Little half-dog: The 3D printer that we ordered from Shanghai on a Sunday in June made its way to South Carolina on Friday of the same week. The Creality3D CR-10S Pro required only minor assembly, and was ready for alignment in a half-hour or so. Alas, that is when the trouble began. We didn’t fully understand the alignment and leveling instructions, although these were well-illustrated in the small user manual. Initial testing, using the sample file that came with the printer, produced globs of twisted filament. Our best result on that first day is affectionately remembered as ‘little half-dog’.

    The next morning we revisited alignment and leveling of the hot-bed, taking care not to overlook any detail in the instructions. This effort produced a small dog that was not headless. (no photo—the original artifact is presently in the hands of a grandchild.)

    From not working at all, to producing a satisfactory print most of the time, was not a linear progression. Over time we learned some ‘tricks’ or maybe superstitions that seemed to improve the likelihood of success. At one point we tried using a glue that promised to cause objects to adhere better for printing, and also to release more easily when finished. However, after several days use, we concluded that while the glue did not help significantly, it did contribute to the post-print cleanup chore.


    First useful print: While waiting for the printer to be delivered I had worked through a few FreeCAD tutorials, and had also created my first ‘original’ object. In the electronic parts closet were several small (2-inch) speakers from Jameco. These 8 ohm speakers produce decent sound in the frequency range of speech, but they do not have any sort of mounting frame—There are no screw holes. After measuring diameters and depths etc. I designed a custom bracket, shown on the right as rendered in FreeCAD. The completed result was encouraging. I have since included this 3D-printed bracket in two projects, one being the example enclosure pictured at the top of this article.

    Mix and match: In the past I relied on manufactured project boxes for most small electronic projects. The boxes themselves are fine. However, the challenge has been to drill holes and make smooth cutouts for switches, pushbuttons, LCD’s or OLED’s etc. I do not have machine tools or the skill required to use them, and the scroll saw does not make a clean cut in project box plastic, at least not in my hands. For some reason I did not immediately consider 3D-printing a complete project enclosure. Instead I thought to replace the lid or top of a manufactured enclosure, the part that would have holes and cutouts.

    Careful measurement: The photo (left) shows the inside corner of a manufactured (Steren) project box. The edge that mates top to bottom has three levels. I used micrometer calipers to measure their widths.  Then I placed an index card along the lip and made a pencil mark on the card for the middle part height. There are surely other ways to measure small things precisely. The point is that in order for a 3D-printed top to fit a manufactured bottom, the measurement should be nearly exact.

    Notwithstanding the care taken, I imagined it was unlikely this try would work, and therefore made a design cutout in the top to conserve filament and printing time (right). As it happened, the lid did fit. It was only necessary to file the corners a little, as I had not thought to filet them in the design.

    On seeing the test top snugly in place on the manufactured bottom, I thought of another idea. Instead of reprinting a no-cutout top, with holes and accommodations for project components, it would be more efficient to 3D-print a separate panel that could be inserted into the large test cutout. In that way, if there were mistakes, only the panel insert would need to be reprinted. By this stage of experimentation the original filament spool had run out, and a different color PLA had been loaded. The result is shown in the photo (below left). It is a little rough, but everything fit.

    Spreadsheet epiphany: Different projects come in different sizes. It seemed there should be a way to specify overall dimensions independently of other details, so that the same design (corner screw posts/holes, top-bottom mating, etc.) could be reused. I Googled this question and found an immediate answer. FreeCAD has a built-in spreadsheet. It is there for the looking, but I tend not to look very deeply when fiddling with something new. So I re-did the enclosure top, and also a bottom, which is nearly the same as the top except for depth, but this time parameterizing the primary dimensions in Spreadsheet format. While FreeCAD’s built-in spreadsheet is not Excel, it does include some very useful features. One can define expressions to compute derivative values, for example. And any value (cell) can be given a symbolic name or alias, so that if additional rows are inserted, or if rows are removed from within the populated part of the spreadsheet, it is not necessary to pay attention to shifting cell ID’s. Another virtue of the FreeCAD spreadsheet is that it can serve to document a design, recording what each value means, or how it was obtained, etc.

    A digression: I thought it would be fun to construct a canonic illustration of spreadsheet usage in FreeCAD. This led to a digression having nothing to do with radio or electronics projects. In another article I described a game called Pentominoes invented by professor Solomon W. Golomb and popularized by Martin Gardner in his Scientific American column (ca. 1957). The game involves fitting Pentomino pieces on a checkerboard (or chessboard). In order for the game to work, the unit square size of Pentomino pieces must correspond exactly to the board square size. Of course boards come in all sizes, so a set of Pentominoes constructed for one board will not fit a different sized board.


    The printer had idle time—not much, but some! Why not print a set to fit our chessboard, during hours when the printer had nothing better to do. The design could be generalized to allow reprinting pieces of any reasonable size. Over time this aside project produced a multi-colored set of the twelve pieces. It may be worth remarking about something that is conspicuous in the ‘T’ Pentomino print photo (left). The piece itself has a border called a ‘brim’ that is generated as an option in the slicing software Cura. Brims help adhere flat objects securely to the printer hot-bed. (Another support structure called a ‘raft’ performs a similar function for objects that do not have a flat bottom.)

    Free downloads: As one might guess, a great many 3D print files are available for download from a variety of Internet sources, the best known of which is Thingiverse. Some downloadable 3D print objects make intriguing demonstrations. Others are merely ‘cute’. Truly useful objects are perhaps a minority. One from the latter category is an enclosure for the Atomic Pi. My wife downloaded and printed an enclosure for her Pi (right). This downloaded enclosure is considerably more complex than any we have designed ourselves.

      Prototyping: Typically before completing the design of a full-scale project enclosure we test the fit of individual parts.  For this we make a small plate or bracket that accommodates just the part and nothing else. Then for the main project it is only necessary to replicate measurements from the verified test in the main project. The composite image below shows one such test, where a capacitor shaft must be positioned to align precisely with the Vernier dial (or conversely). Mounting holes for both components must also be perfectly aligned.


 It may take a few tries to obtain a precise fit, but little material or time is wasted in testing. Occasionally it may be more efficient to test two or more things at once.


       Joining pieces: While I have seen some impressive multi-part constructions that were 3D-printed all in apiece, it is also possible to print parts of a thing separately and glue or otherwise join the pieces together. I have done this a couple of times. For example, the photo at the top of this article shows the GPS antenna slotted in an ‘L’-shaped bracket on the side of the enclosure. That bracket was printed separately and glued on. This approach is similar to gluing a panel in a cutout, or a recessed tube in a hole.

       Material and detritus: So far I have tried only PLA filament. According to its manufacturer, the filament we are using is ‘earth friendly’ and ‘biodegradable’. It is certainly tough, but warps at high temperature. Higher temperature filaments would serve better for some purposes, but are more challenging to work with. Some require venting the printer, which would be a project in itself. While PLA would not be suitable for making (outside) antenna parts, as I once considered, I do plan to test its durability in a limited outdoor setting, namely on our small sailboat, but only for short day sails.



    At this stage, having been introduced to 3D printing just two months ago, we remain focused on basics. FreeCAD has many features that I have not examined or studied yet. The slicing software also has options we haven’t tried. 3D printing is not ‘plug and play’. There is much to learn!

       Photo Gallery: The following photos show test or finished prints at various stages of conception and execution, and sundry related stuff.

---

Project descriptions on this page are intended for entertainment only.The author makes no claim as to the accuracy or completeness of the information presented. In no event will the author be liable for any damages, lost effort, inability to carry out a similar project, or to reproduce a claimed result, or anything else relating to a decision to use the information on this page.