Huh...not like a love story at all. It sounded a lot better in the original write-up. Whatever. Anyway, after sadly turning in my keys at the end of the semester, I thought I'd never play with volatile laser beams again...at least until next year. BUT! My school has a Design, Build, Fly (DBF) team - although we always managed to catastrophically crash just before the competition...go figure - and a team member wants to get an early jump on manufacturing procedures in an attempt to revitalize the project. So whaddaya know, I'm now the go-to guy for lasers. Booya.
So it's time for a new saga! For years we've had these old central air supply and return vents in our hallway and garage. They're old-fashioned, impossible to clean, and bent up from years of hallway soccer and such games with four kids in the family. In short, they're ugly now, and too old to find a direct replacement. Custom or even stock metal covers are pricey...and mostly hideous. So what is the solution? Drab to fab and all that jazz - laser cut a new one out of wood!
The leading image is the final product for our garage vent cover. The second image shows a before-and-after for the hallway vent. Still a bit of work to do on that one - cleaning up the wall around the edges and such - but not too shabby overall. All of this work was done at the cost of materials and a bit of time on a CAD system, totaling perhaps $20 - because my usual wood shop was out of the size stock I needed, so I had to go to an art supply store, where they charge double. Figures.
Step 1: Source a Good Design
And then came the first roadblock. There are thousands of really nice vent cover designs - run a quick Google search for "fancy vent cover" and see what I mean. Granted, there are thousands of hideous ones, too, but while we want something modern-ish and interesting, we don't want a busy design. After all, it's a 15.25" x 7.25" (9.25") cover on a wall at shin level that will blend into the wall. This ain't the Ritz.
And then we found our second roadblock...coincidentally right next to the first. Our garage vent (seen in Step 1) has vertical bars. It hides the largest duct hot-air delivery duct in our house - a whopping 6 inch hole in a sheet metal box. Beautiful, eh? But the vertical lines in the old vent fail to disguise this...and most of the designs in Google searches have large amounts of blank space between straight lines or curvy designs. Not good. Too much blank space in the vent will leave an eyesore very visible - even at shin level - and too little will block our airflow (we use what once was a garage as a den, sort of). Straight lines won't work, and we don't want a Fleur-de-Lis-type pattern that matches nothing in our house. What to do?
Optical illusion! In a random search, a new type of design appeared - one made of broken, overlapping, and intersecting circles. I was immediately reminded of the lovely ripple clock by fungus amungus - the overlapping rings broke up the circular edge until closer look, so this seemed like the way to go. Better yet, it's a simple, geometrical design, so it will be quick to cut and easy to CAD. Long search over! Hurrah!
Step 2: Concept CADding. Circles Can't Coexist! Chaos! Calamity!
In the image series, you can see my second attempt at this design. My first attempt was done in the opposite direction - rather than building up the CAD model and relying on alternating Boolean unions and subtractions, I started with a solid box and cut out the design. The second attempt added rings and then cut appropriate spaces - far more tedious, of course, but considerably easier to tweak and play with spacings. But both methods suffered the crippling issue caused by parametrization - at least, caused by starting with the shorter model and scaling in vertical size. Even though the garage vent is only 2" taller than the hallway, no matter how the spacing between concentric rings is defined, there is at least one ring that is missed by an operation. In the first method, one of the lower layer rings ends up existing within a previously defined hole, and in the latter, an outer ring exists outside the box and cannot be merged. This isn't usually a problem - you can modify the feature tree easily enough in Pro/E - but for this design by either method a failed operation would (by necessity) wipe at least 3 layers of work away, requiring rebuilding of the model from that level onward. Boo.
Ah well. I eventually went back (after I'd found ring and space thicknesses I liked) and un-parameterized the model tree, so this is a moot point now. Anyway, the image series shows my second method. The fantastically colored images show the build-up sequence in stages - these were the model layers I settled on based on the vent image, since I was trying to follow the same general layering effect. Then the CAD sequence shows my feature tree of alternating additions and cuts.
Fascinating, I know. Also, horribly inefficient. But hey, this model took only 15 minutes to build and a few minutes to tweak, whereas the rigorous first method was abandoned after an hour of work...and only half the rings. What can I say? When brute force works, it works well.
Step 3: Conversion to Drawing File and Readable DXF
Pro/E DXF files usually play nice with Adobe Illustrator (the cutter control program I'm forced to use...and despise), so I had no problems here. Simply drop your drawing view into your sized template page (mine was a custom 24" x 12" blank template to match the laser bed) and export as a DXF. I checked the DXF file in my standard vector editor (Inkscape) to make sure the file wasn't corrupted (occasionally happens in exporting from Pro/E Student versions - the program tries to include a "watermark" in the header), and brought the files up to the laser bed.
Step 4: Laser! Laser! Laser!
Anyway, I finally remembered to snag a photo of the laser bed and cutting process. The room is tiny (about 5 by 10 feet) and mostly filled by the laser bed's table and vents - the black towers on the plywood table are part of an old 2-axis foam hot-wire cutter (now broken), and the rest of the room is filing cabinets. Not the best - or most ventilate-able - area if your piece catches fire, but hey, free laser. I also have to be careful with the vents - the room next door houses some sort of chemical lab (one of the few rooms I can't access), so whoever works there has to pop into the laser room to shut the vents...to keep deadly fumes from building up near lasers and potential flame...yay...
So, for my first piece (a 15.25" x 7.25" cover for our hallway) I made a dumb choice. I found a great, cheap piece of 3/16" birch plywood in the laser bed size, but I forgot just how much a pain ply is to laser cut above 1/8" thickness. Even on very low speed and high power (5% speed, 100% power at 1000 Hz) with air-assist, it couldn't cut more than 90% through - I believe the laser needs to be recalibrated with the correct auto-focus package...since it was set up with drivers for a 30 watt model, not the correct 40 W. I ran a number of passes at different speeds and powers to try to get the last bit through, but finally had to flip the board, reposition the DXF image, and miraculously hit perfectly on the lines.
My second piece (a 15.25" x 9.25" cover for our garage) was trickier. I wanted to modify the original CAD model after seeing how the first piece came out - adding a few extra rings and shifting the long bars a bit for a better illusion - so obviously I couldn't do the cuts on the same day. I went back to campus a week later, but it was too early - the art store where I'd picked up the other board wasn't open at 8:30 A.M., I had to meet someone at 9:00 who needed some pieces cut, and it was pouring, so I didn't feel like driving back to the store after it opened. The campus bookstore was out of 12" wide boards, and only had thicker pieces. Argh.
So, while hurrying to that meeting, I bought two 24" x 6" x 5/16" basswood planks. Thicker than I wanted - the 3/16" ply board was sufficiently strong and a match thickness-wise to the old steel vent grate - but a considerably easier material to cut. I figured I had two chances - either the laser room would have a large enough scrap piece I could use as a finishing piece (gluing the half-pattern pieces on the back) or I could come back another day with more wood. I lucked out - there was an old piece of beech ply in 24" x 12" x 1/16", perfect for a facade board (left over by our Design-Build-Fly teams), so I slapped it in the cutter. The 1/16" ply cut in one pass at 50% speed, 100% power (1000 Hz), and surprisingly the thick bass took a pass and a half - at 50% and 75% speed, respectively, for the main cut and a cleaning cut...I'd placed the two 6" boards on the bed side-by-side to get a nice continuous pattern (second picture), so it would match the facing board with no problems.
Step 5: Post-Processing (Gluing, Priming, Painting)
The garage vent cover wasn't bad, either, even with the extra work involved. All cutouts dropped onto the laser bed, so the parts were ready for bonding. I chose a good heavy wood glue since both the bass and beech ply faces had very tight grains and they'd be screwed together in an edge-supported frame, so no gap-filling was needed; also, since I had a large surface area that would be unsupported in the center of the frame, I wanted a strong bond to withstand possible impacts to the center of the grate (i.e. little kids, dropped hammers, etc.). Gluing was as easy as sandwiching the boards between wax paper sheets, lining them up with both pieces slathered, and leaving it overnight with heavy books on a table as a clamp. Unfortunately, I forgot to clean the squeezed-out glue while it was drying (a sudden kayaking trip left it drying on table for an extra day...), so I had to chop off the hardened glue in the hole spaces and around the edges with a razor knife and pliers. Not so fun.
The hallway cover was a perfect thickness and would be below bump-out mouldling on the wall, so we didn't bother to round or chamfer the edges. The garage vent, though, needed some serious chamfering to counteract the thickness of the piece - 3/8" total stuck out too far, and the sharp corners would be right at shin height at the bottom of stairs...not a great combination. So, a few blasts with a belt sander took down the edges and highlighted the central pattern a bit more, and it was on to priming and beyond. You can still see the glue line on the back of the garage vent (second image) in the priming stage - we didn't do anything to fix this since (a) the face board hides this split on the front and (b) the cover is too thin for this to be apparent on the sides (where glue had flooded more fully, so the split is invisible), but you can fill this in with a smear of standard caulk or Bondo.
So, I partially lied. No chamfering occurred during this project. You can also see in that 2nd image that we rounded the corners - we also beveled the front (through the facade plate and into the basswood backing) for about 1/4" inward from the edges. This was not exactly to plan - a nice edge series of chamfers would have lent a more industrial look that would match our wallboard in the garage better - but a bit of over-zealousness with a belt sander led to this. Turns out, the rounded corners are much, much better for shins than anticipated, and it looks pretty good. Our wallboard has wide, vertical stripes with thin, rounded stripes in between (like beading, but continuous), so the beveled edges pick this up nicely.
There was one surprising snag after the beveling process - the garage vent cover had warped considerably overnight, to the point where it had extremely visible curvature about the short sides. There was no warping after the glue had dried and set for two days, so the pressure and friction heating from the belt sander must have caused the warp. Set "flat" on a table, the center would sit about 1/4" off the surface. Uh-oh. A bit of pressure on the edges could flatten the board, though - like the pressure caused by mounting screws - so rather than risk damaging the facade by trying to straighten the plate (between clamped heavy boards, etc.) we decided to leave it, paint it, and see if it would flatten against the vent hole.
Finally, a few coats of paint while the pieces lay flat were globbed on, and away they went to the wall. To avoid unsightly bubbles in the hole sections, we painted with a much thinner brush - an artist's acrylic paintbrush, actually - with a half-inch head, rather than the standard home wall brush (2-3" head). Getting a good coating on the inner edges of those holes was no picnic...they had to match the face color or they'd look like extremely deep shadows, adding far more depth to the circular illusion than we wanted. Luckily, touch-up can easily be done once on the wall. The hallway vent received a nice coat of interior white paint, while the garage vent looks smashing in pale butter-ish yellow.
Step 6: Wall Mounting...and Voila!
Mounting was easy. We simply used the old vent covers as drilling templates. And speaking of pop, the garage vent, once screwed into the wall, no longer pops at all - it sits nice and flush along the vent hole. Sure, it may crack sometime in the future or warp further, but then we can work on flattening or re-cut a new plate (from a single board). A little paint touch-up, and boom - fancy new vent covers with little effort.
You can't see the nice yellow color in the garage shot (1st picture, although you can a little bit in the zoomed-out 2nd shot), but you can see the vivid contrast the black mesh allows. Also visible is proof of the over-zealousness with the belt sander - it forced the left screw off-center in the outer bars, and caused a slight waviness to the top and bottom edges. Oh well - live and learn.
3rd and 4th images are of the hallway vent cover. Here, the old drywall ripped slightly as the original cover was removed, so we still need to do a bit of touch-up around the edges. A small bead of caulk and some paint will cure that easily enough. Interestingly, you can see a little bit of the wall behind the black mesh around the edges (no, that's not paint) - it's not readily apparent from this angle, but from straight-on you can see a second, inner white rectangle that mimics the back-most layer in the concept image. Unintended, but it's kind of a cool effect, and doesn't become visible until you walk close.
Welp, that's all I have for you this time. This saga still needs a bit of work - mostly minor touch-up and maintenance now - but overall we're happy with how the new covers came out, and they look much, much better than the old. Now we've got to replace a much larger walk-on floor register in our adjacent, wood-floored hallway...I'm thinking LASERS!