Our front living room is being decorated in a “natural history library” style, with our bookcases, our antique prints of bats, frogs, sea creatures, and of course a collection of rocks placed nicely in an old printer’s letterpress tray and hanging on the wall.

Looking at the rocks recently, I thought that it would be interesting to create my own stones using a generative approach to their design — morphing from rock shape to perfect cube shape — that I could 3D print and put with the real rocks. Having recently done a fair number of designs utilizing Blender’s various deformation tools, I knew this was not going to be a difficult project to model.

Seven generations in Blender

I started with a cube. This was the shape I wanted to resolve to, but it would be easier to work in reverse. I wanted to keep these small, like gems or crystals one might find in a fast-flowing stream, so I made my cube 10mm on all sides. I duplicated this cube and pushed a pulled a few vertices here or there. I duplicated this new form, subdivided it, and again pushed and pulled some random vertices. The lineage was apparent but it was starting to take on an interesting shape.

I repeated the process, and then started to get into some subdivision surfacing and decimation tool work, letting them smooth and roughen the shape as they wanted. After my 13th version, I stopped; I had resolved into a nice natural shape, reminiscent of any small rock in our collection.

I decided to delete every other generation as the larger jumps made the transition more apparent without being too different from the previous generation. This left me with seven generations of shapes.

As a lot of the rocks we find are slightly transparent, I decided to print these in natural PLA. Depending upon the speed and temperature it is printed at, it can emulate this texture quite nicely, or it can be surprisingly clear and more crystal-like. I also decided to print these with no infill — meaning they would be hollow — but with five extra “shells”. This gave them very thick walls, a hollow interior, and kept the PLA refracting light in even horizontal bands.

The estimation of print time after slicing in ReplicatorG was 36 minutes, so I got right to it. From starting cube to final 3D print and cleanup took approximately two hours.

After seeing my rocks next to the real ones, I decided to scale them up by 1.5x so that they were a little more in-line with the real rock sizes, making the root cube 15mm on each side. I also decided to 3D print them with only four shells instead of five, and at 223°C instead of 220°C, to make them a little more glass-like and transparent. Though the larger size took over twice as long to print, I think the new size is much better.

In the end I like how they came out, and my wife likes the way the PLA came out. I decided to arrange them from cube to stone instead of how I originally thought, as it felt better — it was they way they were created after all. I’ll either leave them in-line on one row of the letterpress tray, or intermingle them into the real rocks. Time will tell.

Winter came. And six hours later it was gone. Whew!

Winter has not stopped me from 3D printing quite a few new things. What I haven’t had time for, thanks to the craziness of day job, is finishing any of my big projects, or writing up stories around my smaller pieces. So, I’ve decided to just post some pictures and quick blurbs on what I have been able to accomplish, to serve as a snapshot in time and maybe some inspiration to someone looking for new things to try.

Yet another mundane thing

The house we bought had one of those dryer vents that had little plastic “fins” that cover the hole, and flip open when there’s air blowing out. One of those fins was missing and I knew months ago that I’d have to get a new one.

Upon measuring one of the existing fins I realized they were too large for my Thing-o-Matic’s heated build platform by about a centimeter. However, rotating the part 45-degrees on the bed allowed just enough space to get it in since corner-to-corner of a square is longer than the sides! Yay geometry!

I had some natural ABS filament which was a pretty good match for the existing fins, except for the years of outdoor wear, tear, bugs, and other “schmutz”. The end result fit perfectly, opens with the others when the dryer is on, and took less than an hour from design to completed (successful) print. (I attempted to print it three or four times without a raft, but my heated build platform seems to have a permanent warp on the extreme edges so a raft was necessary).

Whiskey time!

An old friend of mine lamented on Facebook that these nice whiskey glasses used in Blade Runner were pretty expensive. The rebuttals were that any hand-blown glass object will be expensive if it’s quality. I replied with a slightly sarcastic “Modelling and 3D printing in clear PLA in 3..2..”. About a week later I realized that I should, in fact, try to model the glass and 3D print it. It would be a good exercise in modelling a cool, but seemingly simple shape, testing if my natural PLA which sometimes prints clear can be really clear, and how to design something quite large that is meant for touching, holding, and everyday use.

I started in SketchUp as it was a very simple shape, but realized it was going to be a pain in the butt to get the insides right. I switched to Blender and quickly realized that the simple shape was not so simply. It was easy to create, but getting the subtleties were extremely difficult — lots of beveling, creasing, subdivision surfacing, and boolean operations were required. In the end I had a printable glass that wasn’t an exact duplicate, but more “inspired” by the original.

My first attempt at printing stopped about 45 minutes into it when I realized that I had to leave the house for a while and didn’t want to leave it going (I’m overly cautious like that). I did, however, see that there was no way I was going to get anything close to really clear with the natural PLA at the size I was printing; the small thin-walled items were great for going clear, but not larger thick-walled items.

Second print attempt, when I had plenty of time to let it go, went well until about an hour and a half into it when my filament got stuck on the reel and stopped feeding. What had been printed, however, turned out to be a great learning piece. Having made my original design based on the sizes I found listed online, printing and holding it made it clear it felt wrong in my hand — it was too large. Looking back at pictures from the movie, it never seemed like Harrison Ford was holding it comfortably.

I decided to scale down the object by a few percent, taking it from approximately 8.5cm a side to 7.5cm a side. My third print worked great all the way through its two and a half hour printing and I had a decent replica Deckard whiskey glass.

You can download the STL on Thingiverse.

Next?

I’ve been working on several other personal projects over the past few months, but don’t have any of them in a state ready to show. Most of them involve 3D printing some part of them, but also include a lot of electronics, so there are other things I’m trying to get right. A quick tease is this piece which is a hollow, two-part dried Italian poppy seed pod which will be painted and full of electronics. Why? I’ll explain when done.

One of the other projects needs some really finely-detailed parts which all my attempts on my Makerbot have proved too coarse. I’ve put this project on hold as I anxiously await my Formlabs Form1 3D printer.

Keep looking for inspiration and ideas, and happy designing and 3D printing!

The Pacific Northwest has proven to be amazing, and as we felt it was worth putting down roots here, we purchased a house! Of course, with any house comes the “to do” list. This time, though, I have my trusty 3D printer to help out.

Immediately there were two things that needed my attention, one mechanical, one ornamental. First was a sliding door guide — nothing terribly interesting, but easy to recreate, 3D print, and have it work like nothing was ever wrong. You can view and download this guide piece on Thingiverse.

Second, the ornamental, were some finials—you know those things at the end of the curtain rods—for the dining room curtains. The ones there were not ugly, just not our style. My wife looked around online and found some that she liked, but they were rather pricey. She said they were on Restoration Hardware’s site, so I went snooping. I found a few that were nice, but I saw some “hand forged” ones that were in the $50 range for two, and looked like something I could recreate and 3D print. So for fun I modeled one, 3D printed it, and showed it to my wife who replied “great – we need three more”. So I got busy.

This post will focus on how to create the “hand hammered” style using various tools, and my choice for finishing it the way I did; a bit of design process, and a bit of crafty craft.

Before I actually started my first ball finial, I needed to determine some sizes: the size of the existing screw holding the old finials in place, the diameter of the curtain rod itself, and the size we wanted the finial.

The screw was easy — I measured the diameter with my calipers and added a fraction of a millimeter, knowing to account for plastic shrinkage; what measured as 6.1mm diameter, I wrote down 6.2mm.

The curtain rod was, again, easy — 12mm on the smaller side of the rod, 14mm on the larger side. A diameter of 15mm on the inside of the finial would fit over both and allow a “professional” look.

Finally the finial itself wasn’t going to be an exact measurement, as much as what felt right. They needed to be substantial, but not as big as the old ones. But they couldn’t be too small because the drapes were thick velvet-like material and covered large sliding glass doors.

I started with a 40mm x 40mm x 50mm block as the size test; it wouldn’t take long to 3D print, I could test the screw and rod sizes, and not use too much plastic. In its own way, it was a neat, modern, minimal finial… a minial. It printed fine and gave me a chance to see if all my measurements were correct. They were! But in the end, the block finial wasn’t terribly attractive. Maybe as a cube. Maybe.

Knowing the technical measurements were good, I started looking at the size of the block. I was surprised to think that it was too small. Looking at it in your hand and then up on the rod form across the room, the difference is noticeable and you feel the need for weight. I decided to make the ball part of the finial 50mm in diameter, and the overall height around 60mm.

Design Time – Part 1

At this point I was going to walk through all of the steps to recreate the finial in Blender, but realized that it would be easier to just record a video showing the steps as they happen. The entire process takes less than 10 minutes. It’s probably closer to 5 minutes if I wasn’t talking.

So watch this if you want to see the process from start to finish, and then return for another, much more nerdy way, to create the finial. NOTE – it’s probably best to watch in full-screen and at least 720p if you want to clearly see what I’m doing in Blender.

As I mention in the video, after I 3D printed the finial, I finished it using a Modern Masters’ Metallic Paint named “Smoke”. After placing them on the curtain rods, we really liked the end-result. I feel the stem is a little thick for the rod, but we can always change the rods themselves later if we want.

Design Time – Part 2

As I mentioned above, I have another way to create the “hand hammered” finial that is way more nerdy than using a 3D modelling application: writing code in OpenSCAD!

If you really think about the object, it is a joining of a few very simple pieces: a sphere, a torus, and a few cylinders. These are the types of objects OpenSCAD can create in its sleep. The tricky part was going to be the random-like variations to the sphere to make it appear non-uniform and “hand hammered”.

If you’ve ever had a grinding/sanding wheel, the answer may have already come to you: randomly sanding down very small sections around the outside, rotate the ball, do it again. In real life, this is something I’ve done numerous times by accident, but now I can use the idea for creating instead of destroying. The question is, how to do this in code?

Turns out this was relatively easy. The idea was this: subtract a bunch of small, flat cubes from a sphere, at random points around the diameter. This is basically the digital code version of randomly sanding down bits on the ball. To test it out, I tried the following code:

difference() {
     sphere(20, $fn=24);
     translate([19.5,-5,-5]) cube(size=[2,10,10],center=false);
}

Sure enough it worked:

That’s no moon!

The only thing I needed to do was repeat this process many times at random places around the sphere. OpenSCAD doesn’t do random numbers like other languages, so figuring this part out was the trickiest part of the whole script. Basically, you create an array of random numbers – one for each rotational axis, with as many elements as iterations of sanding. Add a rotation command before the translate in the test above, iterating over how ever many “scuffs” you want, and you’re good to go! Once done, the result was pretty awesome:

“Hand Hammered” ball finial in OpenSCAD

Feel free to grab my OpenSCAD script to play with it, but be aware that rendering the file to create a STL file takes a really, really, really long time; your machine may be faster than mine.

I’ve also uploaded my Blender-created STL file to the same Thing, but I’d encourage you to create your own. As you play with settings and parameters, you never know what inspiration will be sparked and you’ll go somewhere you never thought of.

“Hand Hammered” ball finial in OpenSCAD with the flag set to show process steps

Customizer is a user-friendly web interface into a world of 3D objects created via programming language and not visually like how “normal” 3D modelling works. You see, software exists that let you code objects with high precision, using functions with easily understood names like cube() and sphere(), but also combining these objects in useful ways, like adding, subtracting, etc. These functions allow a developer to write scripts that create objects that can be modified and shaped by simply changing variables – this is sometimes called parameterized design.

One such application is called OpenSCAD and, as the name implies, is open-source software and is free! What’s cool about Customizer is that you can write scripts in OpenSCAD, upload the script to Thingiverse, and automagically have sliders, input fields, and other user-friendly UI items alongside a real-time, interactive preview — giving users the easy ability to use your script to make objects.

I decided to take a crack at making a script to create a most mundane object, but one that is easily parameterized: door knobs and drawer pulls.

While nothing game-changing, it is a good example of how many common things can be reduced to code, and a new way of thinking about allowing end-users to customize objects for their needs and tastes.

You can run my script in Customizer, as well as download the script to look at the code and run locally, here: Door Knob and Drawer Pull Maker.

The first place folks seem to be going to is one where customers are going to want the stretchy bracelets, the tiny squirrels, and the Yoda heads that 3D printer owners love printing. This, to some extent, is true; kids love these things and will certainly get the parents in a store (or a booth at a local craft/art fair). But these are not the big-ticket items that are going to keep a business open (the small versions really should be 25¢ to $1.00 at most; you know – prices kids can afford).

Then, there’s an assumption that they can charge a lot of money for some of these pieces when printed larger. This probably stems from seeing gear hearts on sale at the new Makerbot Store in NYC for several hundred dollars, even though you can already get really nice ones on etsy for $10 – $30.

Finally, the place where I see people’s eyes lighting up: adults are going to want to bring files of things they’ve designed, or want to work with someone on making a custom widget, and have them 3D printed.

I think these ideas are each valid in some way, but not things to start a business around; I think there’s opportunity for something more.

Yeah? Like what?

At this point in at-home 3D printing history, the people who are genuinely interested in, and have the aptitude for designing and printing their own 3D objects, either have a 3D printer of their own or a friend with one; know about and use large, online services like Shapeways; work in a place that has a 3D printer they have access to; or are saving up and are going to buy their own 3D printer.

Unless you are opening a Kinkos-sized service bureau with really high-end 3D printers that can do better prints than what you can at home, these people are not going to be your customers. And that’s okay.

I think one entry point, one way to appeal to the masses, is not to try to sell generic trinkets to them (at least not ones you, yourself, did not design), nor try to get them to design their own items from scratch. Instead, it is the personalized customization of useful things that will get you a market, and keep you in business.

Here’s the Thing

Let’s take a hypothetical example of one, potentially useful, customizable thing: cabinet door knobs/pulls. Generic, spherical knobs will run you about $1 each at a big-box store. While you could print these sorts of knobs, and sell them 2 for $1 to undercut the stores, it’s the next step up that it starts to get interesting. You can buy a cabinet door knob in the shape of a daisy for $4 at a big-box store, and one with a dog’s head run about $17 for a metal one. There is now market expectation of paying a higher fee for getting something fun and special for a kid’s dresser, or for the pantry closet where the pet food is kept.

Rendering of a flower cabinet door knob

Imagine you could 3D print daisy flower knobs in an assortment of colors–perhaps even dual colored with the right 3D printer–and sell them for $1 each, 1/4 of the big-box store price. There are 10 knobs on the generic IKEA dresser I have, so $10 for some cute knobs. That’s not a bad price, as it would be $40 regularly. Do that four times and you’ve paid for your roll of filament, of which these would use only a small fraction of. Double your price to $2 each, $20 for 10, and now you’re starting to pay for the filament as well as your time, electricity, etc.

Now imagine you have more than just daisies. Roses, poppies, petunias, sunflowers, lilies, and so on; very simple designs that all use a standard base. And you have 10 different colors of PLA filament. And you also know how to dip your door knobs into metallic paint, or that liquid rubber coating in a can, or brushing them with various leafings (gold, bronze, etc).

Now, imagine that you also have a Kinect “photo booth” set up in the corner of your store where a person can get their face scanned (or their kid’s face) and put onto a door knob. And you only charge $6 each, or 2 for $10. Can they even do that at a big-box store? No, no they can’t.

Now, imagine that they do this and then you say “not only can you get your kid’s face on door knobs, but we also have toothbrush holders, light-switch covers, ceiling fan chain pulls, and 10 more things you can put it on”. Sold. I’ve seen how people love putting their kids’ pictures on all manner of items — there is, and never will be, a shortage of loving parents who do this sort of thing.

These are things that people can’t get anywhere else, are specific to the customer which makes them feel special, and they create value in what the business owner does.

Cabinet door knobs. Really?

I know that it sounds like I’m suggesting creating a business around a really boring item. What I’m actually trying to get at is that if people start thinking creatively about mundane things, they will find areas that no one has ever been able to find before. Plastic cabinet knobs have been around for decades, but never have people been able to create their own custom knobs as quickly and easily as today. Yes, you could have sculpted a base model, made a silicone mold, and cast resin knobs, but that process takes many hours (perhaps days for things to cure), while scanning a person’s face, cleaning it up, and 3D printing it on a knob in a fraction of the time. (I’d posit that you could still cast items using a 3D print from a scanned item if you wanted)

As I alluded to earlier, I know that plastic filament is not the only expense of 3D printing. There’s also time to print, electricity, time to design, your time, maintenance, and so on. However, part of this is, really, a one-time expense — design a base that has a screw hole that fits common door knob screws, and put 20 different things on top.

Then, get a process down of how to scan someone’s face and put it on that base. Create a “plate” that has 10 knobs on it so you can print 10 at a time, or learn how to use software “multiply” options to do it automatically. Get a flow for taking finished prints and cleaning them up and getting them into cheap zip-lock bags; perhaps use fancy bags with an inkjet printed label if you want. Or display a bunch of them in plastic bins you can get a dollar store.

Finally, try to get these small items into the hands of potential customers. Set up a booth at a local craft/art fair, or maybe talk to any local “quirky” home furnishing stores and see if they’ll sell your items. These are things that want to be touched, and when folks learn they are 3D printed (again, the ones that have never heard of such a thing) there will be an immediate interest AND they see how it fits into their life (unlike a Yoda head).

3D printed door knob on cabinet door

I think there can be creativity in mundane. There are myriad household items that can be redesigned, customized, made better, 3D printed, tested, and in people’s hands with unprecedented speed. Starting there can lead to bigger and better things, and perhaps grow into a business with more longevity.

You can find my example door knob on Thingiverse.

Much to my delight, while I was working on other things, Thingiverse user shutay in Kuala Lumpur, Malaysia took the idea and ran, posting several new derivatives.

First was his awesome snail zookin (good photography too). In his description he says “The first 4 Zookins went down so well with the family, that I thought the collection could do with a snail.” That’s awesome, and the snail design is great! I’m glad to hear that people are enjoying them, and that they are sparking ideas. That’s my goal.

Then he followed up with this cute car, which he called a “carkin”. Perfect. He says he plans on making more carkins, and I hope he does. I hope other people do to!

Today I posted three more napkin rings, made for the season: a pumpkin, an owl, and a cauldron; a mix of fall and Halloween. I was trying to work out a bat, but I’m still having issues with the correct wing proportions — the napkin ring would take up the entire place setting! D’oh!

I really hope that folks keep going with these; they are such simple things, but sometimes simple can be incredibly powerful.

You can find shutay’s pieces on his Thingiverse page, and the new fall napkin rings also on Thingiverse.

[ MINOR UPDATE November 7, 2012 - Added a few points, in BLUE ]

I was prepared to write a lengthy piece about this bit of software, but I decided instead to go with a thoughts/pros/cons-type format, and a summary. Basically the “tl;dr” version, though it’s still kind-of long. I’ll give specific insights here or there, but nothing as narrative as normal. Just how I’m rolling today.

However, I want to start by saying that my previous post looking at Bonzai 3D was based on the then-current talk online about SketchUp’s sale to Trimble, and whether or not a free version might continue to exist. A few months later and it seems that we still have a free SketchUp, there’s even been an update, and there’s no more talk of it going away. I was personally never worried because the version I had still worked and would continue to work as long as my computer lasted. Onward.

General

• ViaCAD 3D is developed by a company called Punch! Software, which is part of a family of software companies (apparently), including Hoyle, Broderbund, Bicycle (cards), and Encore. I’ve had really bad experiences with purchasing Hoyle apps in the past on Windows, including battles with their horrible DRM–for a simple card game–that required many emails to customer support just to get the thing installed. While ViaCAD didn’t have any issues installing on my Mac, I can’t say speak to how it installs on a Windows machine; I’d be cautious just on past experience with another member of the “family” (one bad apple spoiling the bunch, and all)
• The basic version of ViaCAD 3D is $100, but the “Pro” version is $250. There is also a 2D-only version that is $40. The differences are shown in this chart. I am assuming the trial version is for the non-pro version, as there are no rendering actions included, though there are some deformation tools and surface tools; so not 100% which features, exactly, are pro or non-pro.
• Punch! Software also makes a higher-end, “professional” CAD app called Shark. The versions of Shark are $700 for the LT version, and $1795 for the FX version. Interestingly, the UIs for Shark and ViaCAD are nearly identical, based on screenshots on their site, and share a large number of features. There are many different features, too, so it’s nice to have a wide range of pricing/feature options.
• ViaCAD has the ability to import many files types, including SketchUp files. I tried:
  • STL – Success
  • AI (Adobe Illustrator) – Success if version 8 or earlier, error message if not (current Illustrator is V11)
  • SketchUp – Failure. Not one SketchUp file from version 7 or 8 worked. No error, just immediately came back to an empty stage. Bummer. [UPDATE] I was able to save a scene from SketchUp 8 in SketchUp 3 format (the earliest it allowed) and that imported into ViaCAD just fine, so there is a cut-off somewhere, but I didn’t try all variations.
  • The failure of importing files from recent versions of applications makes me wonder if there is continuing development effort to ViaCAD, or if Shark is where they spend their time.
• ViaCAD can of course export as STL, and does so with many new options, including sliders on how fine of a resolution to triangulate surfaces and such. I’m not 100% sure what all of the settings mean, exactly, but I was able to export with ease. The export settings dialog had a few visual issues (not being able to see numbers because the field was too small and the part after the decimal point was too long) but nothing breaking it.
• Punch!CAD has a community forum support site, but most forums are marked “Private” and inaccessible without registering. None seem terribly high post count.

Editing and Workflow

• ViaCAD starts with an empty stage with a simple axis-marker in the middle, and in isometric view mode; you can toggle it to perspective if you want.
• The UI is clean, with nested tools that you can drag all the way through to tear-off that group and place wherever you want on-screen. There are some inconsistencies between icon styles — you get the feeling some haven’t changed since V1 and others were created much later, as features were added.
• You need to turn on grid, workplane, and selection “gripper” options to make the UI much more familiar to 3D model making; I personally think these should be on from start, but I understand these are 3D options and not 2D, which ViaCAD also does. It would be nice in the “New from Wizard” process there was a “2D or 3D” question, and sets the UI accordingly.
• There are all manner of options for viewport settings, including the classic 4-up new (top, left, front, perspective).
• It’s pretty easy to grab a drawing tool and start putting shapes on stage, to push/pull faces, and do basic editing. But that is where the similarity to SketchUp pretty much ends. I found it rather difficult to draw things where I wanted them. For instance, I extruded a simple box and wanted to draw a circle on top to pull up. I couldn’t get the circle drawn on the box, despite messing with workplanes. I finally had to draw the circle elsewhere, and move it into position to then push/pull it.
• It seems that some things you draw you can edit the values in the info panel at the top after you draw them (like SketchUp) and have the new values reflect in the drawn item. However, with some things in ViaCAD, that is NOT the case; which features this works for, I don’t know for sure, and there’s no visual indication if you can or can’t.
• With some tools, if you try to select some items on-screen that overlap, where your selection is ambiguous, ViaCAD will open a selection dialog and let you pick from a list of possible choices. This is good, because sometimes in SketchUp it’s hard to select the correct item, but the item names in the dialog are seemingly arbitrary–”Part_226: Edge 1071″ vs “Part_226: Edge 621″–and can be confusing. I understand what it’s doing, but it’s still a lot to mentally parse when making a decision. What is nice is that as you roll over the names, the item it refers to highlights on-screen so you can select the correct one.
• Speaking of odd names, when you go to the edit menu, Undo and Redo functionality has the names of the last action performed so you know what you’re going to undo or redo; that’s great. The weird part is that these read as “Undo Fillet36″ or “Redo–EMPTY”. Easy to ignore, but messy.
• ViaCAD has the full spectrum of 3D boolean features that SketchUp (free) lacks: add, subtract, and intersection. Additionally there are tools to slice models by arbitrary curves or lines, and other nice modifiers. These are the main features I go to Blender for, and was looking at Bonzai 3D to handle.
• There are some great creation and editing features, including:
  • really, really good 3D blend and fillet/bevel options (creating rounded or faceted edges between to faces). I think the blend feature performed better than Bonzai 3D and has a few more options.
  • a countersink creator tool
  • a “boss” creator tool (think: extruded cylinder you can use with screws to connect two pieces)
  • the ability to hollow-out any object, turning it into a shell
  • lofting between curves
• There are video tutorials built-in for just about everything you can do, though not searchable per-specific feature (e.g.: the “boss” tool can be found in “Editing 3D Solids”). Note that the full install was 1.5GB, mostly due to the videos.

Summary

ViaCAD is a powerful piece of software, and totally worth the $99 for the basic version. However, SketchUp still wins for simplicity in creating and editing objects for a first-time, or basic, user. I found some features in ViaCAD acted in confusing and weird ways, and there was a lot of unfamiliar terminology, but that may be because I’m not a traditional CAD software user (I’m probably more of a “3D modeler”).

I believe with the investment of time, and watching the tutorial videos, I’d be able to become proficient in the tool and make some neat things. The question is if I want to spend that much time learning, as opposed to just doing.

So, I’d say there’s a split:

• If you don’t want to spend a lot of money, but get a powerful piece of software that will require an investment in learning how to do basic drawing, I’d totally go with ViaCAD.
• However, if money is not a concern, I’d go with Bonzai 3D as it is quicker to get up and running for basic things, and has good 3D solid model repair tools.

Both ViaCAD and Bonzai have advanced features (like NURBs) that you can gradually work into as you need.

[UPDATE] I let the trial expire and when you fire it up it will still let you create and import objects, but all save and export functions are disabled. All other functions seem to be enabled still, so I suppose that the trial is the normal version, not the Pro version (see above). After not having opened it up in a while, fresh eyes on the UI made me struggle to remember how to do anything; this is definitely one of those apps where constant use is the best way to remember how to use it, and not rely on UI cues.

ViaCAD is available from Punch!CAD directly, on the Apple App Store (2 stars), and through Amazon (3 stars). Pricing and ratings based on date of post (Oct. 1, 2012)

Here’s a screen-shot of some weird thing I made the very first time I used ViaCAD, as well as a pic of the object on my Thing-o-Matic (yes, I’m back to safety orange ABS).

A little over 100 years ago, a scientist created an early plastic called “Bakelite”. While still in use today for some toys and specialty parts, there was a time in the 30′s and 40′s (and possibly later) where companies crafted various napkin rings in animal form using Bakelite.

Bakelite animal napkin rings in antique store

Today you can find, like I did, these napkin rings in antique stores, as well as on eBay, etsy, et al. While somewhat common, some can still bring big bucks. I came across a rabbit and a bird, and checking online when I got home, saw that the rabbit can go quite high ($90+ in some cases), while the bird starts at $1 but can go up to $80+.

Now, perhaps unlike the rest of the country, my wife and I still use cloth napkins, so I can still see an occasional need for napkin rings (well, very, very rarely). Regardless, the charm of these vintage napkin rings grabbed me: they were incredibly simple in design, basically a silhouette of the animal with a hole in the middle for a napkin, and extruded somewhere between 1/4 – 1/2 inch.

To start to design new versions, I first checked online to see what existed previously. Obviously I knew about the bunny and bird, but was excited to find there were elephants, fish of various shapes, chickens, different birds, and even different styles of rabbits. If imitation is the sincerest form of flattery, I figured I’d start by recreating the bird I saw, as well as an elephant.

I could have grabbed a photo from the internet, brought it into SketchUp and simply traced the edges, but I wanted to start from square one. So I did a search online to see the average inner diameter of modern napkin rings, which ended up being around 1.25 – 1.5 inches (31mm – 38mm). I went with a size on the higher end of that scale — 36mm.

Drawing the circle with a larger circle around it, then removing the inner circle’s fill created the basic ring shape and I was up and running. The bird was first as it was the more simple of the two designs I wanted. A single, long “foot”, a tail, and a circle for a head with a triangular beak. One of the things I liked about the vintage ones were the rounded corners instead of sharp ones; these seemed to add a bit of whimsy to them and something I wanted to copy. Unlike the vintage ones, however, I did not put the small ridges on the top of the head, nor make the beak conical; I wanted to keep the first one really simple. The bird took all of 5 minutes to draw.

I extruded the bird to about 3/4 inch and, after switching to my Safety Orange filament, sent it to my Thing-o-Matic to 3D print. While it was going, I decided to start working on the elephant. I started with the same inner circles, easily added the “tail”, and then tried to work out the head. This proved to be quite difficult to do just by using SketchUp; ironically, I felt I really needed to sketch the shape with a physical pencil to get it just right.

By this point in time, my Thing-o-Matic was halfway through it’s process and I noticed that the edges around the feet (well, foot) of the bird had started to lift off the build platform. But I also realized that where it had gotten to was really as thick as it needed to be. A quick measurement showed around 9mm, so I rounded up to 10mm for the thickness of the next 3D print. I re-sliced, re-leveled the platform, and re-sent the file to the machine. This time — perfect. While the second bird was being produced, I finished the elephant’s head and was ready then to send it on as soon as the bird was done.

Making these two napkin rings was super fun as I got to sketch (both digitally and physically), think about how to turn simple shapes into cool things, and how I just saved myself possibly over $100 for the two Bakelite ones I saw in the antique store. They were so much fun that I couldn’t stop there.

I grabbed another piece of paper and started sketching out some more animals — or “zookins” as I decided to call them (think: zoo + napkins).

Hedgehog and turtle “zookins” in SketchUp

Based on the tail of the elephant, I knew I wanted to do a turtle, so that was first. Something about the turtle sparked a thought of a hedgehog and that little guy proved to be quite easy. My wife suggested a bat, but I totally blew that one despite multiple attempts at drawing different styles. A quick scan of my drawing paper, an import into SketchUp (scaling up so my circles were 36mm diameter), and I had an easy time of tracing my two, new, zookins.

I love when a plan comes together, and I can take something I see and translate it into something I can hold in a very short amount of time; having a 3D printer allows me to look at things around me and see many more opportunities than I had without one. Situations like these are great because they keep me on my toes — you never know when and where inspiration will hit you, and it’s great to be prepared with the right tools.

Click here to view and download the STL files for all four zookins from Thingiverse.

In this day of being social by sitting at home and getting online, it’s still nice to be able to talk to folks in-person, to be able to do filament swaps, or just hold something in one’s hands to understand it better. So when I saw that there was going to be a meetup, and they were focusing on 3D printers, I had to make sure I cleared my calendar and attended.

Meetup attendees viewing a presentation on DIY 3D printing.

The turnout was great, IMO, for a weekend meetup, having 30+ people with 6-7 of them being young kids. While in a few cases the parents were the ones there to see the 3D printers, the kids couldn’t resist playing with the cool toys and bracelets and things that were printing and on display. There was a presentation given on DIY 3D printers — a basic introduction to the concept, how they operate, resources available to learn more, and so on. It was a great introduction for folks new to the 3D printing world.

There were dozen or so 3D printers including Ultimakers, Mendel Maxs, Replicators, Thing-o-Matics, Cupcakes, Rostock, and one made out of scrap wood (more on this later). And, to my surprise, besides Johann who created the Rostock, there were several other 3D printing “celebrities” there — Emmett, the creator of the famous Heart Gears and various permutations, and a woman who I knew of because her horror story postings on the Ultimaker forums were memorable and what helped me avoid buying one . It’s really cool to know that these folks are local; there is a wealth of knowledge just outside my door and they are all into meeting up and sharing. Excellent!

I mentioned that there was one 3D printer that was made from scrap wood. When the gentleman brought it in (sadly, I forget his name… Adam, Andy… Bowers, Bowden… I really need to carry a notebook), folks had a “what is that?!” moment, but also were immediately drawn to it and started asking questions. Basically, it was a couple of long slats of wood with some smaller pieces glued together to hold the motors, some metal tubes to let filament through, and two telescoping arms that held the printhead at the end. The DIY-ness was apparent, with large chunks chiseled out, a mass of electronics screwed onto one of the slats, and pieces of tape here and there, but the ingenuity outweighed any lack of woodworking craftsmanship.

In speaking to the gentleman about his design, he explained that several years ago he was working with automated testing machines for cell phones — ones that would repeatedly move here or there and press buttons on devices — and that he was personally interested in the then infancy-stages of DIY 3D printing, and thought “I wonder if I can put a printhead on the end of one of these testing machines”. Genius!

He assembled his first version from some servos and “bits and pieces of wood and electronics laying around his garage” (I’d love to see his garage!). He said it didn’t quite work right, and eventually bought some stepper motors and other gear to make a version that did work. He got to the point of extruding and moving the arms, but had some heating issues and never got final calibration. He figured that since we hadn’t worked on it for a couple of years, he’d release the design into the wild as open source. Let’s hope we see it on Thingiverse

We had a really good, though short, chat about being inspired by things around us and using that inspiration to make something new. I think this is a powerful concept, one that I go on about all the time, and one that is really important for the young folks growing up with this amazing technology. Don’t just think the things on Thingiverse are the only things out there — imagine, design, and 3D print your own!

What was fun about this 3D printer was looking at it with the Rostock in the background: you could see a sort-of family resemblance — both assembled with pieces of wood and zip-ties, both very tall, both had a radical design unlike anything else around, and both the physical manifestation of someone’s imagination. In a sea of Makerbots and Mendels, these seemed so futuristic, so sci-fi.

The Rostock printer that Johann designed and brought to the meetup was of course a hit; it was the first thing you saw when you walked in (that, and the crowd standing around it). It’s always hard to get a feel for scale when you see things only online, and this was no different — it wasn’t as big width-wise and depth-wise as I thought, but it was taller than I imagined. It was quiet and fast. Johann used the command-line to drive it, so no fancy GUI tools yet

Johann was very gracious in answering any question that folks threw at him, and really making his vision accessible to all levels of user. I shot some video of the machine in action, along with Johann answering some questions about the way it works.

By the time I needed to leave, I hadn’t gotten a chance to talk to many folks. I’m looking forward to future Make:Seattle/Eastside meetups, especially if they get one started strictly around 3D printing. (They handed out survey forms to see if this was a good idea, and I hope the results were positive).

If you see a 3D printing meetup in your area, GO! You never know what you’ll see and who you’ll meet.

I loved my Nixon watch; it told me the time, it told me the date, and it looked really good in its safety-orange brightness. I always had people ask me “is what some sort of medical device”, and even “what does it do?” and I always loved replying “it tells me what time it is!” Then one day it broke. Not the electronics, but a small band of plastic holding the electronics to the band broke in two different places.

Unlike old, boring watches where the band could be changed out at will, this one was had a patented process applied that made the rubbery-plastic band physically attached to the hard-plastic “sled” using little loops of rubber, making for a cool, one-piece appearance. Unfortunately, this process also meant that I couldn’t just go out and get a new band for it, and the Nixon site no longer had the safety orange version.

I opened up my broken watch a month or so ago, and at that time thought that it was too daunting of a task to try to repair — the part had too small of features, I couldn’t get the electronics out without cutting the loops inside, and general worries that I’d totally screw it up. After a month of it sitting in a drawer, however, I realized that I might as well try to fix it; it was either that or let it just sit there, looking sad every time I opened the drawer.

I unscrewed the four screws holding the back panel (battery cover), then four more screws holding what remained of the broken sled to the electronics enclosure. Having it free, I got my digital calipers and started writing down numbers. The values were fun sizes like 3.36mm, and 2.14mm; no whole numbers here. I still worried that I wouldn’t be able to get the precision I needed when I got to my printer, but I pushed on.

Sled design in SketchUp

Taking my numbers into SketchUp, I pushed and pulled, added holes, added inset walls, the works. There was a really elegant curve to the part that went inside the band, and where the rubber loops went through. Since I needed to slice those loops off to get the original out, I didn’t need to put the small, elongated holes in this area and could leave it solid. I was shocked at how fast I was able to get the piece modeled and looking good; a testament to both the easy nature of SketchUp and working with it so much to know how to look at a piece and quickly build it out. There were several areas that I had concern over — the thinness of the inner-walls around the screw holes, and the overhang of the band attachment point. I figured I’d just print and see what happened.

Sled rendering in Indigo

To the 3D printer! I knew that I’d need support, so I added a raft, set the parameters for no extra shells and 100% infill, and hit “print”. D’oh! Errors! Given all the pushing and pulling I did in the model, it was no surprise that there were issues with it that I couldn’t see. I ran the model through Netfabb’s awesome cloud STL repair service and had it back almost immediately. Let’s try the print again. Success!

Long-time readers may notice that I’m printing in white (“natural”) ABS, even though I’ve printed before with the same safety orange as the original watch. I had white already prepped from another print, and just went with it. Part of me didn’t think it was going to work at all, and the other part thought… well… the other part of me was just plain lazy.

The piece took 33 minutes to print, with a portion of that being the slow raft. Still, from the time of sitting down to take measurements to the time the print was done and the raft peeled off was less than two hours. That kind-of thing always makes me happy; it’s just crazy to think where we’ve come.

The screw holes looked good, though the outer overhang didn’t actually touch the raft supports, so there were a few dangling noodles to clean up. Time for the big test — does it come close to fitting? No! Well, the overhang fit perfectly in the band, but the main part wouldn’t fit around the electronics enclosure.

So, I was worried about the inner walls of the screw holes being too thin, but as it turned out, after printing they were actually too wide. Despite printing with no extra shells, there were two extrusions per wall, and that made them just a wee bit too big to fit in their areas around the electronics. I had to take an X-Acto blade to the walls and slice off just a little bit. Once fitted around the electronics, I was able to put the screws back in that held the sled firmly in place. I placed the battery cover back on and screwed in the remaining screws. It worked!

The next, and final step, was to glue the sled back into the watch band. Since I had sliced off the rubber loops (and had no method to actually re-attach them anyway) I resigned myself to just using some super-duper glue substance. I had a tube of crazy-strong “gel” glue, but it was old and had hardened. I also had some model airplane glue (from a previous project, not from making a model) and it seemed fine. I squeezed a little in the band and put the sled in place. And waited. The next morning the glue had dried! But it hadn’t actually stuck to anything. Bummer. And now it sits, waiting for me to go get some better glue options, maybe even an epoxy.

Whatever glue I end up using, the fact that I have been able to repair my most favorite watch with surprising ease, made possible only because I have the most awesome tool ever — a 3D printer, is cause for much celebration.