Puzzle Design With the Engineering Process




Introduction: Puzzle Design With the Engineering Process

About: My name is Troy. I'm a Mechatronics and Aerospace Engineer. I make things out of wood and electronics and spend time outdoors (especially SCUBA diving).

This instructable covers how I followed the engineering process to develop a 3D printed puzzle. This process can be used anytime a design is being developed to ensure the final product is something useful and will meet your design requirements.

The following steps of the engineering process will ensure that you focus your time and effort into iterating a design until your requirements are met:

  • Ask - Define the design. What needs to be made and what requirements does it need to have?
  • Imagine - Brainstorm solutions to the presented requirements. Research what solutions others have come up with.
  • Plan - Select two or three ideas that can be refined. This can be done on paper or in a CAD software package.
  • Create - Build a prototype based on the plan previously defined.
  • Test - Find strengths and weaknesses in the design. If available, have others offer feedback on your prototype.
  • Improve - Based on feedback from yourself and others, brainstorm ways to improve your design and create a new prototype to test. Once you are satisfied with your design, your final product can be made.

If you are interested in purchasing a bolt puzzle instead of printing and finishing your own, you can visit my Etsy Shop where they are for sale.

Step 1: Tools and Materials



Step 2: Sketch


When thinking of designing this puzzle, I gave myself the following requirements:

  • Made from 1 in. bolt
  • Coupling nut must hide seam between bolts
  • Coupling nut must spin freely between both bolts
  • Must stay locked until purposefully unlocked

After researching the difficulty of finding (and expense) of metal 1 in. bolts, a 3D printed solution was the only result I could come up with. So my first requirement was no longer a requirement.


Brainstorm possible ways to lock the puzzle. Write down a list of ideas and draw some sketches of your ideas. Research other existing puzzles for possible locking mechanisms.


Select two or three designs to refine. Draw new pictures and include some rough measurements so you know what scale you are working with.

Shown are a couple sketches of my first design that didn't work in the end.

Step 3: Test Your Idea


Test your idea by making smaller parts of the design you haven't made before. I have never designed or printed threads before so I made a small test specimen to verify my design before printing the whole part.

Step 4: Design Failure 1

Test 1:

My first design utilized a horizontal locking pin to engage the puzzle. This failed my first test for a few reasons. The cross-sectional area of the smaller inner threads were not strong enough to prevent the layers from separating when unlocking the puzzle. Additionally, the locking pin needed to be short in order to fit in the available area. This caused the magnetic force to be too strong to easily dislodge the pin contributing to the likelihood of breaking as shown above.

Step 5: Design Failure 2

Test 2:

This change was promising as it exchanged the horizontal locking pin to an axial locking plate. This strengthened the locking mechanism by spreading any torsional forces through a larger surface area. In testing among my engineering coworkers, a weak link was discovered. The inner threads are left handed to prevent the two bolts from being able to be unscrewed until the coupling nut is moved to one of the ends. Unfortunately, the two half threaded sections weren't strong enough to hold off a determined engineer trying to figure out a new puzzle, they broke off when rotated the wrong direction.

Step 6: Final Design


The final design consisted of increasing the inner threads of the locking mechanism allowing for two 1/8 in. rods to be epoxied into place to stiffen up the locking threads. This design has so far proven to be engineer proof among my coworkers.

Step 7: Assembly


  • Insert the locking plate into the half of the puzzle with the metal pins.
  • Secure the second half of the puzzle through the left hand threads.
  • Tip the puzzle so the locking plate slides against the magnet. You will hear a click when engaged.
  • Rotate the coupling nut to any position on the puzzle.


  • Rotate the coupling nut to the end of the puzzle with the metal rods.
  • Tap the puzzle against your palm until you hear a click.
  • Turn the bolt clock-wise until the two halves separate.

Step 8: Print Your Parts

Print the four attached stl's:

  • Locking Bolt
  • Latching Bolt
  • Coupling Nut
  • 1 in. Socket

Step 9: 3D Printed Socket

I didn't own a 1 in. socket so I simply printed my own that would fit on a 1/2 in. socket adapter.

Step 10: Clean Threads on Bolts

Secure the 1 in. die in a vise. With the socket adapter, cut the threads to improve the look of them. This will also bring the threads into ASME standard specifications. Using mineral oil as a lubricant will result in something that has a nice matte finish and is safe to handle.

Step 11: Clean Threads on Coupling Nut

Holding the coupling nut in a vise, chase the threads using a 1 in. tap in a drill. Be sure to lubricate the tap with mineral oil.

Step 12: Insert Rods and Cut Locking Plate

Cut two 1/8 in. rods to 1.75 in. in length. Apply epoxy to the rods and insert into the holes. Allow to dry.

Cut the locking plate from roughly .035 in. thick metal. Anything from .030 to .040 should work. I cut my locking plate from a piece of thin angle iron. The angle iron was left over from installing a new garage door opener. The dimensions are 0.700 x 1.500 in.

Step 13: Trim Rods

Using the coupling nut to stabilize the bolt, trim the excess rods with a disk sander. Be careful not to take off too much at a time to keep the part cool.

Step 14: Glue Magnet

Epoxy the magnet into the other bolt and allow to dry. A small round metal rod (old bicycle spoke) was helpful for me to reach down into the hole to glue it in place.

Step 15: Finish

I finished my puzzle using sandpaper (400-1200 grit) and mineral oil. This resulted in a wonderful matte finish that is smooth to the touch.

Step 16: Admire and Iterate

Take some time to admire your work. Now apply the engineering process to other designs and projects.

Be the First to Share


    • Trash to Treasure Contest

      Trash to Treasure Contest
    • Stick It Challenge

      Stick It Challenge
    • Make It Modular: Student Design Challenge

      Make It Modular: Student Design Challenge



    3 months ago on Step 7

    I finally unlocked it. Pretty interesting!


    3 months ago

    I bought one and it looks professional and has a smooth finish. But I cannot unlock it. I hear the click but no seperation. Any ideas?


    3 months ago

    Great story of design! Maybe I missed it, but I don't see what material you used for 3D printing, i.e. what kind of plastic, thanks.


    Reply 3 months ago

    I actually didn't specify. But this is printed in ABS. I assume that PLA would work as well, but I'm not sure how it would sand and it may not be quite as strong.


    Reply 3 months ago

    I asked for the very reason you mentioned: sanding. I had some 3D puzzle pieces made with nylon and wanted to sand them but it didn't work very well. And in fact I was thinking that ABS would be better. So thanks for the reply!


    3 months ago

    This has the makings of a wonderful geocaching container - very clever!


    4 months ago

    Really like that you shared your failures along the way to your awesome finished product. Most authors, including me, skip most of the failures…not because we are inauthentic, but to streamline the write-up. But what a great reminder that the failures should be celebrated because they are an important part of the iterative design/build process.


    I really appreciate that you posted the failures, and not just a description of the process and final results.

    Very nice, thank you for sharing


    4 months ago

    Brilliant work as usual Troy. The finish on your pieces looks so good!


    Reply 4 months ago

    Thanks! I'm happy about how it turned out and like that you can only tell that it was 3D printed if you know where to look.