Designing brackets for cLeg cosmetic fairings

A continuation of: Blending and Bending Prosthetic Leg Fairings

I wanted to design a 3D printable file that would allow me to clamp my acrylic fairings to the cLeg. Luckily Kay and Gabe created a pretty solid model of the cLeg itself so I could use its contours in my design. Again, I had to cut the model down manually in Blender because netfabb does an awful job of exporting STL (hadn’t yet investigated if this is an actual or a free-demo-imposed limitation of the software).

I also used Blender to multiply the scale by 1,000 before importing into Tinkercad.

I like Tinkercad because it makes it super easy to perform boolean functions on complex stuff like our leg here. If you’re a tinkercadder, try ungrouping the model so you can see how I made a cross section of the leg: Tinker This

And I went on to improvise a way to screw two parts together — this is very much a test fit, to see if the idea makes any sense, to see if screwing two plastic pieces on with pressure is enough to keep it from falling off.

I went back and forth about 3D printing holes for nuts and bolts, but decided I would drill it out of the finished plastic — “machining” is way more accurate than printing for this kind of thing.

OK, the original Kinect scan was not very dimensionally accurate…I guess I could have compared measurements of the model with measurements of the object, that’s what I get for trusting technology -.- I can still use the blocks of plastic for practice fitting, though. They don’t clamp around the c-Leg, but I tested screwing the pieces together and drilling into the plastic to insert a brass threaded insert, and with a few taps of a hammer that turned our really nice.

Of course, the chunky tinkercad design isn’t so fashionable, but it served its purpose to confirm that I need to press the shape of the leg more accurately (the front fairing doesn’t sit flush with the printed model at all) and that countersunk M5 screws will do a nice job of securing the fairing to the bracket, though I’ll have to play with melting a countersink so the head of the screw doesn’t stick out the back.

Continued here:

Hackerspace on Treasure Hill

Visiting FabLab Taipei and OpenLab Taipei

Though I griped about Chinese internet censorship, FabLab Taipei was the real reason I hopped on a plane for a detour the last few days of my trip. Their facebook group is always sharing amazing projects and a video I saw at Fab11 of their Maker Faire showed a big, active, and creative community. I messaged the group the day before I arrived but didn’t hear back, so I got in touch the old fashioned way and showed up at their door.

I was greeted by a man named Sega Liu, who was working on his iteration of a fold-up 3D printer. He gave me a tour of the multi-room facility, which was punctuated by a number of surprising projects: incredible corrugated cardboard & resin surfboard (Ah cardboard, always reminds me of home), intricate laser cut wood paneling and an impressive cardboard-folded giant frog mask. The lab was stocked with capable machines, including an assortment of 3D printers, the familiar Roland circuitboard milling machine, and a very heavy duty CNC machine that doesn’t get much use (it runs the unfortunate Roland software, though it accepts plain text serial data not so different from GCode, so perhaps the more modern & open source ChiliPeppr could be modified to control it.) There’s a sizable laser cutter, lots of electronics equipment, and in the basement: a comprehensive wood-shop.

I felt immediately at home in the space, which had a lot in common with CUCFabLab: it is open to the community relatively late at night (every day 6–10PM), has every machine you could possibly need to make awesome stuff, and a lot of the furniture, well, has a hand-me-down feel (which I like better than the try-not-to-scratch-the-table newness of some places. Gives it character 😉

I chatted with Sega about my work back at CUCFabLab. It always surprises people to think FabLabbing could be a full time job. FabLab Taipei, like many hackerspaces, is all volunteer. I told Sega about how CUCFabLab has grown from an all volunteer staff to employing a dozen people at times (though I’ll write another time about the trade-offs in establishing a paid staff among volunteers). At some point in all this he decided that Ted, the main benefactor of the lab, would want to meet me and called him back to the lab (I guess I had just missed him).

Ted told me about the history of the lab, and how they use to have membership fees but eventually switched to free membership after establishing that it was a hard sell and wasn’t helping much with rent anyway. Apparently, Taipei is awash with low-cost designers and prototypers to the point where even university students don’t design things for themselves, they can hire someone to work for them for the price of a typical hackerspace membership. It is an interesting problem in Taiwan as in China, when building things is seen as something of a lower-class occupation: how do you catch people’s interest with the prospect of “Do It Yourself!”

Of course, the premise of a FabLab is to create something new, and in my experience learning to use a machine gives you a lot of ideas about what can be created with it — so if you want to design something (even to have someone else build it), if you want to be innovative, you could start by learning the various styles of fabrication. I think as people’s interest is piqued by the surprising inventions exhibited at Maker Faires and workshops, they will be enticed to try it themselves.

After telling Ted that I was interested in learning about as many maker communities as possible, he said “We better get over to OpenLab then, you can come to FabLab any night but OpenLab is Wednesdays, lets go.”

Exterior of FabLab Taipei, Interior of cab across town.

Ted hailed a cab to take us across town to OpenLab Taipei, past a temple and through narrow corridors. I had just read about this neighborhood earlier that day: an illegal settlement of improvised architecture that the city chose to preserve and transform into a collective of artist studios called Treasure Hill. I was surprised to find out that a hackerspace was nestled among the concrete cubes stacked up here, but it certainly fit in among the artist studios.

The most striking project sitting on the shelf was a 3D printed Immorton Joe mask complete with ribbon cable hairdo. A very DIY shortwave radio built with point-to-point soldering on a copper board was being tuned into Chinese music. Scattered on the table were other projects in progress including a recycled CD drive pen plotter. A pile of vintage diagnostics gear completed the hackerspace aesthetic, and an impressive laser-etched stamp stood ready to print more banners for promoting the space.

We were just here for an hour before people were packing up. Ted and I took the metro one stop to the Shida night market to keep chatting about maker communities. There was some fantastic barbeque chicken (was that cinnamon? Five spice?) and instead of bars in this area there are convenience stores with barstools in a sitting area — kind of neat, you pay convenience store prices for a bar experience. We talked about 3Nod underwriting Fab12 and how Fab Academy would work for a Chinese student body (the material will be translated on a rolling basis and Chinese fab labs will work a few weeks behind English language labs).

The next day I visited the Treasure Hill artist village again, this time in the daylight. It has fantastic art scattered around the open areas and on top of roofs. I peaked into a small cafe called Tadpole Point and saw they had burgers with fried egg on the menu so I put in an order and browsed the art books lining the wall. The room had a great view out the screen door overlooking the interstate and the valley. According to wiki this location was originally fortified for anti-aircraft defense, but it’s a good place to enjoy the scenery, too. With a great burger served on a solid wood butcher block and a breeze through a window of the poured concrete room, I reflected on how cheap materials can be arranged so elegantly, in the case of architecture and for food.

Unfortunately I didn’t get to find out more about the Conductive Ink Experiment nor the More Than Useful Detective Lab.

Coding in Shanghai: Javascript Hackathon

One Friday night, googling (or, baidu-ing?) for “things to do in Shanghai,” I found out about a hackathon being held the next day: 24 hours using MeteorJS to build a web app. Perfect! I’d built a dirt simple sign-in sheet with NodeJS and Javascript, but my code was already getting messy so I shelved it until I was ready to rewrite it from scratch. The hackathon was just the excuse I needed to try a modern framework (i.e. someone else’s code that does most of the work for you) to rewrite my sign-in sheet for the FabLab’s many machines. And hey, if it’s polished enough, maybe other FabLabs would want to use it, too.

I was finally getting used to Shanghai’s metro system and found the Agora Co-working space without any trouble. It’s a beautiful office with a garden out back and the wifi was the fastest I’d gotten anywhere in China. I liked the place right away.

This was my second hackathon (after the People’s Music School Hack Music) so I was a little more proactive about introducing myself and what I wanted to do. Pretty quickly, a small team assembled around my laser-sign-in sheet project, but it was a bit every-man-for-himself as far as learning MeteorJS goes. We were all there for the same reason, to try out the new framework, but without any kind of guided introduction / classroom situation, we were left to read the docs and getting starter guides, which isn’t really a group project.

But after a couple hours, I felt like I was learning a lot and my teammembers felt they had worked through enough of the tutorials that were ready to work on something together, but then we were all new at collaborative coding, too! Especially for a one-page app, it’s difficult to divvy up the work. We didn’t figure that out until after an hour of learning to use github (another good learning experience, though, cloning and committing and all that, that’s definitely a good thing to work through with other people in the same room instead of just reading a getting started guide.) Once we could finally push and clone each other’s code we arrived at the question of how to split up the work. We all took a stab at creating a good foundation to work up from, but were all using our own preferred approaches and stepped on each others feet when we tried to merge our ideas together, so when were getting settled in after dinner and my team members asked what we should work on, I said:

“I want to suggest that we all do whatever we want to”

And that worked out pretty well. A sign in sheet is a simple enough app that we could all write it from scratch, and this way we could experiment with the approach that interested us instead of trying to come to agreement on what tools to use together. So from then on out we worked on our own thing, snacked in the kitchen on occasion, and put on our headphones to focus whenever we wanted, and I think we all had a good time. I stayed up late and slept on that giant beanbag in the first picture.

In the morning I showed my progress to a few guys (it works, but it doesn’t have a way of saving the data or reseting each day, yet), but had to bid them farewell before the end of the Hackathon to get across town for the Shanghai Maker Carnival. It was great to find some friends in such a big city and feel like I was on a programming team (however dysfunctional) for a day.

Pictures posted on their Meetup by Julian:

Voice Controlled Web Apps Pt. 2 (Windows)

NodeJS, Client Side Javascript, and Google Speech API

Part 2 in a 5 part series. Part 1 is here.

This section of the tutorial introduces “client side javascript” — running code alongside HTML in a web browser. We won’t be building a beautiful site. Instead, I’m going to introduce as little as possible to get results so you can move onto controlling lamps and motors and other internet-connected-things. Hopefully you can combine this with your own knowledge or other tutorials to build the application you want.

At this point you should have two files, a server.js and an index.html sitting in the same directory.

For this part, we’ll just be working with index.html, so you could simply point your web browser to C:/your/file/path/index.html and it will work fine, but why not get in the habit of starting your node server? You’ll see changes to index.html reflected when you refresh the page either way.

The history of HTML is a jumbled one with many actors. Lots of people were taking stabs at building websites and lots of people were writing web browsers that did their best to render a useable website from everyone else’s code. The modern day result of this is that the web browser is a very forgiving platform that lets you leave all kinds of stuff out, and lets mistakes go without complaining too much. That’s actually one reason why I’ve heard some people discourage javascript as a first programming language: you can make a lot of mistakes and there’s not a compiler to tell you what you did wrong, the web browser will try to run anything it can and ignore whatever doesn’t work.

Use the “Inspect element” to find out how Chrome interpreted our 4 lines of HTML:

Our file only contained <h1> tags and the <script> tags. <html>,<head>, <body>, and this mysterious “#shadow-root” were all added by the web browser before rendering. On top of that, in the bottom-center there’s styling added on to the <h1> element: all of this display: block; font-size: 2em; -webkit-margin-before etc. is added by Chrome to make our barebones HTML code look presentable. So like I said, you can leave a lot out and modern web browsers will figure out how to render your page anyway.

OK, I promised an interactive web page, so let’s change that header <h1> into a button <button>. If you don’t know: in lots of text editors Ctrl+H will bring up a “find and replace” menu. We’ll also add an “id” attribute inside the opening button tag: this is a name you give individual elements so you can write javascript that acts (or listens to) those elements.

Here, ‘onclick’ listens to that ‘myButton’ element, ready to react to a click at any moment. This is different than, for instance, programming an Arduino, where we’d have to check whether a button is pressed continuously in an infinite loop. Instead, we can write a function that executes only in response to the web browser noticing that the user clicked their mouse while hovering over the button element. This is “Event driven programming.” (You actually can do event driven programming with Arduino, if you know about hardware interrupts.)

While chaining those identifiers and functions all together on one line works OK, programming like that quickly gets hard to read. It’s much better practice to break it into pieces.

  1. Create a variable called button and assign it the myButton element.
  2. Create a function called speak, which calls on the alert function to say hi.
  3. Attach the ‘onclick’ event listener to the button variable and assign the function named ‘speak’ as its action.

I included two ways you could create the speak function here that would work equivelantlly in this case. There are specific cases where you should choose one or the other that involve global vs local scope (great info on the subject here), but for now it’s just important to know that there’s more than one way to do it. Hopefully keeping that in mind will make reading other people’s javascript examples less confusing.

Buttons default to ‘block-inline’ style, and appear on the same line

Let’s move on to a button that changes the background — this will set us up for turning a lamp on and off. The buttons have separate IDs, and each button is assigned to a new javascript variable.

Header elements default to ‘block’ style and appear on separate lines

By the way, you don’t have to have a button element to be able to click on it. I like the look of h1 better, so I’ll just click on those. I’ll change the color style of the h1 elements with each click as well. (side effect: button elements default to a block-inline style whereas header elements default to block style)

In the next section, we’ll plug a lamp into an Arduino:

Recommended Reading:

Note: “Handling Events” recommends using ‘addEventListener’ instead of ‘onclick’, which is a better way to do it. Onclick events are limited to one-per-element, whereas you can ‘addEventListeners’ as much as you want. But for a simple example I like the syntax of .onclick better.

Voice Controlled Web Apps Pt. 1 (Windows)

NodeJS, Client Side Javascript, and Google Speech API

This series will introduce newcomers to web programming using a few of the latest tools available. I’ve dabbled in building web sites since Microsoft Frontpage 2003 came out, but always ran into frustrating inconsistiences (casualties of the browser wars) and intimidating combinations of programming languages to start a server and handle user interactions (PHP, AJAX, HTTP POST and GET, yuck).

Recently I’d been hearing a lot about Node.js, software for building web servers with javascript. When it was introduced at JSConf in 2009, it received a standing ovation. When was the last time you heard of a new technology receiving a standing ovation? With all of the hype surrounding NodeJS, I decided to dabble again and was swept off my feet by the fact that I only have to know one language. Javascript is used to create the web server, and javascript is used to make the web page interactive. With added libraries, we can use javascript to control lights, relays, and motors with the use of Arduino, Raspberry Pi, Galileo, and others, and we can use the relatively new HTML5 web speech components to control all that hardware with voice commands!

Part 1 of this tutorial will walk you through installing the bare minimum on a Windows system (linux and MacOS instructions will come later) and serve up a ‘hello world’ HTML page.

Part 2 will introduce client side javascript — that is, programs that run in a user’s web browser to make the page interactive. We’ll start with a button that changes the background color.

Parts 1 and 2 could be switched, you could start with html and then learn how to start the server, as long as you do both before part 3.

Part 3 will use the excellent Johnny-Five library (meant for programming robots with javascript) to allow our webserver to control an Arduino. It could just as easily control the GPIO of a Rapsberry Pi, BeagleBone, or Galileo. Socket.IO will be uesd to send messages from the client to the server.

Part 4 adds voice control to the mix, explaining how to use the Annyang library to execute javascript functions (and therefore control attached hardware) within the web browser. I’ll also cover creating an SSL certificate here so the browser listens continuously without timing out.

Part 5 moves all of this functionality to the Galileo platform, which can run our web server and interface directly with other hardware, via digital I/O and serial communication.

Without further ado:

Install the Basics

Visit the official NodeJS website and download the latest. I’ll be using 4.2.2, hopefully they won’t change too much by the time you read this.

Pick out a good text editor: I’ll use Komodo Edit because it’s open source and lets me use Vi keybindings, but Notepad++ and Sublime Text are popular. Any of these will give you helpful features like syntax highlighting and auto-completion which make programming a million times more pleasant than trying to use notepad.

Go through the Next →Accept →Next →Next →OK →OK →Finish rigamarole of Windows installations to get NodeJS installed.

Next we’re going to do a javascript hello world. NodeJS is simply a program that executes javascript files. We’ll write a command in a javascript file, and tell NodeJS to run that file for us. This first program is just one line of code:

console.log("Hello World");

Type that into your text editor, and save it as something like ‘firsttry.js’ or anything you want as long as it has that .js file extension.

Next, open Powershell (you can hit start and just type ‘powershell’ and hit enter). This is a program you might never have touched, but it comes with Windows and has a lot of powerful features — it’s how we’ll control the webserver. But first, try a few commands, see what ls, cd, and pwd do. They stand for List, Change Directory, and Print Working Directory.

Once you’ve navigated to the folder where you saved your first js file, you can type node followed by the filename. This uses NodeJS to execute firsttry.js, which simply prints “Hello World” to the console — another word for the command line. We’ll use console.log() a lot, it can give status messages from the server running in the command line, and it can also give status messages in the web browser, which has its own console we’ll see later.

So that’s a pretty boring program, but it’s important to make sure that ‘hello world’ works if we expect to run a web server on our machine.

Of course, there are whole books written on and whole careers based off writing and running web servers. For our purposes, I want to provide this working model: a web server is a program that listens for requests, and fulfills certain requests however you tell it to. The ‘listening for requests’ is mostly handled for us, we just have to write code to respond to them.

The NodeJS documentation provides dozens of functions for this purpose, but doesn’t give a simple example. I’ll use a stripped-down example from this awesome instructable to get us started. You could copy paste this example code into your editor, but I recommend typing it in yourself to spend some quality time with each component.

If you haven’t coded much, this is a little crazy, but thankfully we don’t have to come up with this ourselves. This chunk of code actually handles quite a lot: it uses require() to import or include the ‘http’ and ‘fs’ (short for filesystem) features of nodejs. It creates a server object and starts listening on port 80. It answers any request by sending index.html back, and delivers a nice error message if it has any trouble.

To get your first website up and running, save this javascript file, and create another file that contains nothing but the phrase “hello world.” Or, if you want to work ahead:

<h1> Hello World! </h1>

console.log("Hello Console!")

Save that file as index.html in the same directory as simpleserver.js and go back to Powershell. Remember how to use node to execute a file? Try it on the javascript file containing the web server code.

The very first time you run javascript code that creates a server, you’ll get the following message (unless you have Windows Firewall disabled):

Make sure “Private networks” is checkmarked before clicking Allow!

I checked Private and Public, since I want to try serving webpages on networks besides my own. Windows will only ask you once! If you don’t allow both networks now, it’s a bit of a pain to change the settings later.

Once you allow NodeJS to communicate, you can open a web browser and try it: just type in ‘localhost’ in the address bar.

If you added that console.log(“Hello Console”), you get to check out the web browser’s console. That line of code executes inside Chrome and writes its message to Chrome’s console, found in the Inspect element menu. We’ll use this a lot later on, so now’s a good time to find it.

That’s all for part one. Next up is writing our own javascript (don’t worry, it won’t be as complicated as the server function, at least at first 😉


And NodeJS Documentation for FileSystem and HTTP.

Preparing FFMPEG for free screen capture, time lapse, and GIF making.


Put together some info from here and here and here.

Add deb-multimedia non-free to your repo and install ffmpeg with the following commands in your terminal:

echo deb jessie main non-free >>/etc/apt/sources.list
apt-get update
apt-get install ffmpeg

That’s all! You’re ready to make screen captures and gifs.

Windows 7/8/10

Download and Install

First, install WinRAR or 7zip so you can open 7z compressed files.

Download the latest ffmpeg.

Adding ffmpeg to your ‘PATH’

Add ffmpeg to your path so it can be executed anywhere. There will be only cosmetic differences between Windows 7 and Windows 10.

I found that one one version of Windows I was unable to edit the path if I simply hit Start and searched for “system environment variables” — instead I had to open the dialog via System (control panel) and then click “Advanced System Settings” and then “Environment Variables”. I guess by clicking on the ‘Advanced system settings’ with the little security icon it gives me the admin privilege I need to change the settings.

After clicking “Environment Variables…” You’ll get a list of user variables and a list of system variables. Scroll through the System variables until you find “Path” and click Edit, you’ll get the “Edit System Variable” dialog. Here each folder containing executables is listed in the format C:foobar;C:foobar; — that is, full file path, ending in a backslash, seperated by semicolons. So we’ll add our ffmpeg folder extracted from the 7z file, specifically, its bin folder. I dropped the whole thing in to my C drive, renamed the long folder name to just ‘ffmpeg’ and added the following to my ‘Variable Value’ (highlighted in the screenshot). The backslash at the end is important.


After OK OK OKing out of there, you should be able to start powershell and start ffmpeg from anywhere. if you get the following when typing ffmpeg into powershell, then everything worked!

This stackoverflow discussion was helpful in figuring out how to do screen capture on Windows. I ran the following command to grab my Powershell window at 10 frames per second and saving to the file Here is the official documentation which details a few options.

> ffmpeg -f gdigrab -framerate 10 -i title=”Windows Powershell”

I wrote a more thorough guide to screen capture here:

FFMPEG is handy at converting video to gif, too!

> ffmpeg -t 2 -i firsttry.gif

Pretty neat!

I wrote a guide to high quality gif making here:

References linked:

Blending and Bending Prosthetic Leg Fairings

From Kinect scan to laser cutting

Opened the original scan in Blender, selected the portion of the scan we’re concerned with. Separated it from the rest of the mesh and exported as stl.

Used Netfabb to repair the mesh and cut into a top and bottom portion.

Found out netfabb decimates the mesh upon export, from 27000 triangles to 24000 triangles, and manages to make it pretty rough, so it looks like this in 123D Make.

Inspected the files before and after repairing in Netfabb…

Used MeshLab’s Filters → Remeshing → Surface Reconstruction to “Shrinkwrap” the model and fill in gaps and holes without losing the smoothness of the scan.

Used this guide to split the part in two using Blender’s newish “Bisect” function.

Played around with different options in 123D Make…

Drew a circle in Blender against the scanned cLeg and stool, confirmed that the units of the scan are in meters. (The circle is .355 units wide and in real life the stools are too.) But, since many programs interpret these STL units as mm, our objects are 1000 times too small. Let’s scale them up.

Settled on a slicing pattern in 123D Make. Used the Modify Form feature to hollow out the leg and picked out the Radial Slices construction technique. The arrangement has some collisions at the bottom, but I don’t want a closed model anyway, so will do some hand-tuning in Inkscape. Exported the model to Blender to check that the cLeg would sit inside it okay.

Using Inkscape to layout the the pieces onto less than 4 sheets of plywood… (nesting is apparantly a difficult problem for computers, and not one that many softwares implement, certainly not free ones. just have to eyeball it)

123D Make is sadistic and exports the objects as groups of thousands of lines instead of a single vector, so editing the exported PDF is really really slow. (It takes 20 seconds to select a text object and type in a new number. I counted.) Learned things:

On the left: the highlighted list of numbers 4,000 characters long describes a single path. On the right side, two coordinates of a single line segment are surrounded by markup. The file containing 2 objects takes 73 lines of text. The file exported by 123D Make takes up 6151 lines of text for the same object. Inkscape modifies this information in memory every time you move something.

Inkscape operates with SVG, which is plain text, ASCII, one character equals one byte. When a path is exported as 677 individual paths, each one of those line segments has 8 lines of markup surrounding it in the SVG file (between the g tags, with information for position and stroke and fill attributes). So what could be saved (and manipulated in memory) as 20 bytes if it was appended to the list of nodes in a single path instead takes 389 bytes to be kept in memory as a standalone object.

I found this forum post describing how to select all the separate objects, Path → Combine them into a single object, and in node edit mode you can select all nodes and click Join Selected Nodes and it simply merges overlapping nodes. Hallelujah!

So to downsize the files enough that they’re not a pain to work with:

Open the PDF. I learned that if you check the “import via Poppler” option Inkscape can detect the colors of the stroke (otherwise imported PDFs have “undefined” color unless you adjust it yourself). So check that box. CtrlA, CtrlU to ungroup everything. Click a blue line segment. Edit →Select Same →Stroke Color. Path →Combine. F2 (Edit Paths by Node). CtrlA. Wait for a minute. Join Selected Nodes (one of the buttons in the node edit toolbar, 3rd from the left, just after the delete node button. It might take a few minutes for Inkscape to chew on this one. It will probably become unresponsive. Let it do its thing. In my case this trimmed the file size by 97% and I could then translate and rotate and rearrange the objects without any lag.

Two hours later…have my minified cut file, fitting on 1.5 sheets instead of 4. I’ve trimmed the tops and bottoms off the vertical slices so the end result will be open.

20 minutes of lasering + 10 minutes of assembling: Hey I made a thing!

Met up with Shawna and learned the design constraints: how the leg has to move, what areas can’t be covered, etc. We picked out a pattern she wouldn’t mind having wrapped around her leg:

I took a few measurements of the wooden positives and drew the outer shape of the fairing in inkscape by hand (using auto-smooth of course!). A few booleans later I had my cut file, which took about 20 minutes apiece. To get the part ready to shape against the mold, I stuck it in our convection oven to bake at 300F for about 5 minutes (don’t worry, we keep one oven for circuit boards and plastics and a different oven for foods 🙂

So the process was: semi-soften the plastic in the oven, then roughly fold it over the wooden form, then stick the whole wooden form in the oven. If you let it soften all the way when it’s sitting on the metal it’ll stick to the metal. Better to have it roughly on the wood form when it reaches maximum droopiness. When it’s as soft as rubber, we take it out and push the plastic against the form, holding it in place as it cools.

I’m pretty happy with the second prototype. I think the next step is 3D printing some bracket system, and maybe taking more time to plan out how the plastic will fold against the form — this was pretty heavily eyeballed.

Continued with a rough prototype of a bracket here: