871.jpeg
IMG_1872.jpeg
IMG_1873.jpeg
IMG_1874.jpeg
IMG_1877.jpeg
IMG_1878.jpeg

May 20, 2024: Woo hoo!!! I figured I had just enough time to do one more update before the contest closed so I revisited the battery powered concept. I had originally started off using 6 C cells but was having trouble with them. The coil just didn't seem to want to work. I checked the output of my so called 9 volt plug in module and found it was actually 12 volts. I couldn't fit any more C cells in the little battery box so I went with 8 AA batteries to bump it up to 12 volts. HA! That did the trick. We are cordless now!!! I'm a very happy clock builder this morning. AA batteries should give me about a month of run time. I'm curious to see if that pans out. I'll make a bigger battery box on the next design to fit C cells and get more run time. So that's it for sure now! Last update.

By the way... I know I messed up the wire colors going to the coil. I actually got them wrong at the coil. It's ok... the electrons don't know!

SW