To test this feature, visit your live site.

Is the smaller crazy gear clock just a scaled Sp10?

I would like to build the crazy gear clock but I really like the mini version as opposed to the medium or large one. Can I buy the SP10 plans and just scale them down? If so, what scale would you recommened to replicate the one you show in your videos? Thanks!

19 comments

19 Comments

Stu Harrison

Dec 22, 2024

The frames were converted to solids (using fusion)--just since it was easier to manipulate them this way. The gears were made using a free fusion add in--called spur gear. I fine tuned the add in parameters so the gears matched the size and shape of the crazy gear then tinkered with them for the interior bracing, inner gears, thicknesses etc. Once the gear is generated, it is very easy to work with. I can send the files to Steve if he wants them and he can include them--but they are not drop in compatible with the rest of the model...

And I agree about the hands--but truthfully i like the gears moving more than its actual use as a clock :)

ksmietanski

Dec 23, 2024

Replying to

Hi Steve,

Looked for a new post by sosalty62 but probably he has not yet created it.

Steve, would this problem also manifest itself in SP8. I have printed a number of SP8s and they work really well for the first while and then slowly the pendulum amplitude decreases and they stop.

I do not understand why the clock works so well for the first week or so and then the performance deteriorates over time. Why does the problem not show itself immediately?

I have printed all of my SP8s in silk at a layer height of 0.16mm. Could I combat this problem by printing in normal PLA or larger layer height??

Krys S

Steve

Dec 23, 2024

Replying to

Hi Krys,

I have noticed several clocks that appeared to operate perfectly until the performance degrades after a month or two. I first noticed it on the 32 day easy build clock. After a few months I had to reduce the runtime to keep it running. Traditional brass clocks run for 10-20 years before slowing down due to the oil thickening. Something is obviously different with PLA.

I do not know if PLA changes over time. All my clocks use silk PLA. It looks way too good to switch to anything else. My suspicion is that the worst friction is engaging friction that occurs when a new tooth starts making contact. It is especially bad when this contact occurs before the line of centers and the new tooth pushes into the other gear. Adding a very light layer of grease to the pinion teeth seems to resolve the problem for a while at least.

Normal PLA might be better. Or it might be worse. I have no idea what characteristics to look for when ordering new spools. I had a roll of semi translucent PLA that seemed like hard plastic instead of the sticky feeling that silk PLA has. It made great looking gears.

Steve

ksmietanski

Jan 10

Replying to

Hi Steve,

How did your test go?

Did the newer gear profile make much of a difference in friction/runtime??

Krys S

sosalty62

Dec 22, 2024

Wow! Nice work. It looks nice. That sounds like a lot of Fusion work. I wish Fusion handled STL better. I imagine you used the STL's just as a reference and completely created all new gear solids? Or were you able to convert the STL's into usable solid objects? Just curious, because I just went through this, and it would be cool to learn other STL to Solids tricks.

If you are open to any suggestions, I think I would go with a different color for the hands, as the black ones seem to disappear within the sea of black frame and black numbers.

Cheers!

Stu Harrison

Dec 21, 2024

So I did manage to scale this one down. here was my process if anyone wants to follow:

assemble the entire clock in fusion 360 then scale the whole model by 75%
redrew all the gears since I liked the look of chunky clock gears versus the more organic gears that were included. The shaft dia. is fixed in fusion at 2.4mm accounting for some overextrusion a bit. The gear modulus I used was 3 I think. I used the original gears as the reference to give the outer and inner gears the same orientations and to get the thicknesses just right.
the front, back and lower base were also increased by 1% in the x/y direction to give the new gears a bit more play. The base was then stretched in the z direction back to the original height to accomodate the electronics.
the base was also remodelled a bit to make room for the nema17 motor--I could have stretched the legs but I had them printed already and didn't want to redo it :)

Steve

Dec 22, 2024

Replying to

That is super cool. The crazy gear style in the original design requires a relatively large pitch or else it starts leaving gaps and losing all its strength. Your gear style is much better for scaling.

Stu Harrison

Dec 12, 2024

Ok Thanks--I might give it a go and see how it goes--I only have nema17 motors on hand so that might be the fly in the ointment.

Steve

Dec 11, 2024

It's not a perfect scale. The gears may have a 75% scale, so a large gear that printed perfectly with 4 perimeters will print smaller with 3 perimeters. However, gears with minimal clearance around the motor would collide unless the motor was also scaled down to a NEMA 14 or smaller. The designs share many features, but they were each optimized separately without using a linear scale factor.