Justin Dunham

's journal about making things

The original bartender’s guide, some of which is in verse

Filed under: Cooking Journal — Tags: , , , , , , , , , , — Justin Dunham on September 25, 2011

A bartender making a "blue blazer". New York City, 1862, or Portland, 2011?

Here’s a fun book I found – “How to Mix Drinks, or, The Bon-Vivant’s Companion”, published in 1862 (and it’s free!). This is the original bartender’s manual. It has lots of arcane and delicious-sounding recipes, charming pictures, comments (“refrigerate with all the icy power of the Arctic”) and contemporary advertisements at the front and back.

You should take a quick look at it yourself, but there are a few specific things I wanted to mention.

First of all, the method for extracting citrus flavors. Instead of zesting a lemon, the guide instructs that “the ambrosial essence of the lemon must be extracted by rubbing lumps of sugar on the rind, which breaks the delicate little vessels that contain the essence, and at the same time absorbs it” (p.18 of the PDF). This seems like a really tedious and possibly inefficient method to me, but I wonder if it’s still used today in some cases, or if it’s been superseded by the availability of the zester?

Secondly, check out the first few lines of their recipe for mulled wine, which is in verse on page 61 (the whole thing is in the gallery below this entry):

First, my dear madam, you must take
Nine eggs, which carefully you’ll break
Into a bowl you’ll drop the white
The yolks into another by it.
Let Betsy beat the whites with switch
Till they appear quite frothed and rich –
Another hand the yolks must beat
With sugar, which will make them sweet…

The eggs are eventually mixed with spices, wine and nutmeg. I really like that this recipe is in verse, and it seems to have been taken from The Ladies’ New Book of Cookery by Sarah Buell Hale, though that’s just the earliest occurrence of this poem, which shows up in a bunch of cookery books from the 19th century. I wonder why only this one is in verse, though? Is it because it’s a popular holiday drink that many people knew or wanted to know?

Thirdly, the second half of the book is in some ways even more interesting than the first. It’s a DIY manual for manufacturing your own liqueurs, as well as many other important bartending ingredients such as syrups. This would have come in very handy in the 19th century, I’m sure, before the advent of easily-accessible liquor shops all over the country. Here’s their recipe for gin:

How to make gin. Where do I get juniper oil?

I’ll have to try these out one day, though that is a lot of gin. Here are a few more selected recipes from the book (you can click the thumbnails to read them):

Mulled wine in verse.

Ginger wine - one of the many "temperance drinks" listed in the manual

"Arf and Arf"

The many-layered "pousse l'amour"

I like the cooling method of having a giant block of ice in the punchbowl (though it dilutes the drink after a while).

A drink made from eggs and ale.

"Then smile."

What I learned at my first Open Hardware Summit

Filed under: Everything Else — Tags: , , , , , , , , — Justin Dunham on September 17, 2011

Swag bag! Including a copy of Make:Magazine, throwie parts, iPhone screwdrivers, lots of other awesome stuff!

Open Source Hardware means technology for which the plans, schematics, and other diagrams – everything you need to know to make your own copy of the hardware – is freely available. The most prominent example to me is the RepRap 3D printer project, and the Makerbot, which is based on it. But there’s also the Arduino microcontroller, OpenPCR, which lets you replicate DNA at home, Lasersaur, which is a laser cutter you can build yourself,  and lots and lots of other projects.

This year was the second gathering of prominent people in the open hardware space, including CEOs of companies that make open hardware, legal experts, academics, and lots of other people. I think open hardware is going to be pretty big in the coming years, especially as it continues to coalesce and overlap with the Maker movement, and so I thought it’d be useful to check out the conference.

(Also, before I continue, a note about the swag bag from the conference. It was fantastic! Electroluminescent thread for sewing, a full set of LEDs and a battery for making throwies, an issue of Make Magazine (beautifully bound and normally quite expensive), as well as tons of other cool stuff.)

Big Picture

The  first set of talks (full schedule is here) was on the open source hardware landscape in general – the big projects, where the community came from and where it’s going, etc.. The guys behind Arduino, the open source microcontroller, gave the keynote.

Now, you have to understand, a year ago I literally knew nothing about hardware and electronics; all my self-teaching efforts over the past few years have been devoted to web development. Since I joined Hive76 in January, I’ve learned a lot, and earlier this week I bought my first Arduino, and finally felt as if I had “arrived” in beginning to understand how to make things.

So it was great to see these guys up on the stage talking about the evolution of their product, which, in addition to being an open source hardware product itself, is also a critical enabler for lots (most?) of other projects that are going on, and a great model for how an open source hardware company can build products that help educate people.

Arduino is about to sell their 300,000th chip, which is pretty impressive for a project that started with a 700 Euro investment about six or so years ago. Especially given that the product is open source – their source of income is a 10% royalty on the wholesale price from anyone wishing to distribute Arduinos.

It also was the first instance of a theme that resonated for me throughout the gathering: entrepreneurship doesn’t have to be dramatic. Coming from Wharton, and reading well-known startup blogs, I think it’s easy to get the idea that all companies follow the young person-tech product-VC money-fast growth path.

While that model is important, it discounts almost all of the 600,000(!) new companies that get started every year, from which people make their living. It might be kind of the equivalent of living your life based on People magazine, no? It’s a very narrow view of an incredibly broad field. And even Bre Pettis of Makerbot, whose company recently got $10M of VC funding from some very well-known people, looked pretty relaxed when I saw him at the conference. He is either an absolute master of cool under pressure, or simply doing what he enjoys doing and not freaking out about it.

Anyway, the Arduino guys shared a lot of the lessons from starting their business, and I’m told their slides will eventually be posted on the Open Hardware Summit site. A few of the things they mentioned are that it’s important to know who your customer is, know what you want out of it, and expect resistance. Sort of standard stuff but good to hear again. They were asked about VC funding, and they said something else that echoed throughout the rest of the conference: many open hardware companies don’t need VCs (and they like it that way), or VCs couldn’t be made to understand their business model. Sometimes both.

Eric Wilhelm, the founder of Instructables, spoke shortly after the Arduino guys, and talked about how Instructables grew. Amazingly, when the site was first launched, quite a lot of its content was guns made out of K’Nex (which is a building block system sort of like Legos).

Eric showed a sequence of incredible videos of early Instructables members showcasing increasingly formidable K’Nex weapons. While nobody at the conference thought that could be a business, the level of innovation and evolution was incredibly high, and of course the apex was video of this backpack-mountable K’Nex gatling gun, which you should check out sometime.

The implication? Every open source hardware project has a (metaphorical) backpack-mountable K’Nex gatling gun that it is moving toward, because communities can build around interesting projects and evolution and innovation can happen rapidly.

Bunnie Huang, the founder of Chumby, spoke about Moore’s law in relation to the open source hardware community. I can’t claim to have fully understood everything he was saying, but one interesting idea he had was that of the “heirloom laptop” or “heirloom phone” – technology that you pass down because the version that comes out in 18 months isn’t simply ten times faster and therefore better. What happens to technological artifacts when there is no more real technology obsolescence?

Social Change and Communities

Later in the afternoon we heard some talks on how open hardware was being used for important social goals, and on how communities spring up around open hardware.

The “social change” presentations opened my eyes to the incredible range of open hardware projects that are out there right now; speakers talked about DIY sailboat drones that clean up oil spills, open source geiger counters, and an eye-tracking-based visual writing system that’s allowed a paralyzed graffiti artist to produce salable graffiti art using only his eyes.

These talks show one of the massive advantages of open source hardware, which is that it’s designed for dissemination. In the case of sailboat drones, for example, any environmental organization or interested amateur could build these drones (and even improve them and release the improved design) in the wake of an environmental disaster or in preparation for one. Given how ineffective current methods for oil spill cleaning are, this could be a tremendously useful way of allowing individuals to contribute meaningfully to environmental cleanup. There’s no way to do this with closed-source models.

The geiger counter example is another great one – Shigeru Kobayashi talked about building open source geiger counters in Japan and using that data to map radiation levels across the country in the wake of the nuclear accidents there.

Suddenly, it’s possible to liberate massive amounts of human effort and ingenuity in the service of an important cause. And you don’t have at least one major pitfall of open source software, which is widespread availability of competing closed-source products. Who else is building a really cheap geiger counter? It looks like only open source hardware is filling this niche right now.

Certainly this is the case with one of the other projects that was discussed during this segment, the open source book scanner. What company is producing, or would want to produce, a cheap book scanner for public use? In this case, the open design of the machine (it can be built quickly and cheaply) also directly supports one of the machine’s goals, which is getting as much information as possible digitized and transmissible, as quickly as possible.

Starting Up in Open Hardware

The last major topic area the conference covered was starting up an open hardware business. I like how this seems to be a natural milestone for many open hardware projects, since you need a financial and operational infrastructure to really popularize any piece of open hardware – compare open software, which can be distributed and widely disseminated almost for free.

This set of talks covered mostly practical issues related to getting a hardware business up and running. Bryan Newbold of Octopart talked about some of the immense savings that can be had from mass-ordering certain parts – but also the possible downfalls of trying to immediately commission large runs from places such as Kickstarter, which often take pretty large fees.

Nathan Seidle at Sparkfun (where I bought my new Arduino and other parts from) talked about what parts of his business are open and which aren’t. I have been thinking about this issue myself and I’m sure some open source hardware people are too – his answer was basically that they open up whatever helps their customers build better. Spec sheets for parts? Yes. Supplier names? Not really. Private customer data? Certainly not.

Other interesting things

There were a few other talks that I didn’t want to cover in detail, but just mention briefly. Michael Weinberg at Public Knowledge talked about protecting 3D printing (and open hardware) from government intervention.

This is a real, real risk – anyone reading this blog post knows (I hope) that open hardware folks are the good guys. But it is all too easy for lobbyists to get Congress to pass rules squashing the good guys, in the name of protecting the rice bowls of the companies those lobbyists work for. The law is, sadly, a tool for competitive advantage. But Weinberg talked about events he’s organizing to introduce open hardware folks to Congresspeople, so that open hardware has a little more protection when the fight materializes. We’ll see.

Bre Pettis gave a quick talk on the proliferation of designs and “digital mashups” in the 3D printing community, using the many variations on the 3D-printed “gangsta” model (including Yoda + Gangsta = “Yodsta”, and the very meta Gangsta + Gangsta = “sta-sta” as examples).

All I could think of the whole time, as Bre paged through dozens of mashups, was evolution – it was this  massive quantity of different species created through mutation of a common ancestor. Of course, that’s sort of what open hardware is about in some ways, since the DNA (instructions, code, schematics) constantly morphs and evolves, resulting in different phenotypes (completed robots and devices).

Conclusion

What a great way to spend a day, and how impressive, too. I’m excited to think about what next year’s summit will look like! I think the most encouraging thing about the summit might have been that you can both be generous and open with your knowledge, but also build a sustainable business. It’s hard for me to imagine wanting to do one without doing the other.

Designing a 3D printer kit: Packaging

Filed under: Everything Else — Tags: , , , , , , — Justin Dunham on September 14, 2011

This is the third entry in a series of four exploring the design, packaging and user experience of 3D printing kits. Each article will be published to the front page of this blog, but you can also find them, as well as other articles I’ve written about open-source 3D printing, on my 3D printing topic page.

Packaging

Once I collected all the parts, the next step was to start actually packaging everything. The goal of this project was to focus on making a kit that’s accessible to mass market consumers, which I think implies that it’s (a) free of problems that only experts know how to solve, (b) organized, and (c) helpful during the build process – users should encounter parts as they need them, for example, not all at once. Where possible, the kit should also look interesting and inviting, and worth talking about and showing off.

I did a few things to try to move toward these goals.

Staged packaging

The top of box #1, with the first set of numbered envelopes visible.

All 16 envelopes.

The reverse of the envelopes, with their contents.

Closeup of envelope 15.

A few more of the envelopes.

Envelope 15, unpacked.

In my experience with 3D printing kits, parts are typically packaged together by type, instead of by construction step. So, for example, all the nuts of a certain size will be in the same bag. Worse, actually – all the nuts of all sizes are often in the same bag. So are all the washers, all the bolts, etc. This creates a number of problems:

  • You have to dig to find a screw (nut, washer, etc.) of a certain size.
  • You have to manually count out all the nuts you need for a certain step. So, for example, if you need 12 washers, you need to pour out your bag of washers, and then find and count out 12 of the right size.
  • You also have to find somewhere to store these washers while you are using them, which makes it easy to lose important parts, and less smooth to pause and resume assembly (since you may have a large number of different parts lying around your workspace in different places). Sure, it’s easy to find a little bowl or something to hold the parts you’re using, but why?
  • If you don’t have an intuitive sense of which screw is 25mm long vs. 20mm long (for example), it’s easy to accidentally use the wrong part. It may then be impossible to fix your error without a large amount of time-consuming disassembly.
  • It’s slightly intimidating to look at a huge amounts of parts all at once, and know that you will eventually need to fit them together. Like the feeling you get when you unpack a 1,000-piece jigsaw puzzle.

Packaging this way is also a missed opportunity. If, on the other hand, parts are organized by step, the packaging itself can help to lead the user through the build process.
So, I decided to split the assembly up into 16 stages, and assign each stage an envelope that contained the necessary build parts for that stage. There are a few large parts (including 3D-printable parts) that couldn’t be included in envelopes, but these can be found and identified easily as necessary. I chose translucent plastic envelopes, because these can be easily and non-permanently closed, and because the parts can be seen inside before the envelope is opened.

Each envelope contains a card that can be read through the envelope. On the front, this card clearly specifies the step number corresponding to each envelope, and on the back is printed a visual bill of materials listing the parts that should be inside the envelope. Having a set of pictures corresponding to all the necessary parts makes it easier for end-users to identify the parts they’re using, and is also helpful for manufacturing, allowing a packager to quickly sort all the parts into their respective buckets by putting each card in a Tupperware sorting container as a guide. Card and container contents can then be transferred together into an envelope.

Vacuum-molded containers

You end up with a perfectly-molded plastic tray...

Closeup of the packaging.

Often, the 3D-printable parts in a kit like this are packaged unceremoniously together in a Ziploc bag or box. (As mentioned above, I refer to them as “3D-printable” because I used cast resin, not a 3D printer, to make them).

This looks strange, and it also makes it difficult to find a part when you need it – you have to dig through all the parts, every time you need one. This difficulty is compounded by the fact that (a) there are a few quite large 3D-printable parts, which can make it difficult to extract the small ones if they’re all packaged together, and (b) many of the parts are complex shapes that are difficult to identify.

To solve these problems, I decided to use vacuum-formed plastic to package the 3D-printable parts. I talk more about vacuum-forming here, but basically a layer of warm plastic is put over a tray containing the parts, then stretched to fit them as it cooled. The result is a tray with a little molded space for each 3D-printable part. Molded packaging is cheap, but makes it easy to see all the 3D-printable parts at once, so it’s easy to find and extract the correct part when you need it. Related parts can be kept together (such as the x-end-idler and x-end-motor and their matching clamps), and labels could easily be added to the tray to make identification even easier.

Reducing or eliminating soldering, epoxying and crimping

There are several parts that come with kits that require partial assembly before they can be used. This includes cables, which sometimes come as bare wires and connectors and therefore must be cut, stripped, and crimped, and it also includes other parts such as the extruder, which requires that a length of heater wire be wrapped around the extruder nozzle, taped in place, and soldered to connecting wire.

I think companies like Makerbot have largely eliminated these steps from their kits, but in any case I think RepRap kit contributors should be aiming to do so also. It will be hard to gain popular acceptance for a kit that requires learning how to solder (or even use a multimeter) if you don’t know how to do it already.

Tools

Most kits include a set of hex wrenches or other small screwdrivers. I wonder whether, optionally, you could get an electric screwdriver with the correct bits included with your kit? They’re $20 on Amazon, and I’d estimate that a significant portion of the time I spent assembling (perhaps even 10% – 20%) was spent screwing things in manually. Sure, users can go buy an electric screwdriver, but the goal is to make the experience as seamless and pain-free as possible – and including or offering the option of tools like this might serve as a signal to new users that they should buy one.

Jigs are included in the kit I assembled, allowing a builder to quickly determine whether everything is level and whether things like frame vertices are the appropriate distance from one another.

Instructions, and packaging the kit for mailing and storage

While the RepRap community has worked very hard to document all the steps in building one of these 3D printers, and while lots of good documentation is therefore available, what I thought was really needed was IKEA-like documentation that enforced a simple way of communicating each step to the user. Luckily, we are already much of the way there – someone has produced these visual instructions that are very close to what I think the ultimate goal could be. These instructions should probably be printed and bound, to allow several people to easily share them as a build process is happening.

I didn’t give a lot of thought to the larger boxes that the kit would be packaged in (including the trays, envelopes, instructions, etc.); I went with some relatively inexpensive ones that were an attractive blue, and that were the correct size for the plastic trays I had made.

Next steps and customer feedback

While I’m done with this project for now, I designed a survey to use in case I wanted to do further research on user requirements (many of the changes I made were based on informal conversations and my own experiences). I think this survey could be generally useful for product improvement as well, which is why I’m printing it here. Tracking customer feedback post-purchase would also be important for customer satisfaction and to improve future versions. You can click “Show Survey” to see it. Show Survey »

Value-added services

While this isn’t technically “packaging”, part of the user experience could be value-added 3D printer building services. I also think this could (and perhaps will) be an important source of revenue for open source 3D printer companies, and perhaps all open source hardware companies, going forward. Some of these services could include:

  • Build help. When I was building my Makerbot, I got so frustrated at one point that I posted some job listings for someone to help me finish it. Luckily, I became a member of Hive76 shortly thereafter and the guys there helped me out. But what about those who don’t have this option? Or those who just want to get the build done, quickly?
  • Repair and modification. Makerbots and 3D printers in general are designed to be modified and improved. Similar to the build help service, what if you could bring your bot to someone to have better parts installed?

That’s it for the composition of the kit. In the next and final entry, I’ll talk briefly about some of my conclusions from doing this project.

Rose, orange blossom, and elderflower marshmallows

Filed under: Cooking Journal — Tags: , , , , , , , , , , — Justin Dunham on September 13, 2011

Homemade rose, orange blossom and elderflower marshmallows.

Extreme closeup. These marshmallows look slightly different - the holes are a lot bigger and much less uniformly distributed.

Test cards for figuring out what was wrong with my original recipe.

At some point I will write a longer post about the work we did for our wedding, which included a lot of DIY stuff. Today I just want to talk about marshmallows.

I knew I wanted to make a bunch more for two reasons. First of all, I wanted to try out this egg white marshmallow recipe. For some reason, I lost my marshmallow mojo (you can call it “marshmojo”) a few months ago and every batch I’ve made since then has come out tasty, but dense; more like a soft candy than a marshmallow. After running a bunch of tests to see what was going wrong, I realized the most effective approach might be to try a different, more reliable recipe altogether.

Secondly, I also wanted to try making some fancier, more… conceptual… marshmallows than I have in the past. I initially thought about the idea of making them dessert-themed (guinness + chocolate topping, banana + caramel topping, something else) but I eventually decided to make them all taste like flowers. This was easier because I had already tested some elderberry-flavored ‘mallows a couple weeks before, so I knew it would probably work out.

I decided on elderberry, as well as rose and orange blossom. Marshmallows like this are not at all hard to make. You just follow the normal recipe, but at some point, you add a little bit of your chosen flavoring agent. In the case of rosewater and orange blossom water, I think I added less than a teaspoon to each batch; I was able to be more aggressive with the St. Germain.

You should add the flavoring to the cold water in which the gelatin blooms; you don’t want it to boil away! Be sure to taste it (before adding gelatin) to make sure you’re getting enough, but not too much, flavor. Keep in mind that you want it to be fairly strong since you’ll be adding a bunch of sugar and water later.

They turned out great, but actually making them involved three separate disasters which vary in importance depending on your point of view.

  • The markings on the thermometer I was using were poorly designed, leading me to read it as twenty degrees hotter than its actual indication. This ruined my first batch of marshmallows.
  • Rose water and orange blossom water, unlike elderflower liqueur, are really strong! They are completely undrinkable by themselves actually, a fact which I discovered accidentally about halfway through the second batch. I’m really glad I tasted the recipe before finishing it, but this lost me a whole lot of rose water and about a pint of strawberries.
  • I accidentally covered my wife’s Dad’s and stepmom’s kitchen in sugar, and I guess I am so used to a slightly sticky kitchen that I didn’t clean it up that well. Sorry guys. We all had a good laugh about this later. (Well, I did).

It’s good to persevere, though. After wasting a lot of ingredients, I finally ended up with three very large batches of marshmallows that looked and tasted great. I can pretty much guarantee that none of my guests had tasted marshmallows in these flavors before, and they may never again!

I tinted each batch very subtly so they could be told apart by color – orange blossom were orange, rose a light pink, and elderflower purple. We served them along with some wonderful floral cocktails and some normal marshmallows and s’more fixins (egg white marshmallows liquefy if you try to roast them).

Designing a 3D printer kit: Sourcing parts

Filed under: Everything Else — Tags: , , , , , , — Justin Dunham on September 11, 2011

This is the second entry in a series of four exploring the design, packaging and user experience of 3D printing kits. Each article will be published to the front page of this blog, but you can also find them, as well as other articles I’ve written about open-source 3D printing, on my 3D printing topic page.

Sourcing the components

In the last entry, I talked about my general thoughts on producing a user-oriented 3D printing kit. In this entry, I’d like to talk about the logistics of sourcing parts for such a kit (or at least how I did, as an amateur).

To build a RepRap 3D printer, there are 3 major categories of parts that you need:

  • 3D-printed parts (plastic bits like pulleys, frame vertices, motor holders, etc.)
  • Machine parts (rods, screws, bolts, nuts, etc.)
  • A category that I will call “everything else”. This includes the metal extruder nozzle, epoxy or cement, various wires, electronics, motors – basically, as the name implies, everything other than what you can buy from McMaster Carr or print on a RepRap.

Here’s a table to make things clearer:

Category Picture Notes
Printed parts All of these parts can be printed on a 3D printer (or produced in other ways)
Machine parts

Nuts, bolts, rods, etc. Everything metal that can be found easily.
Everything else Motors, optical endstops, wires, aluminum plating, and lots of other random things

3D-printable parts

The parts packaged within the kit.

Alternative coloring.

Instead of printing the 3D-printable parts (which takes a long time and often gives inconsistent results), I used a cast-resin parts kit I developed earlier in the summer. Cast-resin parts, while philosophically somewhat opposed to the RepRap concept, have a number of advantages over printed parts.

First of all, they can be produced quickly, cheaply, and professionally if enough effort is invested and the right molding equipment is used. By contrast, when these parts are produced using a DIY 3D printer, they often have inconsistencies which make them look less polished than you might expect from a professional product.

Secondly, they look cooler than printed parts and can be easily customized. “Making it your own” is an important way to make high-tech products appealing to consumers – color customization and choice is the most obvious example, and by casting these parts, you can easily offer a range of colors and transparency levels, and even other creative presentations. What about embedding little gears or sparkles in the parts, or even Jurassic Park-style insects trapped in faux amber? This is all easy to do with resins.

Machine parts

What I’m calling “machine parts” (rods, screws, nuts, bolts, etc.) are pretty easy to acquire – you just order them from McMaster-Carr. In case you’ve never heard of McMaster-Carr before, they’re a giant hardware store that stocks almost everything you can think of, in many, many combinations.

For example, go check out the listings for screws there. You’ll get literally dozens of option categories – material, metric or SAE, length, thread size, cap type, counter or non-counter, etc. As someone who was new to buying hardware until recently, I was blown away by all the options available, all of which can be sent to you very, very quickly.

To make sure I ordered the right parts, I made a fairly complex Excel spreadsheet with the necessary part listings, quantities and McMaster-Carr part numbers. (Even so, I found I had to return a couple of things and order the correct versions – the Prusa Mendel bill of materials is not always easy to match up with listing on McMaster’s site).

If you’re interested, you can see the relevant parts of the spreadsheet here. Building a bill of materials for a product like this is very hard – since it’s a young open source project, there’s no one, authoritative source that also includes all the information one might want.

In at least a few cases ordering the machine parts is only part of the battle. The threaded and smooth rods for example, and the extruder bolt, have to be modified to be included in the kit. (The threaded rods, in case you are curious, make up the printer’s frame; the smooth rods are the tracks along which the extruder and build platform move, and the extruder bolt is what catches plastic so it can be fed through the extruder).

Getting ready to cut a threaded rod. (See the previous entry for the actual cut). This is a miter saw with a grinding wheel that cuts through metal.

So I had to learn how to prepare all these parts for inclusion in the kits. For the threaded and smooth rods, for example, I had to figure out the best way to chop them up into the necessary lengths – the cheapest way to buy from McMaster is in meter-long lengths, when what you actually need is several odd rods with very precise lengths like 495mm. I started by using a dremel and soon found that this was basically impossible as a way of cutting them – extremely slow, slightly dangerous, and very, very ugly.

What I really needed was a grinding wheel. Fortunately there was a miter saw at Hive76 that could be properly outfitted with a grinding wheel. After it spins up, you bring it down on the metal you want to cut. Amidst a shower of bright orange sparks, you get a nice, clean cut.

Everything else

Where to begin on how to source all these things? Here’s a brief list of the things you need, to give you a sense of the range of objects we’re talking about:

  • Auto cement
  • Thermocouple wire
  • Brass heater block
  • Teflon tape
  • Extruder nozzle
  • PTFE barrel
  • Zip ties
  • Aluminum sheeting
  • Medium-density fiberboard, precut in very specific shapes

Oh, and not to mention electronics and motors, which are the guts of a 3D printer. You can improvise around some of the other parts if they are missing, but not these.

I used a bunch of different strategies to get these parts. Some of the parts are, fortunately, readily available at Home Depot and other hardware stores – zip ties and PTFE tape, for example, fall into this category. Some I managed to scavenge from a 3D printer kit I bought a few months ago; this was particularly important in the case of the electronics, which cost over $100 by themselves and are therefore the most expensive single component. Lastly, I got a few things from sellers on eBay – motors, for example, and some of the lasercut medium-density fiberboard that’s used for the build platform (though it would be good to learn how to lasercut eventually).

In the next entry, I’ll talk about some ideas on how a kit like this can be packaged.

Older Posts »

Creative Commons License
.