Click to email me

Objects de-Art

So far I’ve not had cause to print a single art object but an opportunity has finally arisen in my metal casting experiments.  In essence I am attempting lost PLA casting - essentially the same process as lost wax investment casting.  However I want to simplify the process by placing the PLA model directly in the sand and use the molten metal to burn out the PLA.  If this doesn’t work then I’ll be coating the PLA model with a home-made ceramic investment coating and burning the PLA out in the furnace.

For my initial model I thought I’d try a polar bear.  There are 3D .STL polar bear models (some free and some for sale) on the inter-web, but they were either too expensive, too tacky, or highly stylised.  Maybe I can make a better model from scratch?

To date I have done all of my 3D model design using Autodesk Inventor, however my current version of this software does not readily support free-form 3D objects.  More recent versions of the software have extended 3D tools but Autodesk have changed their software upgrade policy to a subscription only service and I cannot justify the the ongoing cost of switching to this licensing regime, particularly given my initial investment for a perpetual license.

I downloaded Blender (free-ware) and spent a three days getting used to the drawing environment and the command line interface.  In my youth when I was interested in drawing I learned the  technique of sketching animals (including people) using ellipses to represent proportioned body structures and muscle groups, and joining these with free-form curves to form natural body outlines.  .  Blender allows you to do the same thing in 3D using elliptical metaballs (mould-able 3D structures that blend together when in close proximity).  While there are other techniques for building free-form 3D shapes that use extendible regular 3D geometry (cubes, spheres, cylinder and the like) they require a lot of tweaking to form natural curves, and making closed manifolds takes some care and effort.

The process started with a simple 2D pen and paper sketch of a polar bear.



Figure 1.  Simple Polar Bear Sketch


Then I found a couple of polar bear images (side and front view) on the inter-web that, with scaling and mirroring, provided consistency between views and were similar to my initial sketch.  I loaded the images onto reference planes in Blender and set about drawing my 3D polar bear using elliptical metaballs.  The whole process only took a couple of hours (remarkable given that I am a complete Blender newbie-dweeb).

I sculpted the final image and added a particle field to represent fur.  The Blender model looked great but I could not export the particle field in a .STL format so I abandoned the fur.  I figure that the actual surface finish of the casting will be a reasonable approximation to fur, and if necessary I can improve this further with wire brushing and hand finishing.



Figure 2.  Polar Bear in Blender


With the basic model drawn, converted to .STL and scaled to about 70 mm (just under 3”) long I 3D printed it. 

Printer settings in Simplify 3D were high resolution (0.1 mm layer height) with 8 top and bottom layers, 2 external layers, a raft and support, and no infill.  The external wall thickness is about 0.8 mm (32 thou.).

I essence I am trying to make a ridged accurate 3D print with as little plastic content as possible, and with a continuous internal cavity.  The intention is that this will aid the PLA burn-out process by using the least possible PLA  and providing an escape path for combustion gases and carbon residues.  I think you’ll agree that the final image is a reasonable likeness and is well proportioned, albeit I have used black PLA (and this is one reason why I don’t want oil drilling in the Arctic).  The print uses about 8 g of PLA.



Figure 3.  Polar Bear 3D Print
(about twice actual size)


There are some print layer lines clearly visible but these are just 0.1 mm (4 thou.) high.  The surface feels quite smooth to the touch, and in any case case these lines are much finer than the sand grain size in my moulding sand.

I’ll be adding a link to the lost PLA casting process once I have attempted it and written up the experiment.  Here is the link,  Lost PLA Casting, but I have more work to do to get the process to work.

Click here to navigate to laboratory equipment...

[Turtles are Here] [Site Map] [750Z Housing] [Construction] [A20 Housing] [Drawings] [Galapagos] [Data Logger] [Making PCBs] [PCB Drill] [CD Welder] [Current Monitor] [Proportional Controller] [Lock In Amplifier] [NAD C 520] [Kenwood Tape Deck] [Tape to Computer] [Reflow Oven] [Designed to Fail] [M2 3D Printer] [Assembly] [Prints] [Objects de-Art] [Lab Gear] [Surface Finish] [Blocked Nozzle] [M2 Upgrades] [Flexible Prints] [Print Problems] [Extruder Failure] [X, Y & Z Calibration] [Coil Winders] [Changing a Light Bulb] [Squash Racquet Repair] [Dive Photos] [Sand Casting] [Oxy-MAPP] [Lathe Maintenance] [Thermal Adhesive] [M73 Compass Repair] [Calcium Block for Turtles] [Infinite R Network] [ST7820 LCD] [Links] [Contact Me]