Laser Cutting Demo

[Lee DeRaud]

People at work kept asking me what the laser could do, so I made up this widget to show them, cut from 3/16" acrylic. Note that the only thing holding it together is friction: the pieces were cut to a zero-tolerance fit, I had to use a vise to put it together.

The CorelDraw used to cut it is also attached. Note that if you try to cut one of these yourself, do not scale it: the adjustments to the finger/slot sizes will scale also, and it either won't stay together (if you scale it smaller) or won't go together at all (if you scale it bigger). When I get some of the user interface details worked out, I'll post the VBA code used to generate the drawing.

[Jerry Allen]

Very cool design. I'll give it a try, and thanks for sharing.

I just checked out the file and will have to interpolate.
Zero clearance in the width of the finger should actually be a couple of thousandths due to burn width. The depth is based on .1875, but is .212. Does it still come out flush?
I only have .093 and .220 material and will convert to those dimensions to give it a try.

[Lee DeRaud]

Zero clearance in the width of the finger should actually be a couple of thousandths due to burn width. The depth is based on .1875, but is .212. Does it still come out flush?
I only have .093 and .220 material and will convert to those dimensions to give it a try.

The VBA code that generates the drawing makes the fingers wider and the slots narrower on one side of each of the eight mating edges. I used 0.006 as the allowance (it's an input parameter)...the finger width is 0.2, which I think is how you got that 0.212. The finger length is normally the same as the material thickness: there's "kerf" taken from both the end of the finger and the bottom of the slot, so it balances out. (There's also a parameter to add finger length so you can trim the burn off when you're making it out of wood.)

Hmm...I think I said 3/16" thick in my original post, but it's actually 5mm. Sorry about that.

Visually, the edges are flush: you can feel some irregularity, but I don't have anything sensitive enough to measure it. I think it's due to beam divergence between the top and bottom surface of the acrylic. The edges are just as they come out of the laser...the most time-consuming part of making one of these is peeling off the masking tape after cutting.

This cube thing is just a "side effect" of the program: it was actually written to generate cut vectors for some double-wall box-jointed boxes I've been making as gifts. Like I said, it'll get posted Real Soon Now.

[Jerry Allen]

I was a little confused after analyzing the .cdr file. The fingers and slots are not consistent if you measure them between all the sides. I'm sure it worked okay for you but I opted to make all the fingers and all the slots consistent. I used .206 for all fingers and .194 for the slots. My material was labled, .220 but actually measured about .205. I made the depth of the slots .200 but have a mil or two to get rid of next time that is probably caused by the melt in the corners of the slots.
Anyway, It came out great. I was concerned about breaking something because it was so tight and I've shattered acrylic before, but it pressed together by hand with some effort against a flat hardwood surface. It is amazingly solid when assembled.

[Lee DeRaud]

I was a little confused after analyzing the .cdr file. The fingers and slots are not consistent if you measure them between all the sides. I'm sure it worked okay for you but I opted to make all the fingers and all the slots consistent. I used .206 for all fingers and .194 for the slots. My material was labled, .220 but actually measured about .205. I made the depth of the slots .200 but have a mil or two to get rid of next time that is probably caused by the melt in the corners of the slots.

Yeah, the "kerf adjustment" is done on the left and right edges of the "sides" and all four edges of the "top" and "bottom. The "front" and "back" pieces are nominal (equal finger and slot) on all four edges. It made sense at the time because the original program didn't do the allowances and doing it this way localized the changes.

Anyway, It came out great. I was concerned about breaking something because it was so tight and I've shattered acrylic before, but it pressed together by hand with some effort against a flat hardwood surface. It is amazingly solid when assembled.

I haven't had one break going together yet, but I had a friend break one trying to get it apart.

[J Porter]

There's a really neat pattern for a box on J Stephen Spence's web site. Take a look at it and you'll see what I mean. It's in the downloads section of:
http://www.jstephenspence.com/
The design says it was done by a guy at Unversal Laser Systems.

But there's a catch. I've never been able to properly assemble it. It's a great design idea, but all of the pieces don't fit to make a cube... lol

Come on... someone make me feel like an idiot and put one together..

If it really does fit together, then it makes a great puzzle.

[Lee DeRaud]

There's a really neat pattern for a box on J Stephen Spence's web site. Take a look at it and you'll see what I mean. It's in the downloads section of:
http://www.jstephenspence.com/
The design says it was done by a guy at Unversal Laser Systems.

But there's a catch. I've never been able to properly assemble it. It's a great design idea, but all of the pieces don't fit to make a cube... lol

Come on... someone make me feel like an idiot and put one together..

If it really does fit together, then it makes a great puzzle.

I saw an assembled one at the dealer I bought my Versalaser from: it really does work.
Looking at the drawing, it should work...unless you get one or both of the pieces without text on them "inside out" when you're trying to put it together.

[William Capman]

Yes it's a great little puzzle , I know that it should go to together but does it really??

I've had it on my desk for at least 6 months, in pieces. When I need a break I work on it. Six months of frustration and it's still in pieces.

[Lee DeRaud]

Yes it's a great little puzzle , I know that it should go to together but does it really??

I've had it on my desk for at least 6 months, in pieces. When I need a break I work on it. Six months of frustration and it's still in pieces.

<sigh> Yes, it really does work. Sorry this took so long: I don't have access to the laser at my "day job". Stop reading now if you don't want to see the solution.



Ok, look at the original drawing and number the pieces left-to-right, top-to-bottom:
1 2 3
4 5 6
The right edge of 1 mates with the top edge of 2.
The top edge of 1 mates with the top edge of 3.
The left edge of 1 mates with the bottom edge of 5.
The bottom edge of 1 mates with the top edge of 4.
That leaves four free edges and there's only one way 6 can fit.

Took 10 minutes to cut and about 90 seconds to solve, although I'll admit I spent some time looking at the picture while the laser was doing its thing.

The only way to screw it up is to get either 2 or 3 facing outside-in. Of course, if you didn't print the text it gets a bit harder, since you can get the other four flipped as well.) There may well be other solutions: there are quite a few rotational symmetries in the pieces.

Attached is a cleaned-up version of the file, sized for 0.25" material. I'll post a picture of the assembled widget whenever I find the USB cable for my camera.

[Lee DeRaud]

Here's the assembled box, in 1/4" MDF.

[J Porter]

ROFL... thanks for solving the puzzle Lee...

Mine still doesn't quite fit together with a couple of tabs being not quite the proper size, but I see that it does work and will if I use a bigger sledge hammer..

You win .. You get a free trip around the sun...

[Lee DeRaud]

Mine still doesn't quite fit together with a couple of tabs being not quite the proper size, but I see that it does work and will if I use a bigger sledge hammer..

I'd noticed that the tabs were off a few thousandths in several places, and it was easier to recreate the drawing than to fix the original. Plus it was sized to something odd like 5/32" (or maybe 1/8" with room to trim the tabs down later). You should be able to scale my update to fit whatever material you have on hand...I may try acrylic this weekend.

[Jerry Allen]

Again if you don't want to know the solution, don't read this, but I assume most folks make these things for the benfit of others...

Like Lee, I noticed that the original file was not drawn properly. So I redrew it for GE acrylic, .220, which is actually .205. It worked but was too sloppy. The enclosed file has tabs that has tabs increased .008 and stays together better, although I think it could still use another couple of mils to snap together a little tighter. To do this I had to break up the original into the separate pieces.

I also modified the graphics a little and added some.
The blanks side is where you could put your logo or a name, etc.
The pieces are numbered in the corners so you might want to delete that.

hint:
__5_
1234
__6_

[Lee DeRaud]

Like Lee, I noticed that the original file was not drawn properly. So I redrew it for GE acrylic, .220, which is actually .205. It worked but was too sloppy. The enclosed file has tabs that has tabs increased .008 and stays together better, although I think it could still use another couple of mils to snap together a little tighter. To do this I had to break up the original into the separate pieces.

That's 0.008 total for the width of the tab, yes? I've been using 0.006 adjustments for "press-fit" joints in that stuff, but that adjustment was applied to both sides of the joint: total of 0.012. For something intended to be taken apart again, 0.005 each, 0.010 total ought to be just about right.
(And yes, thanks, it's 0.205, not 5mm as I think I said before. The stuff I got from Home Depot was unlabled except for some tiny print on the price tag that I had absolutely no chance of reading.)

[Jerry Allen]

Yes, .008 overall just on all of the larger tabs and one only of the sides with the smaller double tabs. Some of the larger tabs could have used .009 or .010. I only did the tab side to speed up the process. I did not want it to be as tight as the press fit design. As it stands it's real close to being a comfortable snap with disassembly capabilty.
The GE .220 thk (.205 actual) is from Home Depot.