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

Test Jig - Clock - Minutes and Seconds Only

by syvwlch Apr 28, 2011
Download All Files

Thing Apps Enabled

Please Login to Comment

I don't click on "I made one" because I didn't, however... it's funny to see how the jigs look like to the ones in the picture bellow :)

I only wish I had more time

Awesome! Nice print!

I still say that counts as an I Made One... at least of the library if nothing else!

here's the link: http://www.thingiverse.com/thing:7976http://www.thingiverse.com/thi... but no pressure. O:-)

Clockwork Library & Printable Clock Script
by syvwlch
Parametric OpenSCAD Clock Gear Train (Hours, Minutes & Seconds)
by aubenc

You made your own variant of the 4-gear clock with a custom-designed frame and pendulum?

That's awesome!

Any other mods? How does the filament work as a shaft? How does it do with the four tests outlined above? Gah! I have so many questions....

Please tell me you're going to post that as a derivative... at least the frame files? I've been dying for someone to do that. :-)

Pretty p

Nop! It's not a variant, I was just playing a little bit with your scad library. As I said above -dirty way- so, I just print the 4 gears and the escapement wheel (not even cleaned this last one in the pic), and used the pendulum I split in three parts from your 1st scad.

So.... what I did is quite useless because the pendulum escapement doesn't match the size of these wheels (knife :-[).

The filament as shaft won't work neither. It's curled so it will add tons of friction... :'(

I'll do my best ot find the time to take a look to your last version and see if I
can help with the frame but... I cannot promise... I have so many designs pushing to get out of my head...

What/wheres that picture from?

That's a picture I took (today) from some testing I was doing in one of the 1sts derivatives (I don't remember which one... the 1st where the 4 wheels assembly was there). I just did the little gears taller, drawn that kind of support -dirty way- and adjusted the pendulum escapement with a knife :-D

Sweet! 3mm axles! Printing now...

I'm sorry I keep running these out and relying on you to test print them. Thank you again!

Don't be!

My aims aren't totally altruistic, I want a printable clock too! Besides my printer runs nearly non-stop while i'm awake (approaching 15lbs of plastic since I got the printer a few months ago). If not this then i'd be printing something else and most of those things i've been giving away.

Speaking of testing... would it be useful to design a calibration object that would allow different Makers to pick the optimal values for parameters like shaft radius, sleeve thickness, clearance, gear clearance, gear backlash, and the like?

A lot of this design requires some pretty tight tolerances, and each printer made need tweaked settings, I'm thinking... out of my deep experience of Makerbotting, you understand. ;-)

Could be interesting. I've also seen things with recommendations on print settings (like 0 extra shells for gears, etc...).

On a side note I was doing some reading up on gears and saw it mentioned that lower pressure angles are better able to compensate for manufacturing inefficiencies at the expense of weaker teeth.

Additionally it looks like the pressure angle defines the angle of forces against the shaft the gears
are turning around. (Hence why i've been trying to experiment with them). =)

Yep, there are even some people who say that because clockwork transfers little torque, but is very sensitive to friction, involute gears are not the best profile, because the teeth slide past each other for part of the action... creating friction.

Cycloidal gearing is better for this application, and we may have to switch to it before we can scale back up to a full 4-gear clock, let alone the original 8!


I have to disagree with you about cycloidal gearing being better for the application. Here is a nice piece comparing cycloidal to involute gearing:


I agree with msruggles that cycloidal gears probably won't help much. One question I have (think this came up elsewhere) is why is the pressure angle 28 degrees? "Standard" values are 14.5 or 20, and those profiles look much different than a 28 degree profile. I assume 28 degrees is used because it prints better. But it might be worth trying a lower pressure angle.

Lower angles seem to print ok (i've tested down to 20), although the inner pinion could use structural backing. In one experiment I ended up breaking a few teeth. It would also help with making them easier to print as we'd no longer be printing 2-3mm towers. The failure point seems to be in the lamination where the pinions meet the larger gear.

You know, I haven't played with the pressure angle since the first couple designs, and I may very well have left it at a sub-optimal value.

The short explanation is that a smaller pressure angle will produce less bearing load while a larger pressure angle will (generally) give you stronger teeth. 14.5, 20 and 25 are the standard angles, but with a 3D printer you aren't constrained to off the shelf hobbs. Question is; how much time do you want to spend optimizing gear strength and bearing performance for plastic teeth in a low load application?

Hum... I think we'd be very happy to have lighter bearing loads, in this case. Can't wait to hear how rustedrobot's experiments in that direction turn out. :-)

Thank you for that link. I can definitely do without changing the tooth profile if it's not necessary... :-)

And it just so happens that, thanks to mattmoses, we have a public domain library to generate just that kinda profile... it just needs to be wrapped up in a gear generator.


Epitrochoids and Hypotrochoids

As for calibration issues... I'd have to give a plug for my ProfileMaker. http://makerblock.com/profilemaker/http://makerblock.com/profilem... It should allow you to change up your settings without the need for printing calibration objects. Given the small teeth on the various gears, I'd suggest a thread width of 0.4-0.5mm. (This is achievable with a stock MakerBot 0.5mm nozzle).

Nice! I'm sure that will come in handy when I try to get my MakerBot working (psych!).

I was thinking of a calibration print to adjust parameters of the clock script, as in... how small can you make the clearance parameter to eliminate wobble without causing the concentric shafts to seize up. This may not be something everyone has to do, but it might be something a few early adopter
s might need to print, just so we can dial in the design. :-)

Alright, let's test this bad boy. :-)