Hey! This thing is still a Work in Progress. Files, instructions, and other stuff might change!


Test Jig from Clockathon #2 - FIXED

by syvwlch, published

Test Jig from Clockathon #2 - FIXED by syvwlch Feb 6, 2012

Featured Thing!

1 Share
Download All Files

Thing Apps Enabled

Order This Printed View All Apps


Use This Project

Give a Shout Out

If you print this Thing and display it in public proudly give attribution by printing and displaying this tag.

Print Thing Tag


Thing Statistics

12040Views 5328Downloads


Fixed a bug on the hour gear. Download the fixed STL and if you are interested in it, all three OpenSCAD files as well.

This is a cleaned up version of the printed clock we were able to make during Clockathon #2 last weekend. Thank you Makerbot and Bre for setting that up and hosting it, and thank you Ben for all the help, the awesome test print all set up, and the brilliant idea to back off one gear for the power drum.

It was ticking, powered by a weight that would only drop about 9" per twelve hour period, and included printed concentric shafts to support minute and hour hands on the face of the clock.

Here's some video: http://www.youtube.com/watch?v=AH4Tg2tIgfg
I had to inflict grievous harm to the code base at the time and have cleaned everything up since. See details below.


From the makerbot clockmaker google group:

So I cleaned up the code from clockathon 2, and implemented the changes discussed.

  1. Changed the gearing to add another factor of two to account for the fifteen tooth escapement wheel. Still six gears total.
  2. Deprecated the old drumgear routine that could only be used on the front of the clock and replaced it with the core pinion wheel routine, which can now include a non ratcheting drum on any gear, and can now produce a gear with no pinion if you set the number of teeth on the pinion to zero.
  3. Implemented some logic to use one shaft radius (bearingRadius) if the gear has no concentric shaft, and another (pinRadius) if it does. This way, the design automatically assumes you place bearings on all the gears that do not include a concentric shaft.
  4. Changed the parameters to include concentric shafts and hands for the hour gear and the minute gear.

With these changes, it is now possible to generate the same clock we had at the end of clockathon 2, without touching the code or the parameters incessantly. What it currently does NOT do is include a ratcheting drum solution for any other gear than the front one, or a way to turn the fixed drum on for only one gear (right now, you just turn it on while rendering the gear(s) you want with drums, and turn it off for the rest).

The standard config right now is bearings on all gears except hours and minutes, concentric printed shafts resting directly on M5 threaded rod (had to guess at a good radius value there) for the two exceptions, and a 5mm wide non-ratcheting drum on the gear between hours and minutes.

Today's respite from business travel does not extend to tomorrow, unfortunately, so I will post this code base later today, with some STLs and some screen grabs... And you guys can give it a spin. Ha! See what I did there? ;-)

All Apps

3D Print your file with 3D Hubs, the world’s largest online marketplace for 3D printing services.

App Info Launch App

Auto-magically prepare your 3D models for 3D printing. A cloud based 3D models Preparing and Healing solution for 3D Printing, MakePrintable provides features for model repairing, wall thickness...

App Info Launch App

Kiri:Moto is an integrated cloud-based slicer and tool-path generator for 3D Printing, CAM / CNC and Laser cutting. *** 3D printing mode provides model slicing and GCode output using built-in...

App Info Launch App
KiriMoto Thing App

With 3D Slash, you can edit 3d models like a stonecutter. A unique interface: as fun as a building game! The perfect tool for non-designers and children to create in 3D.

App Info Launch App

If there's a way to send you a private note I haven't figured it out yet. (I only came to thingiverse the first time yesterday.)

Your early designs are simple enough to guess (either from the text or the gif's) that you were thinking of using a 60:1 ratio of minutes to hours. I would have expected 12:1 to be required but I don't see a specific mention in the later evolutions. If 60:1 is indeed correct, can you take a moment to explain how that works?

I cannot tell you how long I have been looking for a mechanical clock to 3D print ot laser cut that actually works that my students and I can use to learn how a clock works...how difficult is this to assemble once parts are printed?

Not hard at all, but as we don't have a printed case and the parts aren't numbered or labelled, it still takes a little extra DIY. Mostly we've been mounting everything on a board, with a printed box to hold up the shafts.

When we finalize the design, we'll have a complete set with labelled parts and a case to print. Meanwhile, it's still a little beta. ;-)

Have you found a bearing that works? I've been browsing my bearing catalogs but an 8mm OD 5mm ID bearing is mighty rare.

We switched to those bearings when Ben opted for M5 rod, the M3 having a tendency to bend under the weight on the drum. Ben found these bearings and we haven't looked back... Although we are thinking of standardizing to something easier to find.

I've asked Ben, and the rest of the merry crew on the Makerbot clock maker google group, for the source of these bearings. They should get back to us soon. If you want to drop in and say hi, join the group at http://groups.google.com/group/makerbot-clock-makershttp://groups.google.com/group... .

Alternatively, we could branch th
e clock and have a different version with different hardware... But since it takes time to generate the STLs, it might lag a bit behind on the bleeding edge.

Ok, in my own research the cheapest I found was a pack of ten 5x8x2.5 sealed bearings for $14.95 at http://vbx.comvbx.com.


These definitely look like the ones Ben has been using and bringing to the Clockathons... And that sounds like the price he was quoting.

Once this drum thing is put to bed, we have got to make a version of the clock with cheaper hardware... Like
homemade bushings cut from some tubing. We've been discussing it in the google group.

I live not far from a huge manufacturing center, and I asked around at the industrial bearing supply shops. I got pretty much laughed at, but I did take a look at a really nice tapered roller bearing with a 5000mm ID!!! It was a steal at $122,000.

I ended up ordering a couple sets from http://vbx.comvbx.com, thanks for the link.

Published a ratcheting replacement for part #2, the intermediate gear that carries the drum powering the clock. Should make it easier to rewind the clock without taking it apart.


Racheting Drum Intermediate Gear for Clockathon #2 Test Jig - FIXED
by syvwlch

And... already published a fix for the replacement part. Sorry about that guys...

The hole in the escape wheel doesn't go all the way through. I'm assuming it's supposed to??


Yes, it should go all the way thru. I'll take a look at the code, fix it and republish the STL. In the mean time, since that part rides on bearings and the blockage is halfway down the hole, you can probably just drill a hole for the shaft, put bearings on either side and call it a day. :-)

Is there a need for the thin little shoulder on the bottom of the Hour Gear? It means I have to print with support, and makes a lot more cleanup. Thanks.

Hum... There's something wrong with that part. There shouldn't be that cylinder sticking out the bottom and the hole should go all the way thru. Must be something wrong with my code in OpenSCAD.

I'll rerender the hour gear and post a new STL. Sorry about that!

Ah ha! I have all the pieces in front of me, and thought 'why doesn't this gear have a through hole??'.


Yeah, redid the code on that one since the clockathon. It looked right in F5 view but apparently it rendered weird in F6 and I didn't notice when I exported the STL... You can try to drill it out but I will make a new STL this weekend. Been traveling for work and openSCAD doesn't work on my laptop. :-(

No sweat, glad I could bring it to your attention. There's also a 1mm (or so) protrusion on the bottom that actually renders below the build surface.

Ok, fixed it. Turns out, when you difference a cylinder from another cylinder with a radius of zero, the result is invisible in F5 view, but renders as the first cylinder in F6 view and in your STL exports... Who knew? :-)

Moved the errant object into a error catching routine and it all seems fixed now. This is almost reaching the point where I need to open a bug tracking tool for this thing! :-)

Anyhoo, please download the fixed STL file for the hour gear, and the fixed code if you are interested, and let me know
if that works better. :-P

Thanks for the head's up, and for your patience.

Silly questions time!

I don't see any place to attach the weight to the clock when I render it in OpenSCAD? I'm guessing it has to do with some setting that you need to enable for the escapement wheel?

I guess the biggest challange was getting the escapement to work right and reliabily?

syvwlch - in reply to Daid

No such thing as a silly question. :-)

The weight driving the clock goes on a string wound on the drum built into gear number 2, between hour and minute gears. There is a series of holes in the side of the drum to tie off the string.

The pendulum weight needs to be added to the stub sticking out of the escapement (not the escapement w
heel), on the end of something long and thin like threaded rod or a ruler.

Daid - in reply to syvwlch

Oh, now I see where you put the string... that's one small guide. When I generate a full clock with the OpenSCAD code, all gears have that guide?

This looks like a really interresting print. But I currently only have 608 bearing which might be a bit large...

syvwlch - in reply to Daid

The guide is always there, and it's length along the shaft is controlled by a parameter called spacer. I would recommend leaving it at one millimeter when rendering all the gears except the one you want to put the weight on, and increasing it to something like five or ten millimeters for that one.

There is currently no way to set that value separately for each gear, so in the assembled view, all the gears will always have the same size guide.

What are the outer and inner diameters on a 608?

Oh, in the STLs posted here, I already did render gear number two with a higher value of the spacer parameter...

Ah, indeed, gear2 looks better suited for a weight.

608 bearings, they have an inner diameter of 8mm, and an outer diameter of 22mm. And a width of 7mm. Yes, they are pretty large for a clock :P

This is one of the more awesome projects on Thingiverse. Great job. I see you had to stick a few bearing in it? I guess it has to much friction without bearings?

syvwlch - in reply to Daid

Actually, we're starting to think that, except for the drum holding the weight that powers the clock, radial loads are low enough that bearings aren't strictly necessary... But they do prevent wear, eliminate some slop and eliminated friction as a variable during the testing.

We may publish a no-bearings version once the clock is finished!

I've been watching the progress of this keenly. This is the first iteration that I've downloaded and plan on printing. Is there not parts number 8
amp; 9?

Keep up the good work!

The part numbering scheme stems from the way the script is written, with the possibility of using zero to eight gears between the escapement and the drum. This version only uses six, so two numbers don't get used.

Make sure you have access to some M5 threaded rod and some bearings that fit on it with an outer diameter of 8mm... Before you print these out. Or some plan to make due without, in true maker fashion!

Once cleaned up... any chance of also posting DXF files for us Laser folk? I know it

would be a bit... but the 3D AND Laser communities would both benefit..


Hum... That would be an interesting challenge! Since we're not using the herringbone gears, it should be possible by stacking cuts, but it would require quite a bit of rework and redesign, I would think.

A stopgap would be to extract dxf slices from the existing STLs, which I believe is not that hard to do in OpenSCAD or can be done by hand in something like Sketchup.

Arg! Of all the bearings and threaded rod I do have, 5mm isn't it for either!

Agh, sorry! I now have a bunch of it. You should be able to get by fine with anything else though. Try not to go much smaller on the bearing OD as there are great benefits from the larger higher quality home made on the bot.

Larger higher quality hole in the gear hub, I assume that was before autocorrect gotta hold of it?

Change two parameters, pinRadius and bearingRadius, to whatever you have, and re-run the scad script once per part, and you should be golden.

I'm gonna wait a little longer to update the library on GitHub, as things are still pretty fluid with Ben and RustedRobot working on various enhancements.