One layer at a time
3D printing is not particularly new. It’s not quite commonplace yet, but in the last couple of years it has become sufficiently cheap that even a tightwad like me can consider it.
I could buy a printer or a kit, but that would be too easy. Instead, I’m working on my own design for a printer similar to the marvellous Rostock printer. While most 3D printers are based on a cartesian framework (moving the print head along X,Y and Z axes), Rostock is based on the delta robot design. In this scheme, the print head is suspended from three arms, as if it’s at the point of an upside-down pyramid. By driving each of the arms up and down independently, the print head can be moved very quickly over quite a large volume. Delta robots can move very quickly, because generally the masses of the moving parts can be kept low.
In this spirit of 3D printing, I’m going to print as many parts as possible for my printer. Which is to say that the excellent Rob is going to use his Ultimaker printer to do it for me. To design the parts, I’m using OpenSCAD. it’s not the most user-friendly CAD program on the market, but it has the twin advantages of being both free and highly programmable. This means I can define shapes parametrically, and then alter their sizes with ease. For example, I can change the size of bearings I am using, and make all the parts using them automatically change size to fit. Lovely. It takes a bit of work to define the parts in the first place, but as the saying goes: if a job’s worth doing then it’s best to pay someone else to do it. I won’t be doing that, because I’m cheap.
As an experiment and sanity check, we printed a sample of the first part (the print head). Designing things in CAD is great, but there is no substitute for holding a solid object in your hand to assess it. Here’s a picture of what the object should look like, according to OpenSCAD:
And here’s what it actually does look like, after printing:
I can’t help thinking that there’s something vaguely magical about this.
Hello.
Only to inform you and with my thanks for your explanations well documented with your drawings.
It seems to me that there is an error on the page ” http: // robinsonia.com / wp/? Cat=4 “.
You write ” hb = sqrt ((la*la) – (ac*ac)) ”
instead of
” hc = sqrt ((la*la) – (ac*ac)) ”
Beautiful work!
Again, thank you.
Roland,
Thanks for spotting that. You are absolutely correct, and I have updated the page.
Peter