I have been thinking about where to begin with the story of building a 3D printer. As I write this I am sitting in an engineering works in Holland, and this morning I have been working at the most fundamental part of the process, cutting up 1cm square brass rod into 22mm lengths. I just cut 120, and it’s a messy job; brass dust flies everywhere, you wear all-enclosing plastic goggles to stop the brass getting in your eyes, and that makes your face sweat.
Of course, unless you’re a died-in-the wool engineer, you won’t be doing this, because you’ll buy a kit, and this metal-working bit will already have been done. Usually in China (we ran out and couldn’t wait for the Chinese order to arrive).
But this brass block is fundamental to the business end of a 3D printer, so it seems like the place to start. We print with thermosplastics; plastic that melts at a high enough temperature, and then cools back to solid. It is the job of our printer to make it into an interesting and useful shape between it melting and hardening again.
If you’ve never come across thermoplastic, then pop down to your local DIY store and buy something like this hot glue gun. They are very useful, illustrate the basics of working with thermoplastics, and, sooner or later, despite warnings in this book, and in the instructions that come with the gun, you’ll try and smooth the hot glue with your fingers, and realise just how hot it is. And you won’t do that again (probably; I must have done it at least three times).
What you see is that you have a stick of plastic, and by pulling a trigger you force it through a brass nozzle that is very hot. The nozzle end, in 3D printing terms is called “the hot end”, and reams have been written about it … on the RepRap wiki there are 25 pages written about hot ends, and that’s just RepRap. I’m not an expert on them all, but I have to say that on paper at least my favourite has to be the Budaschnozzle.
So, the hot end consists of a brass nozzle, with a very fine hole (ours is 0.35mm) and the nozzle screws into the brass block. Also in the brass block is a wire-wound resistor that functions as a heating element, and a thermistor (a very tiny sensing device) that reports back to the central electronics how hot the hot end is. The plastic filament arrives at the hot end through a brass tube, that is also screwed into the brass block, and the filament is forced through, melting, and emerging in a soft stream, 0.35mm across.
And that’s it. If you are building from a kit, it’s likely that the whole thing will come pre-assembled (I can assert this with some certainty, since I have spent the best part of today assembling them), so you don’t need to worry, although it’s important to understand what’s happening. If you are really building from scratch, then I suggest that you spend a good amount of time reading through those 25 pages on the RepRap wiki. If you really want to go back to basics, you can eschew the sissy route of buying a wire-wound resistor and buy yourself instead a meter of nickel-chrome wire (it’s what they use to make the elements in old-fashioned toasters), and wind your own. It’s what they did in the leather-helmet and white-scarf era of 3D printers, and I have seen some of those very home-made hot ends. Supporters say that they heat up much more quickly, but I’ve also watched people try to find what’s wrong with their Ni-Cr wound hot end (usually it’s that the insulation has cracked). Search for pictures of hot ends made with NiCr on the wiki, or on Flikr. Your response will tell us immediately what sort of person you are, depending on whether you whisper “cool” under your breath, and immediately order a couple of metres from eBay, or recoil and decide to go for the clean, tidy route of using a wirewound resistor.
I’ll leave you to guess into which camp I fall!
So let’s look at some of the issues of the hot end. At the very (hot) end the job is to melt the plastic so that it flows. There are different grades of PLA, and they melt at different temperatures, varying between 160°C to 230°C, hence the need to have a sensor so that we can control the temperature of our hot end. If we are too cool then the PLA won’t really melt and we’ll have a blockage. (As I write this an e-mail comes in about a guy testing printing with polycarbonate, which is very strong … and he finds he need a hot-end temperature of 300°C. If you want to know more, search for “polycarbonate printing” on You Tube).
However, there is another issue. Think for a moment of our glue gun. We have a trigger that pushes the plastic glue stick into the gun. For that to work, the glue stick has to be stiff, otherwise it’s like trying to … well, all the analogies that I can think of are inappropriate for a technical book, so I’ll leave that to your imagination! Let’s just think about trying to push over-boiled spaghetti up a straw.
This problem is even worse in a 3D printer. Glue sticks are typically 11mm (half inch) in diameter, although small ones are 7mm (1/3″). So, roughly the size of a pencil: plenty to push. Thick PLA filament is 3mm; most of us print with 1.75mm filament (7/100″). That’s not a lot to push with! We’ll come later on to the details of the extruder and how it works, but basically the filament is squeezed between spring-loaded rollers. I have a variety of extruder mechanisms in front of me as I write, and the distance between the extruder rollers and the heater block is at most 5cm (2″). So, at the heater rollers the filament better be cooler than the “glass transition point” (GTP: the temperature at which is stops being rigid and starts being rubbery … about 60°C for PLA) or it won’t push down into the hot end. In fact, the longer we can keep the filament cool and rigid, under the GTP, the better for not gumming up the inside of the extruder hot end. On the other hand, when it gets down to the business end, then it needs to heat up quickly.
Hence the history of the 3D printer is, to a very large extent, one of a lot of research into how you keep the filament cool and rigid (and so capable of being pushed) right up to the heater block, and then coming up to temperature, and melting really fast.
I know that whatever I say here will create a storm of protest in one camp or another, but, in my experience, NiChrome wire is difficult to work with and prone to problems, whereas the brass heater block with a wire-wound resistor glued inside works well (and looks a lot neater!) Again, it all depends on how important it is to you to either go back to basics, and say that you’ve built the whole thing yourself, or how important it is to get it up and running, and printing what you want. As with all things in life, it’s a balance, and we’re aiming for some middle ground here!