Ever since I saw the helicopter cube, I've wanted to make my own, I'm new to using CAD software to design puzzles but after looking around the forum I managed to come up with the design(well at least i hope it works). Of course, I still have a lot to learn to make more complicated puzzles.

I have a few questions about actually making the puzzle and about the method I used to design it.

1. How do I get an estimate on how much printing masters will cost?
2. Also, what information do I need to give?
3. About 3d printing, is the following a worthy alternative?
Since at the moment, I can't afford to print 3d parts, I found a program that takes stl files(also other files) and creates a net to make a paper model. Can the paper models be used to make molds?(I don't know much about the importance of filleting parts, but to make the paper models the piece would have to not be filleted)
4. How do i go on about learning to make molds?
5. What are some puzzles I should attempt first to get used to using CAD?
6. What I did to design the puzzle is make a wireframe diagram of 3 sides of the cube and then messed around with the extrude-face tool & the intersection tools and I also added cylinders to some of the pieces and eventually it worked out(well, I think it did), can this method be applied to other puzzles?
Below is a picture of the diagram I mentioned, and a picture of paper models of the pieces(except for the core, too complicated to make a paper model of)

1 & 2: http://printo3d.com/requestforquote.html
3. I don't think paper would be good for mold making, as the paper might collapse under the weight of the mold and deform. Worth a shot, but have fun filleting every piece.
4. This is the video I learned from.
5. Try a 3x3 first and go from there.

Don't really get what for your saying for 6, can you explain a bit more?

Here's some links to get you started:

http://twistypuzzles.com/articles/
viewtopic.php?f=9&t=11821&p=136371&hilit=printing+alternative#p136371
viewtopic.php?f=9&t=10945&hilit=printing+alternative

A lot of questions you mentioned are covered in other topics. Don't forget to use the search facility. It has been recently repaired so should dig up what you need. Based on your questions, you have a lot of reading to do.

Well what I basically did is look at the face of the helicopter cube, drew it, and then drew 2 more and placed them perpendicular to the first face.Then, I extruded triangular prisms, added and inscribed cylinders. After experimenting with different combinations and using the intersect and subtract tools I got the pieces.

1. I think printing masters will cost about $40 for a set of one of each piece. Printing a core will also cost about $40. (at printto3d)
2. Your STL files, scale, preferred fill style
3. Forget using paper meshes. It's not going to work for two reasons: Tolerances, fillets AND the texture of paper.
4. You can only learn to make molds by screwing up numerous times.
5. I think your design is OK to print. You can learn CAD by just playing around with it.
6. What you are doing is a feasible way of doing things. It's not exactly an efficient method but it works. I used to do what you do a while ago but recently I switched to SolidWorks and that is just sooo much more efficient.

i think you should try to make all the pieces out of paper (except the core) then assemble it, then you'll see if your design works, thats an idea

It seems to me that cores can't be casted because of their spider-like structure, I made two different changes to the core, putting a sphere in the center of the spider and turning it into a sphere, are any of these viable version to make molds of?

Also, does there seem to be anything wrong with the design?

Drewseph makes half-sphere cores, so he can cast the core (he needs to cast 2 parts) then glue the 2 half spheres together.
Or you could print the core.

I did manage to cast a helicopter cube core:

It isn't pretty and of course it required drilling the holes but it works.

If you're only building a single copy anyways I don't see why you would want to cast the core. And I don't think building multiple copies has a point as it is a no sell item (don't ask why I casted the core, long story).

Your design is completely valid and will work just fine. You might want to change the profile of the cuts a little though as your design will have a popping issue. It's not going to ruin your fun playing with it but using a spherical design instead of a cylindrical design will improve it.

Don't forget the filleting issue when finalizing your design!

if you make the core in the shape of a dodecahedron or onything else with 12 sides and drill the holes then would casting a core not be a so big problem

What do you mean?

Also, I want to be able to cast the core because I wanted to give a second cube to a friend.

Hi, new friend!

notice the gaps

I managed to come up with this.
Also, I just realized that if I print half of the spherical core, it would be cheaper, but how does he glue them?
Is this this design better and prevents popping? And how do I address the filleting issue?

As a builder and not a solver, I really don't care what the puzzle looks like as long as it works and works well. (Just look at some of my puzzles like my speed 2x2x1)
p.s. I hope that info was helpfull...friend

I don't know how I should fillet the pieces, and is there a way to avoid the gaps that result?

The best way I can explain it is that you need to extend the parts out the distance you plan on filleting them...which can be tricky.

Well, I extruded the outer face of the center piece and filleted the 3 visible edges, but when I fillet the corner piece its size doesn't correspond anymore to the center piece.


Also, filleting the internal piece is proving to be quite the challenge. What would happen if i don't fillet the internal edges?

I didn't bother to fillet the internals on my version as well. It works fine, but I still regret I didn't take a little extra time and add some fillets. The sharper edges also are subject to more wear.

Even if you're going to make two I think making a mold of the core isn't a good idea. It uses up a lot of rubber and combined with the possibility of failure and the time I would just go with a printed core.

I guess I can fillet the exterior and then sand the internal just a little bit, about the core, it seems that the spider design isn't as strong as a sphere so I'm definitely going to go with the spherical core but the sphere is bigger so it seems that only printing half and then casting would turn out cheaper.
If it's isn't cheaper, please tell me.(If so, I'll probably make a spider core that works, since it turns out the one I made blocks the edges from turning.)

A spider core is plenty strong. I wouldn't go with making half the core and then casting, especially if it's only your first time. Sooner or later you will probably screw up.

Unless you have an ingenious locking mechanism, once you put in the screws they'll just force the two halves apart.

Ok , I guess I'll make a smaller spider core. Or a thinner one, but that would probably weaken it,( Should I still fuse it with a sphere?)

(Removed)
Wow, I really was tired when I posted that! sorry it makes no sense.

Here's my latest design, I think I filleted it correctly. What's a good radius for the holes in the core for the screws?(And where should I get screws and what type?)

i would say 3mm holes because then type a diy screws will fit

Take care of the difference between radius and diameter.

For type A screws, this works:
Holes in the core: 1.4mm radius
OD center: 5mm radius
Large ID center: 3mm radius
Small ID center: 1.9mm radius

I don't quite understand, but I'm probably gonna have to change the sizes again, also, where can I get those screws?

Thanks for pointing that out also, I thought the pieces were ready to be printed. by OD and ID do you mean outer diameter and inner diameter?

You can order the screws (in sets of 6) at C4Y.

I only see 2 choices,
http://cube4you.com/452_Cube4you-Screws+Springs.html
http://cube4you.com/111_Screw+Washer+Spring.html
how do I know what type they are?

Also, what are the most reliable moulding and casting materials?( and where can I buy them?)

oommoo (the six hour one) ruber from smooth-on and alumilite black from alumilite.com and smooth-on casting stuff for white.

OOMOO for making your mold, for sure.

I still don't understand what TomZ meant, I wanna make sure all the holes are the right size before anything.
What did he mean?

What is it that you don't understand?

Well you gave 4 different radii but I only see one place to apply one, the 1.4 one on the core.
What are the other measurements for?(Where do they go?)

What I gave you was:

Large Center ID -> this is the size for the hole which contains the spring/screw assembly
Small Center ID -> this is the size the hole for the screw, but not the spring to pass trough
Center OD -> this is the thickness of the center 'stem'

(I only just noticed how I used ID and OD when talking about radii )

This post shows the design criteria that I use for printed cores:
viewtopic.php?p=137895#p137895

For your first project, I would recommend printing the core. It will give you a more precise foundation for your puzzle.

Ok, if I understood right the 3mm radius can be ignored, this design should be with the correct radii.

Also, I'm assuming these are type A screws? http://www.cube4you.com/452_Cube4you-Sc ... rings.html