This is a small project I made up simply for practice. I've long been interested in machining as a hobby. It simply fits in well with other interests of mine, and is plenty of fun in its own right! I first got access to a machine shop at the University of Akron, as part of the Formula SAE program (lots of fun!), and loved it. Since I discovered and joined TechShop I suddenly have access to things like vertical mills and lathes again, meaning I need to knock the rust off my skills. :)

I decided to make one of these simple puzzles out of aluminum to practice squaring up stock and taking basic slotting cuts. 6061 was the order of the day because it's cheap, available, and machines well.

Burr puzzles can be learned about on their Wikipedia page. I printed a copy of the #305 image off that page to use a drawing, no dimensions necessary due the modular nature of these puzzles. ;)

Step 1: Squaring Up.

First order of business is squaring up the stock. This helps immensely when taking measurements during the machining process, and is usually outright required in order to make parts clamp accurately and repeatably in the mill vise.

I went out and bought a length of half inch square stock (6061 aluminum) and cut off six 1.5inch-plus-a-bit pieces using the cold saw. Hacksaws, bandsaws, etc, would all work fine too. Measuring the stock as it came from the mill, it was pretty close! Just 0.503 or so on one side, 0.504 on the other. Squaring this stuff was hardly necessary for a project like this, but i did it any way to build a little muscle memory.

Learning about the process is easy on YouTube, I used a 3/4inch end mill to skim cut each face, 3 wide in the vice. Once I had blocks squared on 4 sides, I reset the knee (Z-axis) dial to 0. This reduced the amount of measuring I was doing, since I was simply "working toward 0" for the second set.

Once that was done, I milled the blocks to length, making sure to skim one end so it was square, and the other to length. I did it just one at a time because didn't want to set them all up as a gang in the vise. I was concentrating on the process more than speed. One piece resting on parallels, slightly protruding from the side of the vice jaws, with another piece on the other side to keep the jaw from tilting sideways. Easy Peasy.

THEN I had to make one of the blocks over again. :D Good thing I decided to practice, because I made a mistake that could have bit me later on a much more expensive piece. Skimming one end of a piece left a burr on the edge that stuck out differently than the other cuts had. When I turned the part around to mill it to length, the burr formed a pivot point in the jaws. Shortly into the cut, the helical flutes of the cutter lifted the end of the part. I don't have a picture of what the end looked like, and all the physical evidence has been safely eliminated. :)

From then on I started deburring the parts with a file before reclamping them. ;)

The picture shows the finished set.

Step 2: Starting the Main Cuts.

Now I have 6 pieces of aluminum that are all 0.500x0.500x1.500 inches. This is handy because the puzzle design is very modular, with thousands of different combinations. On the Wikipedia page you can see how the pieces are defined simply by which "cubic units" are cut out. They are easy to spot in the drawing I used. The overall piece dimensions I'm using make each of the units 0.250 on a side, with no "extensions" making the pieces longer. That way it's easy to read the drawing, since every cut is some multiple of a quarter inch wide, plus 3 to 5 thou for sliding clearance.

I used a 3/16" end mill for these cuts, allowing to make rough and finishing passes since it's slightly narrower than one "unit" (the narrowest slot I'd need to make). I set a piece on parallels (the tallest ones), centered in the vice, and cranked the x-axis over until the cutter was roughly centered on the slot I'd need to cut. The first piece was the simplest one with a single 2x1 unit (0.25x0.50) slot. I brought the table up a thou at a time until the cutter just barely marked the top surface, and zeroed the dial (actual cutting depth will be the dial reading plus something under a thousandth). Spindle RPM set right (MAX speed in this case :) ), I just took easy passes at .050, .100, .150, .200, .245, and .253 using the y-axis. 5 roughing passes and a finisher. Each pass is so short it hardly takes any time at all.

Once I'm at depth I used the calipers to measure how thick each wall of the slot was and programmed it into the DRO. For example, on the finished part above the left "wall" should be one unit thick. Measuring it I initially got 0.382 (or so, i don't remember really), so I'd set the DRO X-axis to read 0.382. That way I could simply make passes to step the wall over until the DRO read 0.247, instead of constantly remeasuring. I like DROs.

The rest of this Instructable is pretty straight forward. :D

Step 3: Progress!

After that first piece, I then realized coolant wasn't really necessary in aluminum, and that 2 flute cutters were a better idea. Yay practice!

Just remember, half the flutes, half the feed rate.

Step 4: More Progress!

Ok, Ok, I totally lied in Step 2. I didn't remember the DRO trick until about now.

Guess what, it saves something like 2-4 minutes per slot with all the extra cranking needed in order to fit calipers in there.

Practice. Practice. Practice.

Step 5: Done!

Huzzah! I now have a puzzle I can't assemble!

As a note, the two most complex pieces both have a 1-unit-wide slot in them, I definitely had to use a ruler to center the cutter for those. Or in this case, the depth extension on the calipers.

Also, the left hand slot in the piece in the second photo crosses with another slot. In these cases, double check which direction you are starting the slot from so you don't accidentally start climb milling when the cutter crosses into the other slot.

Step 6: Deburring. (then It's Done)

A quick tip on deburring. On parts like this, you don't have to file on the edges at an angle like most people do to knock off the burrs.

Just hold the piece against a fine (clean!) file like in the photo. Ignore my pinky, I didn't have three hands to take the photo. Using super light pressure, any flat surface without burrs will just skate across the file without getting cut. Any burrs will protrude into the file teeth and get skimmed off. Just keep the pressure really light and steady as you stroke and you'll feel the part free up and start gliding when the burrs are gone. Burrs will also get taken off better if you orient the piece so that particular burred edges are facing the direction of cut (towards the handle). You can do this on all the surfaces of these puzzle pieces with a little practice.

Just remember the file only cuts one way, and dulls the other. ;)

Also, I totally cheated and looked up the solution so I could take the intro pic. :D