Justin Dunham

's journal about making things

Making Fondant

Filed under: Cooking Journal — Tags: , , , , , , — Justin Dunham on March 27, 2011

The completed cupcakes. Each has an orange fondant hexagon on it.

Rolling out the fondant...

Filling the cupcakes with lemon curd. You can see I've cut a cone-shaped piece out of the top of each cupcake, which the curd goes into. The top is replaced, and the cupcake is iced.

Confectioners' sugar and cake flour. Both boxes have the same freaking cake on them! In fact, I think the slice on the Swan's Down box may have been taken from the cake on the Domino's box! This, for some reason, is why I confused them.

You may have a hard time finding this. But you definitely need it. Check the skin care aisle.

A couple months ago, I got accepted to a group called Hive 76, which is kind of the local nexus for people who are interested in DIY 3D printing and lots of other cool things as well. To celebrate actually being accepted, I thought I’d bring some cupcakes to their Open House.

The cupcake batter was basic red velvet batter, which has a sort of savoryish taste that I really like. It’s easy to fill cupcakes – you just cut a cone out of the top, add your filling, then put the top of the cupcake back on. So I figured I’d put some lemon curd in as well, since it’s really cheap at Trader Joe’s, tastes great, and helps counteract the dryness that sometimes happens when I make baked goods in my oven.

(I learned a very helpful piece of information about baking in relation to this, by the way – when you are doing the toothpick test, you only need to push the toothpick about halfway through whatever you’re baking. Only halfway! This makes a lot of sense, since what you’re checking is that the inside of the cake is done cooking. Pushing the toothpick through the cake should theoretically work, I guess, since the outer layers of the cake shouldn’t leave any residue on a toothpick if the inner layers don’t. However, it doesn’t seem to work this way in practice, so my cakes always ended up overdone. Good to know.)

Anyway, on top of each cupcake I decided to put a hexagon made of orange fondant, since that is the Hive76 logo – thought this would be an interesting excuse to actually make some fondant. I have to say, a lot of my cooking is actually motivated by curiosity as to how something is made, or a desire to make something, rather than the eating part. Though that is good too.

Making fondant is very similar to making dough, except that every time you’d want to put in a normal ingredient, you put in something sugary instead. Instead of flour, use powdered sugar. Instead of eggs, use glycerin.  One huge mistake I made on the first run was using cake flour instead of powdered sugar – the boxes are virtually identical! I kind of wish I had finished with that recipe, though… would have been interesting to see what happened.

Overall, it’s not that difficult to do, and I was able to finish the fondant without too much trouble (though it does require a hell of a lot of food coloring). One important piece of information is that if you are ever looking for glycerin – and you will be looking for it if you make fondant – you have to go to a pharmacy to get it. Once you’re at the pharmacy, check the skin care aisle.

Why skin care? Glycerin is used primarily as a remedy for chapped lips. This means its safe to eat, and I guess its moistening properties are part of why it’s an essential component of fondant.

Building a Desktop Fabricator (3D Printer), Update

Filed under: Everything Else — Tags: , — Justin Dunham on

It’s been a busy couple of months on this project, and I have lots more to say, but here’s a quick summary of what ended up happening with the bot:

Where I last left off, I had physically assembled the Makerbot as well as the extruder. I ended up trying it the next day…

And it just didn’t work. I tried lots of different experiments – different software on my computer and the various hardware boards on the bot, disassembling things and reassembling them, testing circuits, etc. etc. Nothing. The platform moved, but I couldn’t communicate with the machine via my computer.

One thing that quickly became apparent was that the extruder board (which controls the part that actually melts the plastic) was defective or otherwise busted. This was sort of a shame, but I was eventually able to get a replacement board from the sellers, Makerbot (more on that later).

I ended up shipping it back to myself in Philadelphia to keep working on it. Right away, I replaced my defective extruder board with the new one. Should work, right? Well… sort of. The extruder wouldn’t extrude for more than a minute or so, if that. When it did, it was globby and slow. The build platform would also skip steps, meaning that it would not move as far as it should one at least one of the axes as it was building.

(What does this mean? Since the motor has no idea whether it was successful in moving the platform, this meant that you’d get something built a little bit, and then the next step up would be shifted a little bit – imagine if you were constructing a building, and the second floor started ten feet over to the right of the first, you’d have an overhang, the building wouldn’t look right, or be stable, etc.)

Fortunately, I’m a member of Hive76, which is a local Philly community of makers and 3D printer people. Several people there spent literally hours with me troubleshooting these problems. Here’s what we found:

  • I made a mistake in putting the extruder together – the toothed wheel that grips the filament as it comes into the extruder wasn’t put on tightly enough, so it was slipping. The rod that Makerbot ships (shipped) with each machine to help you get this distance right is actually not helpful for fixing this problem! It just needs to be quite close, so it gets a strong grip on the filament.
  • Applying lubricant to all of the axes – particularly the z-axis – basically fixed the skipping problems.
  • As for simply getting the machine connected to my laptop so I could control it – the key was to keep trying different versions of the extruder firmware and motherboard firmware. I wonder if the fluorescent lights in my fiancee’s Dad’s workshop also may have caused some interference with the electronics.

And that was pretty much it… Oh, except for the very, very long process of calibrating the machine. There are dozens and dozens of settings with completely inscrutable names (“feed rate” controls the movement of the build platform, the setting for the ratio of filament length to width is on the “Carve” menu, etc.). This took me another few weeks of detailed work to get right.

So I got the bot working pretty well, as you can see in the video. The bot is printing a small plastic part that can be used to build another, different, 3D printer. The printer isn’t at all perfect, but it’s also not bad for this level of technology and for my level of skill. I’ve also been thinking a lot about whether it’s possible to build a more reliable kit, and how this could be done, and working on that. I’ll probably have some more posts about that later.

Creative Commons License
.