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


Test Jig #3 - Hours & Minutes Clock

by syvwlch, published

Test Jig #3 - Hours & Minutes Clock by syvwlch May 19, 2011

Featured Thing!

2 Share
Download All Files

Thing Apps Enabled

Order This Printed View All Apps


Liked By

View All

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

15811Views 6972Downloads


This is the current release candidate for the Printable Clock Project. Many thanks to the many Thingiversians who have contributed over the last few weeks!

It is a work in progress and probably will need tweaking to even tick, let alone keep time. It currently has breakage issues in PLA, ABS is recommended.

Test Jig #2 was a major milestone, passing all four tests, featuring in the first video of a ticking printed escapement for this project ( http://www.prototribe.net/vidplay/testjig2.html )and serving as a development base for improvements from multiple Makers. If you want to print something that ticks, print Test Jig #2 with all the latest parts.

Test Jig #3 is intended to regroup all that progress into a new base camp for what we hope is the final push: an actual printed clock that keeps time. If you want to spend plastic and time in support of that goal, print Test Jig #3 and share with us what went right and what went wrong.

This current design is based on feedback from early adopters ssd, rustedrobot and TheRuttMeister, and in particular a very clever rustedrobot design for a ratcheting drum that rewinds the clock: http://www.thingiverse.com/thing:8555 Special thanks to Renosis for rendering gears 02 thru 07 on his machine last night when mine was not up to the challenge!

I've uploaded a snapshot of the code which generated these parts to this page, but the best place to find the latest and greatest is on github:

Major improvements include: the ratcheting drum, countersunk holes for the frame, a snap-fit extendable pendulum, and an escapement design which should tick as printed (little or no hand-finishing required).

I tried to go for a four gear clock, but getting a 60:1 ratio in two gears is just not practical... the pinions get too small. I also went ahead and added hour and minute hands. I can make alternate parts for a hand-less version if the concentric shafts are too much of a problem.


Print one of each STL, keeping track of which gear is which. Since this is a six gear clock, it does not have parts 08 and parts 09. Use four M3 bolts, two to hold the frame, and two to serve as shafts.

The gears are numbered from front to back, starting with the drum (01A and 01B) and finishing with the escapement wheel (07). Gears alternate on either shaft, and the pinions (the smaller gear attached to the center of the larger one) face forward.

The pendulum and the escapement are combined in a single print and snap-fit together (a design idea from Aubenc).

Attach a string by the middle to one of the side holes on the drum, with both lengths wrapped around the drum in the same direction. That way, when one length unwinds as the clock runs, the other gets wound up. Attach a weight (100g, perhaps?) to the length which locks the ratchet against the stops (clockwise from front), and a smaller weight (20g?) to the length which unlocks the ratchet in the opposite direction.

All Apps

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

Print through a distributed network of 3D printing enthusiasts from across the US, at a fraction of the cost of the competitors. We want to change the world for the better through technology, an...

App Info Launch App

Quickly Scale, Mirror or Cut your 3D Models

App Info Launch App

3D Print a wide range of designs with Treatstock. Easy to use tools to get the perfect result. The global 3D printing network that connects you with high-quality and fast working print services nea...

App Info Launch App

I been printing this out over the last week or so in hopes of making it to the clockathon. Syvwlch, this thing is awesome. Reallly nice work and it gets me super excited about learning openSCAD.

I have a lot of thoughts which I'll share in person later today.
Mostly, I agree with TheRuttmeister about geartrain resistance. I noticed that the resistance seemed manageable from the perspective of the minute hand and could be powered from that gear.

As a test I hung a weight (about 500g) fro
m the minute hand gear and it ticked!! You can view it here: http://www.welovephotobooths.com/share/v3-tick.m4vhttp://www.welovephotobooths.c.... Obviously the pendulum length + weight needs to be adjusted.

I'm very confident that if we attach the drum to the minute hand gear we can get this baby to keep time (but only for a few h
ours before winding the weights). I have many other thoughts as well which I'll share later today.

Thanks!!! This rocks.

Awesome! Can't wait to see you at the not cave later today. That is a strong foundation to build on. :)

Ok, feedback.

Does not tick. :'(

Way too much resistance from the geartrain. I've hung quite a lot of weight on it and tried to turn it by hand. I can't generate enough torque either way.

Clearances for shafts generally was a little generous. Was a little loose on M3 so I ran a 3.5mm drill through and used so
me 1/8" brass rod.

Clearances for the concentric shafts was perfect.

Detail generally was too fine. Details like the walls of the concentric shafts and the hands print as just walls, no infill. There are advantages, but it does mean that they are very fragile.

I'm also not sure about spacing betw
een the gears. There might be a better way to do it.


So close! Out of curiosity what were your skeinforge settings related to the width and thickess of the extrusion?

So I've been rushing to get this done (I had wanted to have it done to take to MakerFaire to show off to Bre). All the time expecting to be second... so to satisfy my immature side: FIRST!


W/T settings: Carve 0.3/1.667 so filament of 0.5mm (roughly). I try to limit myself to features of no thinner than 3mm (and even that is pushing things) otherwise SF can't fill them.

Failure point.
My money is on the gear ratio. I can get the escapement to turn by heaving on Gear 3 (the
minute gear). Using that to power the whole clock isn't too silly either, it just means winding twice a day. And to support my gear ratio theory, consider this (as a super rough calculation), if it takes 0.01N (1g) to turn the escapement, the gear ratio being 3,600:1 means that it will take 36N or
3.6kg to turn from the drum end (assuming 100% efficiency, and I'll guess we'll be lucky to be within an order of magnitude).
I suspect that conventional clocks have the drum at a lower point in the gear train, plus a large diameter drum to give more torque.

If I was to suggest a perfect (to me)
test jig to try next it would include:

Thicker walled concentric shafts (probably double the current, maybe even 3 times). Lower profile teeth on the ratchet (it requires a scary amount of force to wind, but I am using PLA). 1 shaft running from the escapement to drive the second hand (probably the
least important for right now). Drum on the minute hand.

I don't think that the friction in the system is an issue (compared to the gearing) and unless the drum is moved back in the system I don't think that an hour clock will work (heck, its just keeping the current problem)
That said about the f
riction, I know ABS doesn't have quite as low a coefficient of friction as PLA so there could be more of an issue with the concentric shafts.

Hum... It is relatively simple to put the drum in the middle of the gear stack, but to put a rewindable drum there will take a little more work. Nothing impossible, tho. If I do it right, we could keep the frame and all other parts the same except for the hour and minute gears.

As for the math on the torque, I beg to quibble: to transfer 1 http://g.cmg.cm of torque to the escapement wheel, the minute gear has to provide 60 http://g.cmg.cm of torque, and the hand gear has to provide 720 http://g.cmg.cm. If the radius of the drum and of the wheel are roughly the same, it takes 720g, not 3,600g for each gr
am delivered to the escapement wheel.

Still a lot, I admit...

I though I had the maths wrong... that'll teach me to try and figure it out at midnight when I've just gotten into bed.

I wouldn't worry too much about the rewindable aspect right now (I think thats trying to fly before we can crawl ;) )
Or at least just include the ratchet drum and I'll poke my fingers in and turn it from the side! Those teeth are way too tall for PLA though :(

Just nobody beat me to a ticking clock!


We'll get there. Thank you for printing this one out!

Sounds like me might need to try metal bushings. I was thinking that a cheap telescopic pointer/antenna might provide some thin metal tubing that could be slipped between shaft and gear and reduce the friction.

Out of curiosity, which gear would you say the clock could be driven from, based on tr
ying to turn each one by hand?

One thing i'd still like to try is an hour hand only clock. I'm still suspicious of the concentric shafts and it might make a nice step that fits in between Test Jig #2 and Test Jig #3 as it would have less gears overall (i think).

Well, the usual suspects (with some overlap, of course) are:

  1. friction

  2. gear ratio

  3. concentric shafts

Because of 2, I think we might want to shoot for a minute clock instead of an hour clock.

Works for me, incremental is good. Plus with a 60 metre pendulum we would technically have an hour clock!!! :-P

Maybe a huge drum would work, with scaled down gears (although I doubt they would be printable on a cupcake). Plus the drum would need to be way bigger than the cupcake build area.

Looks cool though.

Finally, ready to start on this. One question, can I reuse any of the bits from test jig #2?

I'm guessing no, unless the escapement is the same. Everything else looks different.

Afraid not, every part is new.

Counter-intuitively, you can make the distance between the shafts larger when the gear ratios are lower, and every other radial dimension follows from that, including for the escapement. Since it's important to print this as large as possible, each clock script has a different, maximized value for
that distance.

Fantastic work syvwlch! I'm going to make this with SLS very shortly. I'll try a "pocketwatch" version also.

syvwlch - as as exercise in sheer lunacy, I was wondering whether you'd oblige a request; Ccould you please output these files positioned "as is" from the raw Openscad output? I'd love to attempt to make this clock as a "single part" assembled straight from the machine. I see no reason why it woul
dn't be possible and it would be a fascinating exercise in the limits of Rapid Prototyping.

Again, brilliant stuff, can't wait to see what you come up with next.

Thank you :-)

I suppose I could make a variant that imports and positions the existing STLs, with slight mods to the frame and shafts... Trying to render the entire clock directly would probably take hours! It's also going to be a VERY big STL, I would imagine.

Lastly, how big is your printing area, and what k
inda mechanical clearance would you recommend for the gears and for the shafts? Right now, this is sized such that the largest parts barely exceed 80x80x80 mm, the clearance between a shaft and a cylinder able to rotate freely around it is 0.5 mm, and the clearance for a snug snap-fit is 0.25mm. The
clearances may need to be reduced if SLS is much more precise, no?

Damn, looks like the dreaded bald gears struck on part # 04.

The pinion (small gear) has no teeth, as sometimes happens with the latest version of OpenSCAD and the MCAD library.

Dirty socks and kitten farts, I hate it when that happens!

Ok, found a temporary workaround for that OpenSCAD/MCAD bug, uploaded a fixed version of part 04 and deleted the old version, which was non-functional.

Please use the fixed version!

See below for a JPEG of the two versions. Red is defective.

Also got a toothless part 3.

Going to make it hard to get it ticking =-O

Crap, missed that one too. Lemme generate a fixed version.

Ok, found a different workaround because my first one was not working for part 3.

Followed suggestion here: https://github.com/elmom/MCAD/issues/7https://github.com/elmom/MCAD/...

Remaking parts 3 and 4 just to be safe.

Part 4 printed fine...

Ok, I just uploaded a new part 03 and will leave it at that, then. Sorry for the miss, that bug is annoying.

Happy printing!

A. Woohoo!

B. Dangit, going to be busy all day today putting an old car back together, no printing till the evening!

C. How thick is the frame? Because its hard to find m3 bolts longer that 50-55mm.

D. Woohoo!

A. woot!

B. sweet, sweet anticipation.

C. The frame is significantly thicker. I slimmed down the individual gears a smidge, but there are a lot more of them. See the JPEG below for a side-by-side, where red is Test Jig #3.

D. woot!

Looks like about twice the thickness, overall.

Might have trouble bolting it together then. Will have to see if I can get some m3 allthread maybe. Hum.

You know, Bre Pettis just pointed out that, in a pinch, anything straight and round would do.

Afterall, the two frame halves can be glued together, with something to guarantee alignment while the adhesive sets, and the gear shafts would work better without threads!

Specifically, he was gonna raid a closet for coat hangers. :-)

This is quite a bit thicker than Test Jig #2, which only had the drum and the escapement wheel. The frame may not be rigid enough as is, and may need additional front-to-back bracing...

Very exciting. Now I've got more to print out. Initial test of the new ratchet in PLA seemed much better, still requires a bit more force to turn then I'd like but turns with no breakage.

I could make the ratchet teeth shallower. As long as the ratchet arms are solidly held in place when the drum is locked, they would have less far to bend to get over the teeth going the other way. 75% might be a good starting value.

Or I could make the teeth shallower AND the arms thinner.