Brilliant.. !! I am halfway finished printing this clock. Downloaded kit for the table top clock. Instructions and videos are excellent. Your detailed instructions via video and pdf are clear and well thought out. An amazing job .. the debugging has helped me find a couple of mistakes and help size the weights properly. Thanks for the table clock electronics and design, my next project. Modified the wind drum (double diameter) to increase torque without adding weight ... works great.
Steve I have finally completed my build, have a 2.8Kg weight - overkill for the 10 day setting, but I didn't have anywhere else to store the sand! The problem I have is the clock keeps stopping after a few seconds. I've been through you diagnostics, the pendulum seems fine. Without the pendulum engaged the mechanism "free wheels" on the the weight, and seemed to start without issue... well I say without issue, something that other builders may appreciate knowing about is I found the that the rough an ready winding of the fishing line onto the capstan meant that it could bind on the big gear of the winding arbour. I also initially found it caught on the tail from the knot on the capstan too. What seems to be the issue is there isn't enough assistance from the weight ensure the escapement is always progressing far enough. If I add pressure to gear 3 then it seems to help, but I can't believe the 10 day is needing upwards of 2.8Kg?
2.8kg of drive weight might not be enough for 10 days of reliable runtime. A pound per day (around 0.45kg) is a good starting point until everything is tuned and broken in. A simple test is to hang the weight directly on the line without using the pulley. This doubles the effective weight at the expense of cutting the runtime in half while debugging the clock. The pendulum should free swing for at least 5-10 minutes. If the clock stops after a few seconds, then the escapement may be interfering with the pallet instead of pushing it with each swing. Is the escapement rotating with each swing? The escapement should start spinning quickly instead of sluggishly. It is OK to add a thin coat of grease to the pinions to reduce friction in the gear train. Does the tick and tock occur at approximately equal angles on each side? Also, the drive weight will lose a lot of energy if the line is binding on a gear or onto itself. Start the debug with fewer wraps around the drum to reduce the possibility of the cord binding.
@@stevesclocks the free pendulum decay seems to be pretty close to what you have in this video - so I think that is ok. Initially it runs correctly, then the pallet file to move on a tooth in the escapement, but it makes the next, then within a few more oscillations the escapement and pallet are simply rocking in place, the lower end of the pallet seems no longer move out far enough to let the escapement round a tooth. I will try the weight direct rather than back to the clock chassis see if the extra weight helps, if does there is a little space left, in the shell, otherwise I'll have to print an extension. Thanks for replying so quickly
@@tomsimmons4482 It sounds like the pendulum is decaying faster with the escapement than without. The escapement is not adding any energy to the pallet and interference is actually taking away the pendulum's momentum. Maybe try going back to the "free wheel" test without the pallet. But stop the escapement with your finger and watch how quickly it spins back up when you release. It should be almost instantaneous. If it is sluggish, then figure out where the friction is coming from. Start and stop the escapement many times. It should always start spinning quickly. Here are a few things to try and some potential fixes: Winding cord binding up? Wind carefully for this test. Gear 9 binding up? Remove it for this test. Is the frame pinching any of the gear stacks? Trim the height of one or more gears. There should be some end shake in all gear stacks. Gear tooth roughness? Look for burrs or elephant foot. Grease the pinions. etc. Your clock should be very close to being functional. It should run great once you find the friction source. And you may need another kg of weight to get it running reliably. For now, just loop the cord back onto itself to bypass the pulley and run these tests with double the effective drive weight.
@@stevesclocks I will redo the decay test and confirm, but I fairly sure it was about 10 minutes. I might have something that could shed some light on matters however. I suspect there might be a friction issue in the main train. When allowing it to freewheel so as to unwind the fishing line and allow me to wind it back up "nicely" I re-checked the ability to restart itself if stopped, it's pretty good, it always starts and is up to speed all but instantly. However, as I noted before, the hands don't turn. If I gently push on gear 5 it will starts to turn and hands turn, but only for a few seconds then it stops. As you can see from this video th-cam.com/users/shortsuyJHEJ6qHM0?si=fsbp5a1dDllsz_Ic gear 4 and the friction drive are turning, but not gear 6. Could this be the cause? I am currently printing a 3.2" quarter section so as to increase the weight.
@@tomsimmons4482 The large gear 4 allows power to be transmitted from the weight shell to the escapement. The spring in the friction clutch is supposed to push against the small gear 4b to turn the hands. So, the problem is either not enough pressure in the spring or too much friction in gears 5 and 6. Gear 6 has a printed shaft that passes through the front dial. Make sure it spins freely. Sand or file parts if needed. This is often a problem with different printer or filament tolerances. The parts are designed with loose tolerances that work great for some builders but are too tight for others. Designing with looser tolerances would make a sloppy fit for some builders. Stretch the spring if needed to make sure it applies pressure against gear 4b. There are a lot of gears stacked onto each other on the central arbor. Make sure there is a small amount of end shake so the gear stack does not get pinched by the frame. There is some gear wobble in your video. It's probably OK, but make sure none of the gears are rubbing along their side walls. The extra weight shell may or might not be needed after reducing friction in these gears. It shouldn't hurt to add it.
Steve, I built your earlier design, which is now working perfectly. My issues were ALL with the print quality with a slightly squashed first layer giving some of the gears a rough edge which led to too much friction. Not an issue with 995 of my prints giving good bed adhesion but with gears for a clock it was. A quick 3d print of some small sanding sticks and an hour in the sun sanding out the defects solved this. I intend to build this clock soon and when I do I'll print the escapement and at least the 1st 2 gears on rafts to solve this. Thank you for your designs. Andy
Thanks Andy, Getting a perfect first layer is a balance between too much squish or not enough adhesion. The bed needs to be heated, but PLA starts to melt a bit if it is too hot. Slicers can often compensate for the elephants foot. I usually keep it set around 0.12mm. And the gears have a fair amount of backlash built in. You can try a raft, but watch the first real layer. The gear teeth start printing with a single narrow stripe and work outwards. The bottom edge of the tooth might not form properly if the center stripe does not stick to the raft. It should resolve within a few layers, but may print better without a raft. It is easy enough to try both ways. The worst case is an hour with a sanding stick or small file. Steve
In Cura 4.8 , it is the slicer I use, there is a string in "Shell" , initial layer horizontal Expansion, can bet set to a negative value to offset the elephant foot While this depends on the printer, the leveling, the layer, the temperature, the filament... I found that -0.25 works for me. I usually use .2 mm layer, printing on a glass bed Great video on tuning BTW
Steve I wonder if I might once more call upon your assistance in debugging the clock? As noted previously the gears were rather "wobbly" when running free, added to that where the face screws to the back supports started to spit as did where the string join to the front back support. The to top it all off the whole thing stopped after about 2 weeks and I couldn't get it to run for more than an hour since. I therefore having improved my printing knowledge and set up decided to reprint all the gears and the back (with a slight mod to create a peg or hook for the string to hook to on the bottom support and increased the infill where all the screws are). I included all the updates you have done, and the tolerance on the arbours etc is much much better. The issue I have is the drive (weight/pendulum/escapement) is running well, but hands are reluctant to move. I'm guess for half of the mechanism to running while the other half isn't it must be something to do with the friction section on the central arbour? The issue existed with the escapement removed so the mechanism was free running on the weight alone, it can be coaxed into running, but soon stops, might start on it's own for a while...
It seems like there may be too much friction in the central arbor. The most common places are gear 6 binding where it passes through the dial, or the gear stack getting pinched by the frame. Check for a small amount of end shake on the central arbor. Trim the height of one of the spacers if needed. Frame pinch will steal a lot of the drive power. Then look at the escapement to decide where to debug next. If the escapement is sluggish, focus on gear train friction. If the escapement moves quickly with each tick, then double check the pendulum free swing test. Sometimes the bearings need a second cleaning. This clock has a lot of runtime options. 32 day mode will run with the largest drive weight, but the pendulum will have a very shallow amplitude. 8-10 day modes will be significantly easier to keep running, even with smaller drive weights.
@@stevesclocks Thanks. I did wonder about the entire arbor being trapped, but it felt like these was some play. The pendulum is free swinging for over 10 mins. The free run test was smooth and the stopping the escapement and letting it run again it was quick and smooth at all points. I'm running the 10 day option currently. Is there an updated consolidated guide?
What setting on the slicer will make the pendulum look like that with the spiral front. I’ve used spiral vase mode setting on cura and it makes the part hollow and fragile.
Concentric bottom layer in Prusa Slicer. Cura should have the same option. It works great on simple circles or ovals. The back side with screw hole openings doesn't work as well.
Hi Steve, having now printed your clock, and made all the adjustments, I take my hat off to you, bloody amazing, truly a thing of beauty. I also have to thank you for introducing me to Prusa Slicer , what a world of difference from Cura, the quality of printing is now amazing. I have not quite figured out the pendulum adjustment, especially at the top two adjustment nuts. If you could make that clearer I would be grateful.
The purpose of the top adjustment is to allow the pendulum shaft to rotate slightly left or right relative to the pallet arms. The clock should be adjusted to tick and tock equally when moving the pendulum left and right. This is called setting the beat. There are two ways to set the beat in this clock. 1) Use two mounting screws to hang the clock on the wall. This keeps the clock body solidly held in place. Use the top adjustment nuts to move the pendulum forward or backward along the angled groove to change the angle of the pendulum arm. 2) Use only the mounting screw to hang the clock on the clock on the wall. Set the top adjustment nuts near the center of the adjustment range. Tilt the entire clock frame slightly to change the beat of the clock. A small shift in either direction is usually all it takes. Method number 2 is the easiest to set up and adjust. It is the method I use for clocks in my house. One slight downside is that the clock can shift when winding and you may need to reset the beat. You want the clock to tick...tock...tick...tock instead of tick.tock.....tick.tock. Steve
@@stevesclocks Hi Steve, many thanks for the reply. The clock is up and running very sweetly. I'm absolutely amassed that a 3d printed clock runs so well! Hats off to you. She's running a little slow about a minute over 24hrs. This I will reduce by raising the pendulum weight, in a few weeks once the clock has run in and settled. Fantastic little project thank you.
Remove the rubber seals with a safety pin and soak them in 90% or stronger alcohol. Mineral spirits or other solvents also works well. This should loosen most of the grease. A blast of compressed air speeds things up. You can add a drop of dry silicone lube or very light sewing machine oil to help prevent rusting.
They should still work to test the clock. It might need some extra weight later if the grease thickens over time. Or it might be fine. I have never really tried it. The easiest bearings to find are 623RS with a rubber seal that is easy to pick out using a small pin.
I have printed the 32-day clock for the 7-day time, it has been running for 4 days and it loses 30 minutes in 24 hours it was printed on my Prusa MK3 printer the files were printed with a 0.4mm nozzle, 3 perimeters and 30% infill. I printed the updated #3and 4 gears, the clock is perfectly level to the wall also left and right. The tick and tock are equal I checked with a digital sound meter 5.73db I have checked the gears without the pendulum attached and they rotated smoothly, I am using a 9 lb. weight.
The best news is that you have the clock working. The only thing left is to calibrate the rate. I have found that the deadbeat escapement is fairly tolerant of changes in drive weight since the escapement only pushes the pendulum for a short amount of time. On the other hand, an anchor escapement pushes the pendulum a lot more and the rate varies wildly with various drive weights. The rate on this clock is set by the pendulum length. theindex.nawcc.org/CalcPendulumChange.php has a simple calculator that implies a 15" pendulum needs to be reduced to 14.4" to change the rate by 30 minutes per day. There are shorter pendulum arms that can be used if there is not enough adjustment left below the bob.
Great Jobs Steve.. Well engineered, You must have been a clock repairman for you knowledge, I bought your design and its great .. just a few things that made it hard was which gears for what project. It may make it easier for the zip file to have folders of which gears you need for which clock.. Eg Small clock , Med clock, Large clock and that way a person could just print everything in that folder for which clock they want. Im sure you know it all inside out , but for a new person it just slows the whole process down.. :) Im just saying that is all but a great project.. Thanks :) !!!!
Thanks. I am an engineer with no clock repair experience. I design each part so it works as a 3d printed part. I never feel the need to design the traditional way that might have been used for a brass clock. Each clock has its own zip file. One or two of the older designs had both imperial and metric parts in the same release. The newer designs are metric only, but you still need to follow along with the assembly manual because there could be multiple gear sets for different runtime options. Myminifactory combines everything into a single zip file. Hopefully you found the files you need. I will try to add better documentation to try to make things clearer.
@@stevesclocks Hi Steve, Sounds great, Yes the zip had lots of files, sorting the groups of which day time runs and the main parts into groups may help others just go to a folder and print all the parts for say a 21 day , or a 10 day etc.. would simplifies it for people who have no clue of what they are looking at, may make the zip a little larger but easier. I just think the project is very cool and a lot of work in designing your instructions are very good and well laid out in many ways, as I say great Job ! Merry Christmas :)
@@stevesclocks Hi Steve, Im just wondering of you can give me some info. I made the easy build and it will work I think once i debug it as it seems to need more weight for the 15day version, but Im wondering how come the wind gear is so large as it seems that if you could shrink that gear you could get a bigger drum the thread winds on to for the weight, and then that would have more leverage for the weight so in turn you wouldn't need as much weight to run the clock..? The gearing seems plenty for the wind handle since its so easy to turn even with alot of weight on it . ? Just wondered while I was trying to not have too larger weight but it gets tricky to make everything friction free.
@@keeweeles I considered changing the winding drum diameter and winding gear to create different runtimes. Instead, the ratchet gear and minute hand gear ratios were changed. The net effect is exactly the same, but the print times are really long with any option that needs a new winding drum. I have been running my version of this clock with either the 7.5 or 10 day options. I also added a light coat of grease to all the pinions. This allows the clock to run significantly more reliably. The 32 day option ran for a month or two. The pendulum amplitude was around 2 degrees in each direction which is just barely enough to keep the clock running, but eventually the gears started sticking. It runs so much stronger with the shorter runtime option. The pendulum amplitude is more than +/-4 degrees now.
@@stevesclocks Hi Steve thankyou for your input it is a great design and appreciate your efforts as my wife loves clocks. I have just put some lithium grease to help the gears on the pinions also to help it out. The winding drum I was mentioning wasn't to change the gearing just the diameter of what the thread winds on as the larger that is the more leverage the weight has to turn the gear, The only issue with that is the gear that winds the clock up gets in the way, so was suggesting if that was smaller it could be done and after all that gear is only to wind the clock.. Merry Christmas :)
I have assembled and dissembled this clock over six times and had my son do it to, we printed two clocks and have the same problem all the gears are not in a bind, when you put it on the wall and attached the weight to it the weight will not make the gears move the clock is locked. I have made open gear clocks on my CNC programed and driven by Arduino and been 3d printing for over 9 years. What could be locking it up.
The most likely problem is what I used to call "sticky gears" and can be seen in this video th-cam.com/video/pnpYs-SuytI/w-d-xo.html The issue is that a tiny amount of sideways thrust against the escapement can stall the clock. New helical gears were released at MyMiniFactory to help push gear 3 away from the escapement. The issue can be easily tested by manually pushing gear 3 away from the escapement. If the escapement starts rotating again, then grab a fresh copy of the new gears and print just the 2 new ones. If this doesn't work, then more info is needed to help debug the issue. Does everything spin with the pallet removed? How about with the pallet and the escapement removed? Also, what runtime option and drive weight are you using? I have switched my clock to a 14 day runtime with around 10 lbs of weight. 32 day mode runs, but it is just too sensitive. 14 days is much more stable. 8 or 10 days would be even better.
@@normhuntley5552 Maybe a bit more information is needed. Is it 5 minutes of pendulum free-swing time? The clock should still be made to run with a bit of extra drive weight if this is the case. Or is it 5 minutes of clock run time? If so, what are the symptoms when the clock stops? Does the escapement turn when the pendulum is manually moved back and forth? Check the position of the tick and tock when doing this. They should be approximately symmetrical in both directions.
It was 5 minutes of free swing your video says 10 minutes; I have the clock running now with 9 pounds of weight the only thing I did was shake it before I hung it on the wall, I figured since I had it flat on my bench maybe the gears were the problem it works.
The STL files are at www.myminifactory.com/users/StevePeterson You can download the detailed assembly guide from my web site if you want to see what is involved in each clock before purchasing the design from MyMiniFactory. I believe the cost is incredibly small relative to the hours put into each design and the assembly guide.
In my area of the US, music wire can be found at Ace Hardware or many smaller hardware stores that carry larger selections than the big box stores. Remote control hobby stores also carry it. The name is sometimes labeled as piano wire or spring steel. Most of the clocks can also use regular steel or stainless steel wire if that is all you can find. Try to find something with reasonable stiffness.
I built the small ez build clock and the gears wont rotate with the weight on. Pendulum swings only for a couple minutes. With the escapement removed and a smaller weight attached the gears wont rotate by themselves, however if i start the spinning it will spin for a minute or so. With the clock on the table the gears will spin for a moment or two once I spin the gears by hand. Taking it apart now but not sure what to look for.
The symptoms sound like what I used to call "sticky gears", but "sticky escapement" is a more accurate description. The escapement does not rotate on its own when the pendulum is manually moved from side to side. I believe it is caused by gear 3 tilting slightly and applying sideways pressure against the escapement. It only takes a small amount of pressure to stall the lightly loaded escapement. Push gear 3 away from the escapement to see if the clock starts running again, at least for a short period of time. I am getting ready to release a new set of gears that should solve the problem. It should be ready in 1 or 2 days. Gears 3 and 4 were re-designed to add a gentle helical angle that pushes gear 3 away from the escapement. This issue only affects the easy-build clocks as they are the only designs with two gears sharing the same shaft.
Brian, I updated gears 3 and 4 on MyMiniFactory with the fixes. Try downloading and re-printing these two gears to see if your clock runs better. It takes a few days for the files to make their way into the zip file, so you may need to download them manually.
@@stevesclocks Thx a lot. I did as you said in your first reply and got a bit more out of it. Reprinted gear 5 as it didnt print perfect and got it to run a couple minutes. So hopefully the new gears will work better. Thanks again.
@@stevesclocks Hi Steve, I downloaded and printed the 2 gears and it didnt help. I reprinted some other gears as well. When testing without the escape and pendulum, the gears rotate well with the weight attached. However, stopping the gears from turning and releasing, i find there is one spot the it stalls, trying to isolate the problem but not sure how to diagnose it. Any hints would be appreciated. Thx.
@@briant444 It sounds like there is an issue that is specific to your printing if the gears stick in one spot. It could be elephants foot, a blob on the gears, out of round gears, a bent arbor, or something else. Maybe try taking gears out one by one to see if you can feel any difference in the friction.
beautiful and mesmerizing!! hopping to start printing soon. my machine is a homers tevo tarantula (235x235x250mm) do you think is enough to print this?. thanks
Thanks. Yes, it will easily fit on that printer. I design on a 250x210 Prusa, but also make sure it will fit on a 220x220 Ender3. Height is never an issue.
Steve, I just finished printing and assembling this but I can't for the life of me get the escapement to move on it's own. Everything is nice and loose, the bearings are good, and pendulum has almost 10 minutes of swing on it. I can pull really hard on the weight cord and still the escapement doesn't move. If I move the winding drum myself then the clock ticks nicely. It seems to me as though the drum that the cord is wound around should be much bigger in diameter, as this would give more 'pull' on it and therefore more turning force, resulting in less weight needed?! I would appreciate your thoughts on why I can't get my clock to move on it's own, thank you.
My best guess is the "sticky gear" problem described in this video th-cam.com/video/pnpYs-SuytI/w-d-xo.html Adding extra weight only jams against the escapement harder. There are new copies of gear 3 and gear 4 posted to MyMiniFactory to solve the problem. Making the drum larger would provide more force to the escapement and the clock should become more reliable. It would also reduce the runtime. There are a few gear combinations that provide the same result of increasing power to the escapement while reducing runtime by the same proportion. Each clock design is a tradeoff between runtime and reliability. I prefer a long runtime. If I had a clock than needed to be wound every 12 hours, it would very quickly become a static work of art. :)
@@stevesclocks thanks for the reply, and sorry for the delay in getting back to you. I've now got the clock running nicely. I already had the updated gears on my clock, but I underestimated just how heavy the weight needed to be. More seriously, I found that one of the spacers was slightly too long preventing it's associated gear from turning as easily as it should. I trimmed it down and the clock started working. I then noticed that the hands weren't moving! So I swapped out the 10 day gears for the 7.8 day ones and it's been running successfully for the last 2 or 3 days 🙂. I really don't mind winding the clock every few days as long as it's actually working as it should!
@@trance_trousers Thanks for replying back with successful results. There are so many interactions that can cause a clock to stop running. It is a good feeling when you discover them on your own. Winding once a week is tolerable. Designs that need to be wound twice a day would quickly become static pieces of art.
@@stevesclocks yes, that's why I chose your clock design as it seemed to have the longest run times of any I've seen. Just curious, but would you be ok with someone selling your clocks (fully printed and assembled) on somewhere like Etsy?
@@trance_trousers Unfortunately no for several reasons. The files are released under a strict non-commercial, personal use only license. It is OK to give them to family and friends, but they are not allowed to be sold. The other important reason is that the designs are intended for each user to build their own clock and experience the joy of making it work.
Hello sir.. thanks for sharing your wonderful design... I just recently printed your smaller 32 day clock.. while I can get the gear assemblies to work flawlessly with little to no friction, I cannot say the same for the pendulum.. The pendulum is down to a half degree in about 4 minutes.. I've tried several different bearings, as well as cleaned, dried, PTFE dry lube, no lube at all... I have even gone the extra step and removed the phenolic retainers and still seeing around a 4 minute complete degradation.. I've made sure the arbor isn't to long as well as I thought that could be the issue.. Is there any advice you could give me to increase degradation times?.. I thank you for your time Steve..
There should be some end play in the pallet as well as the arbor. A small amount of frame warping could pinch the pallet. You probably already checked that. When you say you removed the phenolic retainer, do you mean the internal ball retainer? The balls may fall out if you do that. The next thing to check would be the bearings. They should spin freely for a few seconds when placed over a 3mm shaft. Cheap no-name bearings from AliExpress or eBay typically have around 5 seconds of free spin time after cleaning. They work great in my clocks with a pendulum swing test of 15-20 minutes. Even sealed bearings with the original factory grease intact run almost 6 minutes in the swing test. I have not tried expensive bearings in my clocks. Super tight tolerance bearings might be needed to run at 100K RPMs, but loose tolerance (cheap) bearings seem to work just fine at the speeds of a clock. I wonder if you got a bad batch. Do they feel gritty? One user reported better results by opening up the bearing mounting holes slightly. It seems that the bearing holes were tight and the bearings were mounted at a slight angle. Enlarging the holes allows the bearings to float a bit and not apply any side pressure when they rotate. Another thing to try is the 8 day runtime set of gears. This should give 4X as much power to the escapement as the 32 day option. Steve
@@stevesclocks thank you for the quick reply Steve . Yes I removed the phenolic ball retainer in a last ditch effort to reduce as much friction as possible.. and yes if you are not careful the bearing will fall apart.. I've since removed those as I still wasn't getting optimum pendulum swing time.. Also what I wrote about the degradation can be confusing, apologize for that.. should have been more specific.. A 5* swing will degrade to 1* within 4 mins... Complete stop at 7 mins.. so upon further inspection, I did find the bearings I purchased are filled with something similar to dielectric grease.. The alcohol and laquer thinner is just not pulling it all out... Will grab some brake cleaner from my shop tomorrow and try that... But it surely seems silicone based.. While I understand the 8 day gear set will run "easier" I'm determined to make this clock run for 30 days, a personal challenge I suppose.. As a side note, I was wondering.. if the winding drum was increased in diameter behind gear 9 (because there is space there) wouldn't the amount of weight needed to run the clock be reduced from the increased leverage applied?
@@camaroboy383 I use low odor mineral spirits to dissolve the grease. It is fairly weak compared to the old stuff, but will eventually loosen it enough that a blast of air gets rid of most of it. Brake cleaner should work better. Yes, increasing the winding drum could reduce the drive weight needed to run the clock. It would also reduce the run time. Changing two gears gives a bigger range of run time vs drive weight options. If both changes were used, it would be a bit like the gears on a 21 speed bike. :)
The first step is to build one from plans and study how it works. Break the design into small pieces that are easier to understand. The escapement is the starting point. Get it working first. Then figure out gearing that looks good. There are lots of books, mostly written about traditional brass clock design, but still plenty of good info in them.
@@stevesclocks Thanks for the reply. I've finished printing all the smaller parts from your revised pendulum clock. Do you think a wooden clock design book would translate to 3d printing?
@@Noold Most of the features are similar. Wooden clock are typically larger, so it will need to be scaled. 3D printing allows additional optimizations like printing the pinions integrated with the main gear. Frame pieces can also be combined into single prints.
Very promising kit, thank you! What maintenance should we expect? I also see, that is prone to collect dust from air. Is there a proper way to clean it/keep it dust-free? (While this exposed design is very sexy, an encapsulated one might also be interested.)
No maintenance is required except possibly blowing off any accumulated dust. The only lubrication recommended is a dry Teflon type that doesn't attract dust. My oldest clock is over two years old and still running strong. A case could be built, but due to the large size, it might be easier to build one out of wood.
Hi / just printing this clock but have built you other upright model clock / Have found it easier as regards the bob weight to do away with the suspend wheel for the 2 cord and just hang the weight on a single cord thus cutting down on the size and weight of the printed shell case, Clock runs well with 3.8 lbs and drops about 1 inch an hour / Your comments on me doing this I would like.?
Hi Janet, A single cord works just fine with the obvious limitation that the clock only runs half as long using half as much weight. The weight distribution is slightly off on the vertical clocks and the frame may want to tilt. This is easily fixed with a small nail. The pendulum bob acts as a counterbalance on the horizontal clocks and they seem to stay oriented better. A minor feature of the weight shell pulley is to help keep the cord near the center of the barrel when winding. Watch the cord position when winding to make sure it doesn't go over the edge of the barrel.
Hi Steve, I built two models of your previous 8 day watch. It was a challenge with a lot of problems, but now I can say with great satisfaction that both models have been running great for a few weeks now with 2.7 kg and 3.5 ° of oscillation. In the second model I took liberties and changed the aesthetic appearance of the dial and inserted a bearing in the minute wheel and two bearings in the ratchet wheel. In both models I have modified the holes for the shafts which are now pushed by force, in this way I do not use spacers. I use very smooth SAPPORO R2 bearings. I have now started building this new model and would like to do the 15 days. I have a question. Do you think that using bushings can reduce friction?
Hi Guido. Bearings can reduce friction, but are not really needed. I placed them near the weight shell to support the heavy loads and at the pallet to support the pendulum. Most of the arbors in the middle don't really need them because the speed is so slow. The escapement might benefit due to the high speed even though the load is very low. Simple music wire arbors in PLA have very low friction. Bushings used in my first clock have about the same amount of friction. The biggest benefit is that they should last longer. I don't think the friction is reduced enough to justify the extra effort. 15 days of runtime is the upper limit for the large version of the clock, but should be easy in the medium size clock. The biggest factor to achieve a long runtime is the pendulum support bearings. Steve
Hi Steve, I am very interested in the watch shown. Your video is very insightful. I live in austria and I cannot find a shop nearby. Pleased can you help me, to find an opportunity to find this clock in austria or in the european Union. Thanks very much and best regards from austria, Erich
One of the biggest comments on the last design was the difficulty of finding the stainless steel tubing for bushings used throughout the clock. This design gets rid of them and the assembly process is considerably easier. However, it is a real clock that needs a little bit of precision to get it to function properly. Bearings are used in the places where they are truly required, the weight shell and the pendulum support. Music wire typically comes in a hardened state. It is very stiff and more than strong enough to support the ratchet. Music wire is sometimes called piano wire or spring wire. It should be readily available worldwide. Soft steel wire may bend easily and might not work very well. One builder reported that he used a paper clip by clamping one end and spinning the other end with a drill. It work hardened enough to barely work in the clock. But seriously, find some music wire and it will work much better. Start with the shorter runtimes with smaller weights if needed. Winding a clock once a week is still a very functional clock.
Brilliant.. !! I am halfway finished printing this clock. Downloaded kit for the table top clock. Instructions and videos are excellent. Your detailed instructions via video and pdf are clear and well thought out. An amazing job .. the debugging has helped me find a couple of mistakes and help size the weights properly. Thanks for the table clock electronics and design, my next project. Modified the wind drum (double diameter) to increase torque without adding weight ... works great.
Steve
I have finally completed my build, have a 2.8Kg weight - overkill for the 10 day setting, but I didn't have anywhere else to store the sand!
The problem I have is the clock keeps stopping after a few seconds. I've been through you diagnostics, the pendulum seems fine. Without the pendulum engaged the mechanism "free wheels" on the the weight, and seemed to start without issue... well I say without issue, something that other builders may appreciate knowing about is I found the that the rough an ready winding of the fishing line onto the capstan meant that it could bind on the big gear of the winding arbour. I also initially found it caught on the tail from the knot on the capstan too.
What seems to be the issue is there isn't enough assistance from the weight ensure the escapement is always progressing far enough. If I add pressure to gear 3 then it seems to help, but I can't believe the 10 day is needing upwards of 2.8Kg?
2.8kg of drive weight might not be enough for 10 days of reliable runtime. A pound per day (around 0.45kg) is a good starting point until everything is tuned and broken in. A simple test is to hang the weight directly on the line without using the pulley. This doubles the effective weight at the expense of cutting the runtime in half while debugging the clock.
The pendulum should free swing for at least 5-10 minutes. If the clock stops after a few seconds, then the escapement may be interfering with the pallet instead of pushing it with each swing. Is the escapement rotating with each swing? The escapement should start spinning quickly instead of sluggishly. It is OK to add a thin coat of grease to the pinions to reduce friction in the gear train. Does the tick and tock occur at approximately equal angles on each side?
Also, the drive weight will lose a lot of energy if the line is binding on a gear or onto itself. Start the debug with fewer wraps around the drum to reduce the possibility of the cord binding.
@@stevesclocks the free pendulum decay seems to be pretty close to what you have in this video - so I think that is ok.
Initially it runs correctly, then the pallet file to move on a tooth in the escapement, but it makes the next, then within a few more oscillations the escapement and pallet are simply rocking in place, the lower end of the pallet seems no longer move out far enough to let the escapement round a tooth.
I will try the weight direct rather than back to the clock chassis see if the extra weight helps, if does there is a little space left, in the shell, otherwise I'll have to print an extension.
Thanks for replying so quickly
@@tomsimmons4482 It sounds like the pendulum is decaying faster with the escapement than without. The escapement is not adding any energy to the pallet and interference is actually taking away the pendulum's momentum.
Maybe try going back to the "free wheel" test without the pallet. But stop the escapement with your finger and watch how quickly it spins back up when you release. It should be almost instantaneous. If it is sluggish, then figure out where the friction is coming from. Start and stop the escapement many times. It should always start spinning quickly.
Here are a few things to try and some potential fixes:
Winding cord binding up? Wind carefully for this test.
Gear 9 binding up? Remove it for this test.
Is the frame pinching any of the gear stacks? Trim the height of one or more gears. There should be some end shake in all gear stacks.
Gear tooth roughness? Look for burrs or elephant foot. Grease the pinions. etc.
Your clock should be very close to being functional. It should run great once you find the friction source. And you may need another kg of weight to get it running reliably. For now, just loop the cord back onto itself to bypass the pulley and run these tests with double the effective drive weight.
@@stevesclocks I will redo the decay test and confirm, but I fairly sure it was about 10 minutes.
I might have something that could shed some light on matters however. I suspect there might be a friction issue in the main train. When allowing it to freewheel so as to unwind the fishing line and allow me to wind it back up "nicely" I re-checked the ability to restart itself if stopped, it's pretty good, it always starts and is up to speed all but instantly. However, as I noted before, the hands don't turn. If I gently push on gear 5 it will starts to turn and hands turn, but only for a few seconds then it stops. As you can see from this video th-cam.com/users/shortsuyJHEJ6qHM0?si=fsbp5a1dDllsz_Ic gear 4 and the friction drive are turning, but not gear 6. Could this be the cause?
I am currently printing a 3.2" quarter section so as to increase the weight.
@@tomsimmons4482 The large gear 4 allows power to be transmitted from the weight shell to the escapement. The spring in the friction clutch is supposed to push against the small gear 4b to turn the hands. So, the problem is either not enough pressure in the spring or too much friction in gears 5 and 6.
Gear 6 has a printed shaft that passes through the front dial. Make sure it spins freely. Sand or file parts if needed. This is often a problem with different printer or filament tolerances. The parts are designed with loose tolerances that work great for some builders but are too tight for others. Designing with looser tolerances would make a sloppy fit for some builders.
Stretch the spring if needed to make sure it applies pressure against gear 4b.
There are a lot of gears stacked onto each other on the central arbor. Make sure there is a small amount of end shake so the gear stack does not get pinched by the frame.
There is some gear wobble in your video. It's probably OK, but make sure none of the gears are rubbing along their side walls.
The extra weight shell may or might not be needed after reducing friction in these gears. It shouldn't hurt to add it.
Steve, I built your earlier design, which is now working perfectly. My issues were ALL with the print quality with a slightly squashed first layer giving some of the gears a rough edge which led to too much friction. Not an issue with 995 of my prints giving good bed adhesion but with gears for a clock it was. A quick 3d print of some small sanding sticks and an hour in the sun sanding out the defects solved this. I intend to build this clock soon and when I do I'll print the escapement and at least the 1st 2 gears on rafts to solve this. Thank you for your designs. Andy
Thanks Andy,
Getting a perfect first layer is a balance between too much squish or not enough adhesion. The bed needs to be heated, but PLA starts to melt a bit if it is too hot. Slicers can often compensate for the elephants foot. I usually keep it set around 0.12mm. And the gears have a fair amount of backlash built in. You can try a raft, but watch the first real layer. The gear teeth start printing with a single narrow stripe and work outwards. The bottom edge of the tooth might not form properly if the center stripe does not stick to the raft. It should resolve within a few layers, but may print better without a raft. It is easy enough to try both ways. The worst case is an hour with a sanding stick or small file.
Steve
In Cura 4.8 , it is the slicer I use, there is a string in "Shell" , initial layer horizontal Expansion, can bet set to a negative value to offset the elephant foot
While this depends on the printer, the leveling, the layer, the temperature, the filament...
I found that -0.25 works for me. I usually use .2 mm layer, printing on a glass bed
Great video on tuning BTW
Steve I wonder if I might once more call upon your assistance in debugging the clock?
As noted previously the gears were rather "wobbly" when running free, added to that where the face screws to the back supports started to spit as did where the string join to the front back support. The to top it all off the whole thing stopped after about 2 weeks and I couldn't get it to run for more than an hour since.
I therefore having improved my printing knowledge and set up decided to reprint all the gears and the back (with a slight mod to create a peg or hook for the string to hook to on the bottom support and increased the infill where all the screws are). I included all the updates you have done, and the tolerance on the arbours etc is much much better.
The issue I have is the drive (weight/pendulum/escapement) is running well, but hands are reluctant to move. I'm guess for half of the mechanism to running while the other half isn't it must be something to do with the friction section on the central arbour?
The issue existed with the escapement removed so the mechanism was free running on the weight alone, it can be coaxed into running, but soon stops, might start on it's own for a while...
It seems like there may be too much friction in the central arbor. The most common places are gear 6 binding where it passes through the dial, or the gear stack getting pinched by the frame. Check for a small amount of end shake on the central arbor. Trim the height of one of the spacers if needed. Frame pinch will steal a lot of the drive power.
Then look at the escapement to decide where to debug next. If the escapement is sluggish, focus on gear train friction. If the escapement moves quickly with each tick, then double check the pendulum free swing test. Sometimes the bearings need a second cleaning.
This clock has a lot of runtime options. 32 day mode will run with the largest drive weight, but the pendulum will have a very shallow amplitude. 8-10 day modes will be significantly easier to keep running, even with smaller drive weights.
@@stevesclocks Thanks. I did wonder about the entire arbor being trapped, but it felt like these was some play.
The pendulum is free swinging for over 10 mins.
The free run test was smooth and the stopping the escapement and letting it run again it was quick and smooth at all points.
I'm running the 10 day option currently.
Is there an updated consolidated guide?
@@tomsimmons4482 The latest assembly guides would be on my web site at www.stevesclocks.com/sp5
Almost done with your motorized clock - you do a great job designing them
Amazing.
What setting on the slicer will make the pendulum look like that with the spiral front. I’ve used spiral vase mode setting on cura and it makes the part hollow and fragile.
Concentric bottom layer in Prusa Slicer. Cura should have the same option. It works great on simple circles or ovals. The back side with screw hole openings doesn't work as well.
Hi Steve,
having now printed your clock, and made all the adjustments, I take my hat off to you, bloody amazing, truly a thing of beauty. I also have to thank you for introducing me to Prusa Slicer , what a world of difference from Cura, the quality of printing is now amazing. I have not quite figured out the pendulum adjustment, especially at the top two adjustment nuts. If you could make that clearer I would be grateful.
The purpose of the top adjustment is to allow the pendulum shaft to rotate slightly left or right relative to the pallet arms. The clock should be adjusted to tick and tock equally when moving the pendulum left and right. This is called setting the beat.
There are two ways to set the beat in this clock.
1) Use two mounting screws to hang the clock on the wall. This keeps the clock body solidly held in place. Use the top adjustment nuts to move the pendulum forward or backward along the angled groove to change the angle of the pendulum arm.
2) Use only the mounting screw to hang the clock on the clock on the wall. Set the top adjustment nuts near the center of the adjustment range. Tilt the entire clock frame slightly to change the beat of the clock. A small shift in either direction is usually all it takes.
Method number 2 is the easiest to set up and adjust. It is the method I use for clocks in my house. One slight downside is that the clock can shift when winding and you may need to reset the beat. You want the clock to tick...tock...tick...tock instead of tick.tock.....tick.tock.
Steve
@@stevesclocks Hi Steve, many thanks for the reply. The clock is up and running very sweetly. I'm absolutely amassed that a 3d printed clock runs so well! Hats off to you. She's running a little slow about a minute over 24hrs. This I will reduce by raising the pendulum weight, in a few weeks once the clock has run in and settled. Fantastic little project thank you.
How do you clean and dry a bearing?
Remove the rubber seals with a safety pin and soak them in 90% or stronger alcohol. Mineral spirits or other solvents also works well. This should loosen most of the grease. A blast of compressed air speeds things up. You can add a drop of dry silicone lube or very light sewing machine oil to help prevent rusting.
Can you tell me which bearing I need. The one I have has a seal in it and it won’t come out it seems to have a metal ring below the seal
They should still work to test the clock. It might need some extra weight later if the grease thickens over time. Or it might be fine. I have never really tried it.
The easiest bearings to find are 623RS with a rubber seal that is easy to pick out using a small pin.
I found some open bearings. I don’t think the ones I have work for the pendulum it only swings for 4 minutes when the pallet is flipped backwards.
I have printed the 32-day clock for the 7-day time, it has been running for 4 days and it loses 30 minutes in 24 hours it was printed on my Prusa MK3 printer the files were printed with a 0.4mm nozzle, 3 perimeters and 30% infill. I printed the updated #3and 4 gears, the clock is perfectly level to the wall also left and right. The tick and tock are equal I checked with a digital sound meter 5.73db I have checked the gears without the pendulum attached and they rotated smoothly, I am using a 9 lb. weight.
The best news is that you have the clock working. The only thing left is to calibrate the rate. I have found that the deadbeat escapement is fairly tolerant of changes in drive weight since the escapement only pushes the pendulum for a short amount of time. On the other hand, an anchor escapement pushes the pendulum a lot more and the rate varies wildly with various drive weights.
The rate on this clock is set by the pendulum length. theindex.nawcc.org/CalcPendulumChange.php has a simple calculator that implies a 15" pendulum needs to be reduced to 14.4" to change the rate by 30 minutes per day. There are shorter pendulum arms that can be used if there is not enough adjustment left below the bob.
The clock keeps correct time after further adjustment of the pendulum.
Steve, Where do I buy one of your 32 day pendulum clocks.
They are all available at www.myminifactory.com/users/StevePeterson
Great Jobs Steve.. Well engineered, You must have been a clock repairman for you knowledge, I bought your design and its great .. just a few things that made it hard was which gears for what project. It may make it easier for the zip file to have folders of which gears you need for which clock.. Eg Small clock , Med clock, Large clock and that way a person could just print everything in that folder for which clock they want. Im sure you know it all inside out , but for a new person it just slows the whole process down.. :)
Im just saying that is all but a great project.. Thanks :) !!!!
Thanks. I am an engineer with no clock repair experience. I design each part so it works as a 3d printed part. I never feel the need to design the traditional way that might have been used for a brass clock.
Each clock has its own zip file. One or two of the older designs had both imperial and metric parts in the same release. The newer designs are metric only, but you still need to follow along with the assembly manual because there could be multiple gear sets for different runtime options. Myminifactory combines everything into a single zip file. Hopefully you found the files you need. I will try to add better documentation to try to make things clearer.
@@stevesclocks Hi Steve, Sounds great, Yes the zip had lots of files, sorting the groups of which day time runs and the main parts into groups may help others just go to a folder and print all the parts for say a 21 day , or a 10 day etc.. would simplifies it for people who have no clue of what they are looking at, may make the zip a little larger but easier.
I just think the project is very cool and a lot of work in designing your instructions are very good and well laid out in many ways, as I say great Job ! Merry Christmas :)
@@stevesclocks Hi Steve, Im just wondering of you can give me some info. I made the easy build and it will work I think once i debug it as it seems to need more weight for the 15day version, but Im wondering how come the wind gear is so large as it seems that if you could shrink that gear you could get a bigger drum the thread winds on to for the weight, and then that would have more leverage for the weight so in turn you wouldn't need as much weight to run the clock..?
The gearing seems plenty for the wind handle since its so easy to turn even with alot of weight on it . ? Just wondered while I was trying to not have too larger weight but it gets tricky to make everything friction free.
@@keeweeles I considered changing the winding drum diameter and winding gear to create different runtimes. Instead, the ratchet gear and minute hand gear ratios were changed. The net effect is exactly the same, but the print times are really long with any option that needs a new winding drum.
I have been running my version of this clock with either the 7.5 or 10 day options. I also added a light coat of grease to all the pinions. This allows the clock to run significantly more reliably. The 32 day option ran for a month or two. The pendulum amplitude was around 2 degrees in each direction which is just barely enough to keep the clock running, but eventually the gears started sticking. It runs so much stronger with the shorter runtime option. The pendulum amplitude is more than +/-4 degrees now.
@@stevesclocks Hi Steve thankyou for your input it is a great design and appreciate your efforts as my wife loves clocks.
I have just put some lithium grease to help the gears on the pinions also to help it out.
The winding drum I was mentioning wasn't to change the gearing just the diameter of what the thread winds on as the larger that is the more leverage the weight has to turn the gear, The only issue with that is the gear that winds the clock up gets in the way, so was suggesting if that was smaller it could be done and after all that gear is only to wind the clock.. Merry Christmas :)
I have assembled and dissembled this clock over six times and had my son do it to, we printed two clocks and have the same problem all the gears are not in a bind, when you put it on the wall and attached the weight to it the weight will not make the gears move the clock is locked. I have made open gear clocks on my CNC programed and driven by Arduino and been 3d printing for over 9 years. What could be locking it up.
The most likely problem is what I used to call "sticky gears" and can be seen in this video th-cam.com/video/pnpYs-SuytI/w-d-xo.html
The issue is that a tiny amount of sideways thrust against the escapement can stall the clock. New helical gears were released at MyMiniFactory to help push gear 3 away from the escapement. The issue can be easily tested by manually pushing gear 3 away from the escapement. If the escapement starts rotating again, then grab a fresh copy of the new gears and print just the 2 new ones.
If this doesn't work, then more info is needed to help debug the issue. Does everything spin with the pallet removed? How about with the pallet and the escapement removed? Also, what runtime option and drive weight are you using? I have switched my clock to a 14 day runtime with around 10 lbs of weight. 32 day mode runs, but it is just too sensitive. 14 days is much more stable. 8 or 10 days would be even better.
I changed the gears and cleaned the bearings with an ultrasound cleaner still only 5 minutes.
@@normhuntley5552 Maybe a bit more information is needed. Is it 5 minutes of pendulum free-swing time? The clock should still be made to run with a bit of extra drive weight if this is the case.
Or is it 5 minutes of clock run time? If so, what are the symptoms when the clock stops? Does the escapement turn when the pendulum is manually moved back and forth? Check the position of the tick and tock when doing this. They should be approximately symmetrical in both directions.
It was 5 minutes of free swing your video says 10 minutes; I have the clock running now with 9 pounds of weight the only thing I did was shake it before I hung it on the wall, I figured since I had it flat on my bench maybe the gears were the problem it works.
@@normhuntley5552 Cool. Things may loosen up a bit and it could be running even better after a few days.
Steve, I am confused, when I look at your website but there is no way do download the SDL file so I can build one.
The STL files are at www.myminifactory.com/users/StevePeterson
You can download the detailed assembly guide from my web site if you want to see what is involved in each clock before purchasing the design from MyMiniFactory. I believe the cost is incredibly small relative to the hours put into each design and the assembly guide.
Steve I just purchased your ez build clock plans. my question is about .0625 [1/16] music wire, where can you buy 1/16" straight music wire, thanks
In my area of the US, music wire can be found at Ace Hardware or many smaller hardware stores that carry larger selections than the big box stores. Remote control hobby stores also carry it. The name is sometimes labeled as piano wire or spring steel.
Most of the clocks can also use regular steel or stainless steel wire if that is all you can find. Try to find something with reasonable stiffness.
I built the small ez build clock and the gears wont rotate with the weight on. Pendulum swings only for a couple minutes. With the escapement removed and a smaller weight attached the gears wont rotate by themselves, however if i start the spinning it will spin for a minute or so. With the clock on the table the gears will spin for a moment or two once I spin the gears by hand. Taking it apart now but not sure what to look for.
The symptoms sound like what I used to call "sticky gears", but "sticky escapement" is a more accurate description. The escapement does not rotate on its own when the pendulum is manually moved from side to side. I believe it is caused by gear 3 tilting slightly and applying sideways pressure against the escapement. It only takes a small amount of pressure to stall the lightly loaded escapement. Push gear 3 away from the escapement to see if the clock starts running again, at least for a short period of time.
I am getting ready to release a new set of gears that should solve the problem. It should be ready in 1 or 2 days. Gears 3 and 4 were re-designed to add a gentle helical angle that pushes gear 3 away from the escapement. This issue only affects the easy-build clocks as they are the only designs with two gears sharing the same shaft.
Brian, I updated gears 3 and 4 on MyMiniFactory with the fixes. Try downloading and re-printing these two gears to see if your clock runs better. It takes a few days for the files to make their way into the zip file, so you may need to download them manually.
@@stevesclocks Thx a lot. I did as you said in your first reply and got a bit more out of it. Reprinted gear 5 as it didnt print perfect and got it to run a couple minutes. So hopefully the new gears will work better. Thanks again.
@@stevesclocks Hi Steve, I downloaded and printed the 2 gears and it didnt help. I reprinted some other gears as well. When testing without the escape and pendulum, the gears rotate well with the weight attached. However, stopping the gears from turning and releasing, i find there is one spot the it stalls, trying to isolate the problem but not sure how to diagnose it. Any hints would be appreciated. Thx.
@@briant444 It sounds like there is an issue that is specific to your printing if the gears stick in one spot. It could be elephants foot, a blob on the gears, out of round gears, a bent arbor, or something else. Maybe try taking gears out one by one to see if you can feel any difference in the friction.
beautiful and mesmerizing!! hopping to start printing soon.
my machine is a homers tevo tarantula (235x235x250mm) do you think is enough to print this?.
thanks
Thanks. Yes, it will easily fit on that printer. I design on a 250x210 Prusa, but also make sure it will fit on a 220x220 Ender3. Height is never an issue.
@@stevesclocks thanks for your quick replay.
i am ordering that gold filament now.
wish me luck
Steve, I just finished printing and assembling this but I can't for the life of me get the escapement to move on it's own. Everything is nice and loose, the bearings are good, and pendulum has almost 10 minutes of swing on it. I can pull really hard on the weight cord and still the escapement doesn't move. If I move the winding drum myself then the clock ticks nicely. It seems to me as though the drum that the cord is wound around should be much bigger in diameter, as this would give more 'pull' on it and therefore more turning force, resulting in less weight needed?! I would appreciate your thoughts on why I can't get my clock to move on it's own, thank you.
My best guess is the "sticky gear" problem described in this video th-cam.com/video/pnpYs-SuytI/w-d-xo.html Adding extra weight only jams against the escapement harder. There are new copies of gear 3 and gear 4 posted to MyMiniFactory to solve the problem.
Making the drum larger would provide more force to the escapement and the clock should become more reliable. It would also reduce the runtime. There are a few gear combinations that provide the same result of increasing power to the escapement while reducing runtime by the same proportion.
Each clock design is a tradeoff between runtime and reliability. I prefer a long runtime. If I had a clock than needed to be wound every 12 hours, it would very quickly become a static work of art. :)
@@stevesclocks thanks for the reply, and sorry for the delay in getting back to you. I've now got the clock running nicely. I already had the updated gears on my clock, but I underestimated just how heavy the weight needed to be. More seriously, I found that one of the spacers was slightly too long preventing it's associated gear from turning as easily as it should. I trimmed it down and the clock started working. I then noticed that the hands weren't moving! So I swapped out the 10 day gears for the 7.8 day ones and it's been running successfully for the last 2 or 3 days 🙂. I really don't mind winding the clock every few days as long as it's actually working as it should!
@@trance_trousers Thanks for replying back with successful results. There are so many interactions that can cause a clock to stop running. It is a good feeling when you discover them on your own.
Winding once a week is tolerable. Designs that need to be wound twice a day would quickly become static pieces of art.
@@stevesclocks yes, that's why I chose your clock design as it seemed to have the longest run times of any I've seen. Just curious, but would you be ok with someone selling your clocks (fully printed and assembled) on somewhere like Etsy?
@@trance_trousers Unfortunately no for several reasons. The files are released under a strict non-commercial, personal use only license. It is OK to give them to family and friends, but they are not allowed to be sold. The other important reason is that the designs are intended for each user to build their own clock and experience the joy of making it work.
is the 32 day version of clock accurate?
It is typically accurate to around a minute per week when it is in a house with heating and air conditioning.
Hello sir.. thanks for sharing your wonderful design... I just recently printed your smaller 32 day clock.. while I can get the gear assemblies to work flawlessly with little to no friction, I cannot say the same for the pendulum..
The pendulum is down to a half degree in about 4 minutes.. I've tried several different bearings, as well as cleaned, dried, PTFE dry lube, no lube at all... I have even gone the extra step and removed the phenolic retainers and still seeing around a 4 minute complete degradation..
I've made sure the arbor isn't to long as well as I thought that could be the issue..
Is there any advice you could give me to increase degradation times?..
I thank you for your time Steve..
There should be some end play in the pallet as well as the arbor. A small amount of frame warping could pinch the pallet. You probably already checked that. When you say you removed the phenolic retainer, do you mean the internal ball retainer? The balls may fall out if you do that.
The next thing to check would be the bearings. They should spin freely for a few seconds when placed over a 3mm shaft. Cheap no-name bearings from AliExpress or eBay typically have around 5 seconds of free spin time after cleaning. They work great in my clocks with a pendulum swing test of 15-20 minutes. Even sealed bearings with the original factory grease intact run almost 6 minutes in the swing test.
I have not tried expensive bearings in my clocks. Super tight tolerance bearings might be needed to run at 100K RPMs, but loose tolerance (cheap) bearings seem to work just fine at the speeds of a clock. I wonder if you got a bad batch. Do they feel gritty?
One user reported better results by opening up the bearing mounting holes slightly. It seems that the bearing holes were tight and the bearings were mounted at a slight angle. Enlarging the holes allows the bearings to float a bit and not apply any side pressure when they rotate.
Another thing to try is the 8 day runtime set of gears. This should give 4X as much power to the escapement as the 32 day option.
Steve
@@stevesclocks thank you for the quick reply Steve .
Yes I removed the phenolic ball retainer in a last ditch effort to reduce as much friction as possible.. and yes if you are not careful the bearing will fall apart.. I've since removed those as I still wasn't getting optimum pendulum swing time..
Also what I wrote about the degradation can be confusing, apologize for that.. should have been more specific..
A 5* swing will degrade to 1* within 4 mins... Complete stop at 7 mins.. so upon further inspection, I did find the bearings I purchased are filled with something similar to dielectric grease.. The alcohol and laquer thinner is just not pulling it all out... Will grab some brake cleaner from my shop tomorrow and try that... But it surely seems silicone based..
While I understand the 8 day gear set will run "easier" I'm determined to make this clock run for 30 days, a personal challenge I suppose..
As a side note, I was wondering.. if the winding drum was increased in diameter behind gear 9 (because there is space there) wouldn't the amount of weight needed to run the clock be reduced from the increased leverage applied?
@@camaroboy383 I use low odor mineral spirits to dissolve the grease. It is fairly weak compared to the old stuff, but will eventually loosen it enough that a blast of air gets rid of most of it. Brake cleaner should work better.
Yes, increasing the winding drum could reduce the drive weight needed to run the clock. It would also reduce the run time. Changing two gears gives a bigger range of run time vs drive weight options. If both changes were used, it would be a bit like the gears on a 21 speed bike. :)
Steve, where would I start in designing my own clocks to 3D print?
The first step is to build one from plans and study how it works. Break the design into small pieces that are easier to understand. The escapement is the starting point. Get it working first. Then figure out gearing that looks good. There are lots of books, mostly written about traditional brass clock design, but still plenty of good info in them.
@@stevesclocks Thanks for the reply. I've finished printing all the smaller parts from your revised pendulum clock.
Do you think a wooden clock design book would translate to 3d printing?
@@Noold Most of the features are similar. Wooden clock are typically larger, so it will need to be scaled. 3D printing allows additional optimizations like printing the pinions integrated with the main gear. Frame pieces can also be combined into single prints.
Very promising kit, thank you! What maintenance should we expect? I also see, that is prone to collect dust from air. Is there a proper way to clean it/keep it dust-free? (While this exposed design is very sexy, an encapsulated one might also be interested.)
No maintenance is required except possibly blowing off any accumulated dust. The only lubrication recommended is a dry Teflon type that doesn't attract dust. My oldest clock is over two years old and still running strong. A case could be built, but due to the large size, it might be easier to build one out of wood.
Hi / just printing this clock but have built you other upright model clock / Have found it easier as regards the bob weight to do away with the suspend wheel for the 2 cord and just hang the weight on a single cord thus cutting down on the size and weight of the printed shell case, Clock runs well with 3.8 lbs and drops about 1 inch an hour / Your comments on me doing this I would like.?
Hi Janet, A single cord works just fine with the obvious limitation that the clock only runs half as long using half as much weight. The weight distribution is slightly off on the vertical clocks and the frame may want to tilt. This is easily fixed with a small nail. The pendulum bob acts as a counterbalance on the horizontal clocks and they seem to stay oriented better. A minor feature of the weight shell pulley is to help keep the cord near the center of the barrel when winding. Watch the cord position when winding to make sure it doesn't go over the edge of the barrel.
Hi Steve, I built two models of your previous 8 day watch. It was a challenge with a lot of problems, but now I can say with great satisfaction that both models have been running great for a few weeks now with 2.7 kg and 3.5 ° of oscillation. In the second model I took liberties and changed the aesthetic appearance of the dial and inserted a bearing in the minute wheel and two bearings in the ratchet wheel.
In both models I have modified the holes for the shafts which are now pushed by force, in this way I do not use spacers.
I use very smooth SAPPORO R2 bearings.
I have now started building this new model and would like to do the 15 days.
I have a question. Do you think that using bushings can reduce friction?
Hi Guido. Bearings can reduce friction, but are not really needed. I placed them near the weight shell to support the heavy loads and at the pallet to support the pendulum. Most of the arbors in the middle don't really need them because the speed is so slow. The escapement might benefit due to the high speed even though the load is very low.
Simple music wire arbors in PLA have very low friction. Bushings used in my first clock have about the same amount of friction. The biggest benefit is that they should last longer. I don't think the friction is reduced enough to justify the extra effort.
15 days of runtime is the upper limit for the large version of the clock, but should be easy in the medium size clock. The biggest factor to achieve a long runtime is the pendulum support bearings.
Steve
Hi Steve, I am very interested in the watch shown. Your video is very insightful. I live in austria and I cannot find a shop nearby. Pleased can you help me, to find an opportunity to find this clock in austria or in the european Union. Thanks very much and best regards from austria, Erich
You can only buy the plans to make your own clock using a 3d printer.
❤😍🔥👍
Previos model working great. This - not. I try repair, but this design without bearings and 1.5mm shaft in rachet - it really - bad design.
One of the biggest comments on the last design was the difficulty of finding the stainless steel tubing for bushings used throughout the clock. This design gets rid of them and the assembly process is considerably easier. However, it is a real clock that needs a little bit of precision to get it to function properly. Bearings are used in the places where they are truly required, the weight shell and the pendulum support.
Music wire typically comes in a hardened state. It is very stiff and more than strong enough to support the ratchet. Music wire is sometimes called piano wire or spring wire. It should be readily available worldwide. Soft steel wire may bend easily and might not work very well.
One builder reported that he used a paper clip by clamping one end and spinning the other end with a drill. It work hardened enough to barely work in the clock. But seriously, find some music wire and it will work much better. Start with the shorter runtimes with smaller weights if needed. Winding a clock once a week is still a very functional clock.
@@stevesclocks
I will reprint , and will add bearings on rachet and 3mm shaft . In addition 3mm shafts must be 91mm (recomendet 89mm).