Nautilus 608

by David_L_G Apr 30, 2017
Download All Files

Thing Apps Enabled

Please Login to Comment

Any chance of getting a 607 bearing version?

Great design! I found that the single bar makes it somewhat unstable, so I added a second identical bar to the back and printed two small cylinders to stick them as pins in the bar holes and hold them together. It took me a little trial end error to get the right diameter (a precision caliper gives uneven readings for the bar holes) but 4.5 mm in diameter by 12 mm height works; it's tight enough and can be pused without much effort.

OK, I did a remix, if you are interested in a wider, thicker or parametric set of connectors. I did the connectors in openSCAD, so you can download openSCAD and modify any particulars of the connector bar. I have included STLs for nut and screw side connectors for either a 2 nautilus bar or a 4 nautilus bar, using standard 3mm hardware.

OK, I printed the middle version with a brim and it was hard to get working - the damn thing was also way tight on the 608 bearing, so I used a little heat (from a hot air SMD rework station) to fit it. The brim required a lot of fitting, cutting and filing and when all was said and done, the mechanism had an asymmetrical lump in it, and I had to use enough heat in fitting the bearing that I thought it might have distorted things.

I tried again, this time I used the original file, and I tried to press fit a 608 into it - snap - the ring around the 608 snapped. It still kinds works.

So I printed an extra "shell" and used just a little heat - and I quit printing the brim, and I mirror flipped one of the shells (to make any printing issues kind of offset) - and it was perfect - just a little heat to fit the bearing, they mesh well and the only rough spot is where the shells shift long to short - and only if I pull the shells in a certain way. That is right, I still had to heat fit the bearing with the largest version.

So, I let my wife play with them for about a minute and she handed it back - it had fallen apart. No idea how to keep the "bones" fit - except that I got a 3mm screw, 16mm long, with a nylon insert nut.

All that said, these are not pleasant to play with. Every edge is sharp. gear teeth, the 3mm screw shafts are too long at 16mm and too short at 12 mm.

And my wife said, "Look, these things need a bearing cap like a spinner - thicker and inset. Then you could use one of those threaded inserts and completely countersink the hex bolt head, and the 16mm screws would fit. Then you would have a place to rest your fingers that was not either rough or sharp."

I have played with one of these for a couple hours and, wow, it has made my fingertips sore - I want everything rounded - except I know that rounding bottom edges is hard to print, and rounding just top edges would make it impossible to flip the shell - then I wondered, what would this be like with a bit more inertia? How about printing this with every other chamber solid? It is interesting how it has an asymmetrical period when used as a pendulum, especially across the break. A heavier "shell" might make that more noticeable.

So I am sure that I can use a caliper and figure out what the center distance is on the "dog bone" and whip something up in openSCAD that would be at least somewhat parametric and would allow me to have the room for an insert and to countersink the hex bolt head. Or maybe someone can do a dog bone that locks together without any hardware, with larger fingerpads. (OK, I have an openSCAD model for the dogbone with fingerpads and countersinks and chamfering and rounding. It was a pain, since I used Minkowski sums to do some of the rounding and chamfering, but it is done, I am printing the first trial bones now. If you use the bones I would want to put up the openSCAD source as well so that it would be easy for people to modify to their needs as well as whatever STL files you or I settle on. I presumed 3mm hardware and made one version that allows for trapped hex nuts while the other version presumes round heads like on an allen head screw, so there would need to be two STL files for the bones (for people who don't want to use openSCAD.)

Would you prefer to do the honors yourself, or have me send you my "bones" once I am 100% sure the stuff is right and send it to you, or should I release a remix?


First of all : WAW! You wanted these real bad !

Please, release a remix ! This model i got from someone else, who got it from someone else's maths ..... let's create a magnificient world together !
Personally i didn't like these Nautilus gears without bearings because of the rough spots, i released a remix, please feel free to do the same if these dogbones are **** to you !
I'll see if i can work that sharp edge problem. Im using fusion 360 and the last time it crashed when trying to fillet these.

Anyway, thanks for writing all your story, it's great that you didn't give up.

EDIT : also i didn't want the pads to be larger than the inner ring as i think the beauty in it is is mechanics, so having the ball bearings appearent was one of my criteria.

Aha! David, I didn't try the second smaller file. I just printed the 0.15 undersized spiral at 98.5%. This made the bearing tighter and loosened the gear mesh. Now it spins like a champ. I left the 'dog bone' sides at 100%.

Really ? no hard spots at all ?

I was having a closer look at my CAD file and i thought/felt the ' dog bones' would need to be shorter (like .5mm shorter) to get things smooth ... you seem to be saying the opposite, i'll have to give both techniques a try ..

Thanks for the interest & participation anyway ! I appreciate

I printed the smaller dia gear and it is much better, however still a little to large and the bearing still won't fit snugly. I measured my 608 bearing OD to be 22.05 mm and your inner hole to be 22.22 mm. I'm printing with HTPLA that usually prints bearing pockets just fine on other spinners and fidgets. I think another 0.1 off of the radius would do it.

Well, here comes the extra tight version then !

Fusion said diameter is 22.15, so it should print 21.95 - 22.05, i hope.
Please don't give up now :D

Love it. But bearings are too loose. I had to wrap with tape to make it work. I suggest smaller bearing hole by 0.1 mm radius or more.

I also had to trim the teeth at the ends of the spiral. They were hitting.

Hey thanks a lot for your feedback, i just uploaded a version of the gear that is .15 mm smaller in diameter.

Maybe that one will work ok for you ?

Very interesting design! I like it!

Original conception files can be found here, if you want to have a look :