A-Mazing Box Detour

by VeryWetPaint, published

A-Mazing Box Detour by VeryWetPaint Feb 2, 2011

Featured Thing!


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

13723Views 4528Downloads


This is another maze-box with a lid that must navigate through a maze to take it off (or put it on).

It's been reconfigured to use a little more maze-solving psychology so the user tends to make wrong turns while trying to open the box. Most of my testers made wrong turns at every intersection except one. (Closing isn't as difficult as opening, but it a bit more complex than the original model.)


Fixed! The original maze8.stl file contained a number of errors; I have uploaded "maze8_fixed.stl" to correct them.

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

Quickly Scale, Mirror or Cut your 3D Models

App Info Launch App

How did you design this? Im trying to make a similar project

It's a straightforward derivative of A-Mazing Box by Wizard https://www.thingiverse.com/thing:1481
All I did was edit the maze and re-run his scripts.

A-Mazing Box

When I slice with slic3r, the depressed parts (the actual maze tracks) are a separate loop than the outside part, so rather than going in a circle continuously, it does two shells of outline as a circle, then goes back and does a small loop for each maze track. This makes it so the tracks aren't actually connected to the walls of the shell. Anyone know how to fix this? Does the "Vase" seting in slic3r fix it or should I use a different slicing engine?


I have the same problem. Horrific layering issues, can see light through the walls. How'd you fix?
Using cura 2.4

Awesome - printed one today. Only problem, and this might be just my printer's calibration, the nubs wouldn't turn into the fully closed location. Not just over the "latching" bump part, but they wouldn't turn down that last channel at all. I had to trim it some with a knife, and now it "locks" shut before the actual intended "locking" indent.

a transparent top... brilliant! I'll try this as well :)

bummer I can;t print this without it being an unfilled pair of walls :(

I get a load of slicing errors :(

SF35 reports 'dangling edges' and creates partial solid layers across the centre of the box to try and fix it.

from about 11-14mm Z. and around 35mm too.

I'm getting this too. Ruined an otherwise cool print!

Sure enough, VisCAM View detects 12 flipped triangles, 78 unmatched edges, and 27 fragments. Lots of errors!

I use commercial build software (SDView) that didn't even notice the errors. Skeinforge probably isn't so tolerant.

Where did all these errors come from? I exported the mesh directly from OpenSCAD (created with Wizard23's python script) so I'll try running it through again. Or should I just re
pair the exported mesh?

Okay, I refactored the OpenSCAD mesh and rechecked it. Now it has 0 errors: no flipped triangles, no unmatched edges, no intersections, no overlaps.

It should build okay now. The new file is "maze8_fixed.stl"

What is this maze solving psychology? Do you have a link where I could read up on it? Seems like it would be useful for game design....

It's ad-hoc psychology based on observing users' behavior, not deep understanding of why. Luckily I've got a huge supply of office-mates who like to test my toys, and I like to study their reactions.

Most solvers who picked up Wizard23's original design sailed through with only one wrong turn at the six possible opportunities, and the one wrong turn tended to occur at the same branch. (A likely problem was the "right" path was straight at most junctions, and most users tended to go straight.)

So I tried to invert the 'right' and 'wrong' choices at as many junctions as possible while retaining the basic structure of Wizard23's maze. That permitted me to exploit whatever psychology biased the solver toward a particular choice at each junction without deeply understanding why.

My tester
s seemed to confirm its increased difficulty. Those who'd trivially opened the original A-Mazing Box now made many wrong turns, often repeatedly. Average solvers now had to pause and study the maze as it was gradually revealed. Some less-proficient solvers got stuck and ultimately gave up.

But h
ere's a discussion of maze-design theory and psychology http://www.web-books.com/Classics/Books/B0/B873/AmuseMathC15P3.htmhttp://www.web-books.com/Class...

PS: My text was written a bit hastily because I'd accidentally published this 'thing' with a blank description, so I was hurrying to fill in the blanks!

is there a way to make it one big maze or does the script need to duplicate it 3 times?

would be cool to have one giant maze. would be so hard to open :-D

you dont need to duplicate it. but if you have only one maze you'll need to adapt the other part so that it also has only one knob. with only one knob its also a little unstable in the movement.

Could it be possible to get the png file used to create this? I would like to customize the model further and it would be much easier based on the original python scripts than editing the already built files. Thanks in advance.

Sure, I just forgot to include it. Thingiverse interprets PNG as an image, not a project file, but you can find the raw file by clicking the image and choosing "Original Image" from the picture choices, or by using this URL...