All posts by Ryan

likes lasers, geometric design, typography, and video games

123: Semi-Precious Stone Engraving

The smoky quartz gives away its engraving with visible lines in the star.

A recent client sent me a few small precious gemstones to see how easily they could be engraved. I did some tests on them, with varying results, and sent them back her way. We went with a really simple five point star to test the engraving quality—with the smallest one just .07″ in diameter!

The tiniest piece, topaz, didn’t seem to show the same engraving pattern as the quartz.

Out of this initial batch, the topaz and quartz both engraved beautifully, but the opal’s engraving was basically invisible. This is probably due to a combination of factors. Opal is completely opaque, so no light can come through the material to highlight engraving. It’s a consistent color throughout, and it doesn’t burn, which means the untouched surface is only different from the engraved surface in texture. It’s very hard to capture in pictures.

This image was processed to try to show off that the opal was engraved. The star is there, I promise! Look closely!

My client received her gemstones back, and made some observations: in one piece, the horizontal lines that make up the engraving (because of how raster engraving works) were visible even at the laser’s highest engraving density setting. She also felt that the engraving wasn’t very deep and was wondering if a deeper engraving would affect the overall quality of the engraving.

Conveniently, I found out that Jennifer had done some tests on gemstone beads before, so we had some quartz of our own, as well as some garnet, to do tests. I became determined to solve these problems and answer these questions without requiring a new shipment of gemstones.

I was pretty shocked to see that the individual lines of the engraving were visible; this is something that happens on lower density settings (like 4, which is used to reduce engraving time but can leave ultra-fine gaps between the lines in certain materials). At the max density, this normally isn’t visible at all. I decided to try to engrave my pieces slightly out of focus, increasing the laser point’s diameter so that each pass overlaps somewhat.

My later engraving tests on beads of quartz and garnet were… less than successful.

As it turns out, though, none of my testing worked quite the way I’d hoped it would. Maybe the gemstone beads’ facets were impeding the engraving, or both of my solutions weren’t solutions at all, but all I got out of six different attempts were excessive chipping and almost unrecognizable engravings.

They’re supposed to be hearts! While I engraved the first batch with stars, I figured hearts would be a good shape for the quartz and garnet beads. I don’t think the shape mattered much in the end.

I engraved three pieces of quartz. From left to right, I gave them one engraving pass, then two, then three, at the same settings I used for the client’s quartz seen above. This time the chipping was so severe that you almost can’t even see the difference between the three tests, but upon close examination the heart that was engraved three times was actually a deeper engraving—completely moot with how poorly it turned out.

For the garnet, I was testing whether engraving out of focus would result in a smoother engraving, and it certainly doesn’t seem to be the case. Perhaps the offsets (focused, 0.5″ out, and 1″ out) were too severe a difference to see any value, but even the focused heart on the left was nowhere near as clean an engraving as on the client’s gemstones.

So I think the moral of these disappointing results is to always perform subsequent tests on the same exact material. I wasn’t able to give my client a good answer to her questions with these tests. Still, I’m fairly convinced now that chipping will prevent multiple passes from getting a deeper engraving without reducing the engraving quality, but the jury’s still out on whether adjusting the focus can have a positive effect. We’ll find out eventually!

121: Anodized Aluminum Moai

When Jen and I were first researching buying a laser, we didn’t plan on buying the one we ended up with. We were looking to get a small, semi-portable solution. But my brother Johnny found a family out in Terre Haute, Indiana, who had owned their laser for about a year and realized they weren’t really using it as much as they’d like and they wanted to sell it. Buying the used laser meant I inherited some of the previous owner’s mistakes—like a damaged lens and rulers that had art engraved on them—but I had plenty of time to cover up those mistakes with my own.

One of the benefits of buying used is that the previous owner didn’t need their material stock anymore either. Bundled with the machine were dozens of small samples of various laser-ready materials, including a piece or two of black anodized aluminum. I dug up one of those old pieces while cleaning out the office, and realized that I hadn’t spent much time engraving on the medium, so I decided to give it a try!

I’ll probably never find out who these two are, but I’m sure they had fun engraving this picture!

This piece of anodized aluminum had a remnant from the previous owner: an engraving test of their own that they didn’t keep. It also had one corner clipped off, so maybe the leftover engraving I have was a test run and they kept a more successful engraving. Either way, the laser is way too low-power to actually cut through this metal, so I’d be limited to engraving the surface. I don’t have the tools necessary to cut this into a better shape, so our two buddies will, for the time being, remain a part of this experiment.

Full color, desaturated, and then inverted so that it would engrave properly.

I settled on a picture of the large moai head Jennifer took some years ago. This picture converted into black and white really well, and as you have to with all materials that engrave from dark to lighter colors, I had to invert the picture so that the laser would fire on white, not on black.

My first engraving test was too powerful: I typically use 35% power to lightly engrave surfaces, but the laser overpowered the black, creating a blown out image. I settled on 20% power—recommended in the laser driver’s settings for this medium—and the contrast was much better. That’s what I get for ignoring Universal’s built in material settings!

Because it only took one pass to get an acceptable result, I decided to use this project to compare a few of the settings available in the driver. When engraving art that has shades of grey, the machine has to dither to convert those to art it can engrave. The machinist can choose to have the laser convert those shades to a patterned halftone, use an error diffusion method, or use a threshold to convert the art to black and white. You can also select from seven different Image Density steps; the lowest step is the quickest and skips the most space on the material between horizontal lines, and the highest takes much longer and leaves very little space between each pass. The difference between those settings is pretty remarkable, but I didn’t want to do seven passes for each, so I decided upon odd numbered settings.

In the end, I made a matrix of engraving settings: each row is a different Dithering option, and each column is a different Image Density. Twelve engraving passes later, here is the result.

The full grid of engravings. Most of the difference in detail is lost in these photographs.

As expected, the image density settings took different amounts of time to finish. Density 1 took only 47 seconds to complete. Density 3 took 2:08, while the default density setting of 5 took 4:10. The maximum density setting took a whopping 15:13. Since the engraving speed isn’t really affected by the dithering option, the results were almost identical down the columns.

Lower density engravings make the dither pattern more obvious.

It’s clear in person (and less so in this photography) that error diffusion is the best option to get a good clean photograph engraved onto this medium. I’d be willing to bet that will be true for most materials, as the halftone pattern (while great for low color artwork) just doesn’t blend a photograph’s natural shades well. This is even true in the lower image density settings.

While the dither pattern all but disappears at image density 7, the time spent is too costly.

I think the default image density setting, 5, was well chosen. It’s an excellent middle ground between the time spent and the resulting image. You can’t really get a sense that the image is comprised of multiple horizontal lines like you can in setting 3, but you’re also not spending 15 minutes on a 2″ by 2.25″ photo.

120: Fluorescent Samurai FightStick

Somehow for all the plans I’ve had over the years to laser engrave a fight stick, I never actually got around to it until this week! Last week, a coworker at the library was planning on making a custom artwork label for his arcade fight stick in the library’s makerspace, and it didn’t take long discussing materials and fabrication processes (including just printing on paper to replace the stock Ryu artwork) before I practically demanded to laser engrave the hardware myself. Since I kind of stole his project from him in order to complete one of my own, I hope there’s enough juicy details herein to cover the process! Thanks, Mark!

The stock acrylic has a lip almost certainly CNC milled. These aren’t easy to recreate with a laser!

We briefly discussed engraving the acrylic panel that is included with the controller, but we decided against that so that he’d always have an unaltered panel handy. Besides, we were already talking about using a fluorescent orange acrylic that we’ve used in the library before for making bright, eye-catching tokens and trinkets. It’s a fiery equivalent to the green acrylic I use for Ingress badges; edges practically glow with this material! But the material we chose was only available in 1/8″—about 50% thicker than the stock acrylic—so I’d have to do some measurements to make sure it’d still fit. There were a few other tricky issues to figure out as well, so I knew I had some prototyping to do. But first, the design itself!

The layout is about 13″ by 8″, features eight buttons and zero button labels. Pro!

Mark had already done some of the hard work and whipped up a layout of the artwork he wanted to engrave. It was only roughly the size of the fight stick’s acrylic cover. Featuring an illustration of a kneeling samurai and seven bushido virtues, he told me he’d always been enamored with samurai and the code they lived by.

“How that warrior code and Buddhist philosophy can coexist interests me.  … I think anyone who is interested in an applicable ethic for living, who understands that the world can be violent, but hopes not to have to actually be violent, will find plenty to study in the samurai.”

The layout didn’t have accurate measurements or the button and joystick layout, though, so I’d have to see if someone put one together online. As it turns out, we didn’t need to rely on the community; late product manufacturer Mad Catz provides a PSD template for users who want to print out their own art. Its pixels are no good for laser cutting, but it has the exact dimensions and throwing together some vector circles where the PSD says they should be was quick and easy.

The chipboard prototype fit perfectly, and looks pretty snazzy on its own!

Once I had the layout prepared, I ran several prototypes. The first was cut out of cardboard, which was slightly too thick to be used easily in place of the acrylic, but it did help me discover that I accidentally nudged the bottom row of buttons out of alignment when creating the vector template. I corrected those issues and cut out a second full size prototype in chipboard. The chipboard fit perfectly and even looked pretty good under the stock acrylic!

Several small material setting prototypes. Please ignore the Triforce!

I had to run several small engraving tests on the actual material because of a few techniques I planned to use that need really precise settings. First, I’d need to recreate a lip on the entire outside rim of the acrylic; I’d have to engrave very deeply in order to do so. I also wanted to make sure I created a nice depth effect on the engraving, so I’d need to combine the halftone raster engraving on the greyscale samurai art with selective vector engraving on the darkest strokes. Finally, because of the thicker material we’re using, I wanted to round the edges so they wouldn’t stick out quite as much around the buttons. This involves pulling the laser out of focus, so I’d have to make sure the depth offset was an appropriate amount.

Detail of the letter engraving, including the deep vector stroke outlining the glyph above.

During this prototyping, I figured out that I’d need to engrave at 100% power, 10% speed to burn away enough material to recreate the stock acrylic’s outer lip. That’s ridiculously slow, and across an entire 13″ by 8″ design it would have taken over three hours to engrave, so I ended up breaking the engraving into four jobs. By doing so, I prevented the laser from passing by the entire inner area of the acrylic, reducing the engraving time to about 35 minutes. What a relief!

This acrylic really shows off fingerprints, doesn’t it?

I also determined that pulling the laser out of focus by exactly 1″ creates a nice rounded effect while being a nice easy number to remember. I used the same settings that would cut through the material with the proper focus. The slight misalignment of my #2 mirror struck again and required me to move the out-of-focus vector lines just a tiny fraction to the left to compensate. One of these days I’ll get that fixed.

My final acrylic lip was close enough, and check out that smooth rounded corner on top!

After I engraved the lip and cut out the button holes, I flipped the material, removed the back face’s protective mask, and got to engraving the art! It might not be obvious at first, but if you’re going to engrave something you’ll be resting your hands on often, you should engrave the back face and leave the top smooth and untouched. Otherwise the engraving might irritate your skin and you’re going to have to clean your filth out of the artwork more often than anyone should have to deal with. Besides, the fluorescence of the material is best shown off through the material; back face engravings capture much more of the color and appear less frosty.

Samurai on metal plate. Great shadow effect, but with plenty of distracting screws.

Amusingly enough, it didn’t dawn on me until I installed the fight stick that I’d need to make some kind of thin black background layer so that the metal button plate wouldn’t be visible through the design. It’s kind of a neat look, but it detracts from the theme, so I used the only material I had on hand (some black/silver LaserLights; don’t tell Jen!) and cut the template out of that. It fit just fine beneath the acrylic, so it was time for put Mark’s fighting game weapon of choice back together for good.

One small detail: the acrylic’s rounded corners needed to be about 0.1″ radius, quite a bit rounder than I first estimated.

The final piece was simple to install thanks to the hinged design of the FightStick-branded fight stick (that’s not confusing at all, right?). I was a little worried that the originally-pencil samurai design would suffer from being printed “in reverse” (i.e. light lines on a dark background), but the way the deeper vector engraving works with the light halftone raster engraving almost makes it look like our calm, kneeling samurai is fashioned out of fire rather than graphite. As mark put it, “Simultaneously peaceful and ready for battle.”

The joystick just barely fits between the samurai and his blade.
Detail of shallow halftone raster engraving combined with deep vector engraving.

117: PC Case Engraving

To ring in the new year, and celebrate the ten year anniversary of my previous build, I decided to build a new PC. Back in 2007, it was two years before we even started playing with lasers. This time around, I knew for sure I’d be laser engraving some piece or another. I’ve engraved a few macbooks and other portable devices, and I’ve even engraved a custom wood faceplate for a friend’s ATX midtower. So I’ve been pretty excited about the idea of engraving something on my own machine!

Over a few weeks in January I did the research, collected the parts, and then planned a small “build party” with some of my local PC enthusiast friends so we could put the machine together together. Hey, it only happens once a decade or so, that’s a pretty good excuse for a shindig, yeah?

From left to right, Brenn, Jen (<3), myself, Maul, Ray, and Mark. Also not pictured: Maul’s bro Joe! Thanks for the photography, Mark!

Together we had dinner, built the PC, played some couch games, and mulled over a few remaining questions. What should this new build be called? What part of the case will be laser engraved? What are we going to engrave on it? I was so wishy-washy on the name decision that I couldn’t even settle on it before the party was over. Furthermore, I wanted the engraving design to be related to the name, so I couldn’t really come to any decisions on that front, either. But we were able to figure out what part to engrave, and as it turned out, the answer was nothing.

Fractal Design’s Define C case is sexy, but made out of questionably engravable plastic.

The Fractal Design Define C is a sexy, sexy midtower ATX case. I love the shroud, I love the quiet, and I love the flat textured front. I like simple, unassuming case designs, and I wanted to continue down that road after my last build in an Antec P180B. But when we finally dug into the case, I learned a few laser-unfriendly things I could probably have sussed out from reviews online if I had been more thorough.

The front of the case is not an anodized aluminum plate, and it’s also not easily detached from the surrounding plastic chassis that covers the front exhaust system. It’s made out of the same plastic—it’s very pretty, with a subtle vertical brushed texture, but it’s still just the case plastic. Because the textured surface isn’t repeated anywhere on the inside (or indeed on any other external surface) I wasn’t going to be able to do an inconspicuous engraving test. So I wouldn’t be able to engrave the front plate, but what about the window?

On a quest for extreme sound dampening, my previous PC build didn’t have a window at all. But over the years I’ve kind of missed being able to peek in on my parts, so this time I bought a case with built-in acrylic window.

With a power shroud for modesty and excellent cable arrangement, who wouldn’t want to peek inside?

Unfortunately, there wasn’t going to be an easy way to test that material inconspicuously either, and with the  engraving quality difference between cast and extruded acrylics, I didn’t want to gamble.

When I looked closer at the acrylic window, I noticed there was a lip on the inside, one that would fit a secondary piece of acrylic just fine as long as the measurements were correct. So I did a couple of sizing tests with some old pieces of acrylic, got my measurements spot on, and settled on a solution: cut a separate piece of cast acrylic and snap it into the existing acrylic window. I wouldn’t technically be engraving the PC case after all, but the finished piece would still look as good. As a bonus, I’d be able to easily change out the acrylic in the future if I wanted to change the design.

Amusingly enough, it was mulling around design ideas that led me to my final decision on the name of the machine. I’ve always been a fan of the Metroid series, you can see it in some of my other projects. Most game servers I host have names based on “Maru Mari”, and you’ll be connecting to “Varia” if you try to stream content to my television. I had a feeling I’d end up going with the Metroid theme again, but it wasn’t until I thought about how much fun it would be to engrave the cold steel corridors of Tourian into acrylic that I really landed on it.

The full map is too big; I’d have to fit it in this cyan rectangle

Tourian is a big map. Well, it’s not big, but its hallways are long and the vertical shafts are all a daunting climb. I’d have to compress the map pretty significantly to make it fit the relatively tiny space I had for my acrylic window. To make matters worse, halfway through the design I realized I had laid out the template wrong and was designing for the measurements in landscape instead of portrait. But after cutting a few rooms in half (and completely excising the hallway before Mother Brain’s chamber) I was able to make it fit.

The final compressed map, corrected to a portrait aspect.

I added a few additional details (the opening text scroll and an excessively big title in the original typeface in the corner) and the design was finished. I cleaned up the acrylic, seated it in the window’s lip, and used a tiny bit of clear packing tape on the inside corners to make sure it wouldn’t somehow come loose.

A mockup of what the case might look like with the final design.

The panel looks great when it’s not connected to the computer, but as it turns out, I should look into buying some motherboard-powered LED strip lighting to brighten up this design. Most of the photography here is cleaned up to make the engraving visible, but it’s much more subtle than that when properly installed on the PC.

This door is all that’s left of a completely deleted room. Don’t tell the purists!

The end result may be disappointingly dim, but I still had a blast manipulating the Tourian map in a way that wouldn’t compromise the basic layout, and I will definitely be using what I’ve learned on this project to make some more “window inserts” for this case in the future. Once it’s lit up, the design itself should really shine, but for now it still makes for some pretty fun close up photography!

The engraving fights with the inner bits just a little more than I’d hoped.
I love that the escape shaft coincidentally has its own murky yellow and green lighting.
This example clearly shows how dim the engraving is compared to the LED-lit components inside.
A two character 8 segment LED readout hides in Tourian’s O.
One Metroid, permanently frozen.