posts tagged with the keyword ‘openscad’

2019.06.15

stand-01

I’ve found that I prefer firing up OpenSCAD and writing some lines of code to actually trying to sketch simple 3D objects on paper. I grabbed a notepad and thought I could knock out a sketch, but then decided I preferred a model I could spin around and easily edit, and in no time I had what I needed.

stand-02

I’ve found that mocking up cabinets in OpenSCAD works for me. These are not final plans to fabricate something, just quick sketches to communicate ideas with others.

2019.04.28

mold-05

I made wax balls, and it worked, and I did it using 3D printed molds. I won’t get into why I want/need wax balls in this post, but I swear it has nothing to do with candle making or bath bombs. (These balls are about 12mm in diameter.)

model-02

I originally modeled one ball with one sprue, and then used the loop function of OpenSCAD to make a series of them in a row, slightly overlapping the sprues.

model-01

My original plan was to make silicone molds (like I did with this wax stick) and went as far as creating a positive and a mold box, but along the way I thought about just using a 3D printed mold…

mold-01

Here’s the 3D printed mold created using PLA filament. The holes are for bolting the two pieces together using 3mm hardware. (I used tape in the earlier versions, but it did not work well.)

mold-02

I didn’t need to fill all the bolt holes, but wanted a few options so I could get tight clamping. Wax doesn’t have the same low viscosity of something like water, but when melted is a bit runny, so I just want to make sure I can keep it from leaking out too much.

mold-03

Once the mold is assembled it’s just a matter of melting some wax and pouring it into the mold. I’ve had a few balls with air pockets when demolding, so I’ve taken to sticking a thin piece of wire in to stir around the wax in an attempt to remove the air. (I do have a vacuum pump which I’ve considered trying to use, but the chamber is currently too small to fit much in it.)

mold-04

Hey, wax balls! Originally I tried spraying the mold with mold release, but I don’t think it helped much. What does help is putting the cooled molds into the freezer for a bit (this is a known trick for getting wax candles out of glass jars.) It helps solidify the wax enough to make it come out fairly easily. I do break a few every now and then but a lot less than before I used the freezer method.

The other great thing about using 3D printed molds is that I can very easily (and cheaply) make a whole bunch of molds, which is good, because I may need a few thousand wax balls…

2018.08.11

sifter-3d-printed

I needed a sifter to remove some small pebbles from concrete mix, so I designed and 3D printed one. Now, like many things that get 3D printed, you don’t always get it right the first time. After testing the sifter I decided it needed larger holes, so. I started on version 2.

inkscape-drawing

For version 1 I used Inkscape to quickly create a grid of circles (using the clone feature) and then differenced them from a circle. Once I had this done I exported as a DXF file and used that within OpenSCAD to create the bottom of the sifter. I added a ring and Bob’s your uncle. I use this Inkscape/DXF/OpenSCAD/extrude method sometimes, because it seems like the right way to do things, or because it’s pretty fast. But it’s not always the best.

The problem with the Inkscape/DXF/OpenSCAD/extrude method is that you have to go back to Inkscape and redo your DXF file if you want to make changes or tweak the design. This may not be an issue for many designs, but for some, I want the flexibility to easily change things.

openscad-holes

So for the next version I redid the holes portion of the design in OpenSCAD. The great advantage here is that you can very easily tweak things like the size of the holes, and the hole spacing, and automagically see how it looks. (Note: in the image above you’re seeing the positive “solids” of the holes. In the final file I differenced them from another solid object to make them “holes” in the design.)

While the method of doing it all within OpenSCAD has advantages, the one way it suffers is when it comes to render time. I should note that in OpenSCAD you can choose how “smooth” circles are by applying a number between 1 and 200. It basically sets how many “sides” a circle has. You can typically use 100 and circles will look pretty circular, but you can also drop the number down to 6 for hexagons, 8 for octagons, etc. That said, at a setting of 100 rendering the sifter took nearly 15 minutes on my 6 year old MacBook Pro. Changing the circles to hexagons with just 6 sides took about one minute to render.

This is where The Cloud™ should save us, right? But while there are various versions of OpenSCAD running on public servers, no one is running a version that worked properly. And since OpenSCAD is a niche piece of open source software, I don’t know if this will change. But since there is a command line interface to OpenSCAD, maybe I can do complex renders on one of my more powerful computers. (Anyway, I’m getting off-topic, so let’s continue.)

openscad-sifter

Hey, it’s a sifter! Yes, I designed and 3D printed a sifter. I often design things and then print them overnight so I have them the next day. 3D printing is awesome, but it’s often not fast. I did consider trying to make a sifter plate with a drill press, and then I remembered I had a laser cutter, and considered using that, but ultimately I was not in a hurry and 3D printing one seemed like a good idea.

tinkercad

Speaking of good ideas, I’ve been looking at Tinkercad recently, and while I wanted to be convinced it wouldn’t be easy to do what I wanted, it actually was pretty easy to do what I wanted, so I redesigned my sifter using Tinkercad. The smart duplication feature made it fairly simple.

While I’m a bit more impressed with Tinkercad than I thought I would be, I still have the problem I had with Inkscape, that if I want to tweak some values, I need to redo work. In fact, I have to pretty much create a whole new model. For simple designs, this isn’t a huge deal, but it is a bit of a pain.

What’s a bigger pain is the fact that while I will always have copies of Inkscape and OpenSCAD to use, Tinkercad (and the files I create with it) may disappear. Tinkercad is own/run by Autodesk, and while it’s a great tool for beginners to get started with 3D modeling (without having to install any software) ultimately I’m concerned for its long-term existence (like any hosted/cloud service.)

I can easily edit the OpenSCAD files I created seven years ago, which is something that is important to me. That said, I do want to explore other software, because reasons, you know?

2018.07.10

My last mold making experiment was a good experiment, and if I learned one thing, it’s that the mold should be flexible to allow for the removal of the piece you are casting without destroying the mold or the casting. (Duh!) Instead of ABS I decided to print a mold using flexible filament. Now, to print flexible filament I had to do a small modification to my extruder lever to allow the insertion of a piece of PTFE tubing so that the filament wouldn’t bend over before it fed into the feed gear. Once I did that, I was printing flexible filament (very slowly.)

path-01

I decided to start with something less complex, and used a 2.5D object instead of a 3D object. I had an illustration of a hammer, which seemed to be a simple shape, with no weird inside parts or sharp corners. I started by opening the vector file in Inkscape.

path-02

Once I had the file open I used Inkscape’s dynamic offset feature to extend the shape into a larger piece so I’d have “walls” for my mold.

path-03

I did a Boolean difference between the original and the scaled part to get the wall piece, with the wall pieces being approximately 4mm wide.

path-04

I also created a bottom plate that the wall piece would attach to. I did not make it fit with pins or any other alignment method, I just ended up using tape to hold the pieces together.

piece-outside

piece-bottom

Here are the modeled pieces. The wall piece was easy enough to flex around, but seemed strong enough to hold up for the casting.

hammer-mold-1832

And here are the printed pieces. The flexible filament did not print as well as PLA does, and there were some rough parts on the top of the walls, but it wasn’t going to matter for these purposes.

hammer-mold-1833

hammer-mold-1834

I put the two pieces together and applied some masking tape to hold them. Since I was going to use plaster I knew that it didn’t have to be too water-tight to hold the plaster in place.

hammer-mold-1840

I then filled the vessel with the plaster. (Yeah, I have a big bag of plaster on hand for weird reasons, so I occasionally use it, though eventually I want to use concrete for these experiments.)

hammer-mold-1882

After letting the plaster dry for a few days I removed the tape and then pulled off the bottom…

hammer-mold-1885

…and then was able to flex the wall piece enough to get the plaster out of the mold, so that worked pretty well. Again, this was not a complex shape, which really helped as I just had to push the piece straight down through the mold.

hammer-mold-1890

And here’s my (tiny) plaster hammer. I like how it turned out. The edges are not perfect, but then, what is, Right?

hammer-mold-1852

I ended up making two of them, and with one I found that it was fairly easy to shape the edges using tools, so my 2.5D hammer (sort of) became a 3D hammer.

2018.05.02

wire-bender-1894

We’ve got a project at Brinn Labs where we need to bend some 16 gauge wire. The wire bends very easily, in fact, too easily, and you can bend it by hand, but you can’t really get nice curves. I looked up “wire benders” and found “fret benders” which people use to curve the frets for guitar building.

wire-bender-plan

So I found this video titled DIY FRET BENDER – $5 USD FRET BENDER and I was in such a hurry I didn’t even realize the guy provided a bunch of design files! I guess I just often assume people don’t supply files, so I took a screen shot of the design and then dropped it into Inkscape and…

wire-bender-2d

I just whipped up my own design quickly. Since I didn’t want to screw around with using the CNC machine for this, I just exported the DXF and extruded it in OpenSCAD so I could create an STL file suitable for 3D printing. Since there’s a slot and not just holes, it’s not the most fun thing to make with a drill press. Typically slots require a bit more work than holes, with filing and other time consuming hand tool work that is often best left to machines…

wire-bender-render

I designed three parts, and then printed the body and three spacers and six guides. You’ll notice a small lip on the guide piece. That’s to just touch the inside of the bearing so they can roll smoothly. The bearings? Yeah, tear apart that fidget spinner! We’ll need three bearings.

wire-bender-1893

We’ll also need three 5/16″ bolts and nuts, though you could certainly use 8mm if you’ve got those handy. Hey, look, we’ve now got a wire bender!

wire-bender-1901

There’s a little room for improvement on this version… The slot could be a little narrower, and I’ve found that without pliers it’s a bit difficult to tighten the nut. I fixed that by 3D printing some nut knobs so it can easily be tightened by hand. (I already had my own nut knob design file, but you can find plenty on Thingiverse and Youmagine.) No photo because I added it later. :/

wire-bender-1898

This was a really simple build, and since fret benders often cost $50 to $100 (though I saw one for $25 on eBay) this was pretty dirt cheap. I don’t know if it’s up to the task of bending frets, but it should work fine for the wires we need to bend.

wire-bender-1911

If I get around to it I’ll clean up the files and release them. You never know when you might need to bend some wire!

« Older Entries |


buy the button:

Buy The Button