I was thinking about chilli flakes this morning, which naturally leads to clay printing. Extruding and printing with clay/cob is difficult enough, but taking variable subsoil, removing stones and organics, identifying clay clods and breaking them up, accepting additive clay or sand, mixing it precisely, consistently, quickly enough to keep up with the printer AND within the limits of a lab space is a hard problem. On large sites, area hungry methods of wet treading, treading with a cow, or treading with a JCB make sense. For this post, we’re discussing large scale demonstration printers that may use 12kg of material a minute, but aren’t operating on a large scale site.
At Cardiff, mixing has been done by hand or using a small cement mixer. It’s cheap and well established tech. Mixing powders dry was found to give a quite even mix, water being added at the end. The cement mixer approach has no ability to cope with stones or clay clods and these need to be separated and hand ground before mixing. Hand mixing can cope with these, and I encourage it for the first batch to help students develop a feel for the material. It does get tiring quickly, isn’t accessible to all, and isn’t very consistent between people.
The main issue in use was transferring the cob from the mixer into the loading tubes for the extruder. Unlike concrete which is designed to flow to fill formwork, 3d printable cob is designed to sit there unless really pressed. There’s an upper limit on the amount of straw that can be mixed in without it balling, but that appears to be lower than the maximum straw in the mix the extruder can handle without jamming, so in practice, for that setup, it is not a problem.
Cambridge use a pug mill to mix their clays. These are a well established approach where clay is added in a hopper, and chopped to pieces and pushed forward by a very strong propeller blade. It gives you a very dense, deaired clay, and can directly load extruder tubes for smaller printers (3mm diameter nozzles and below). The below video demonstrates the process well. The first clay to pass through the machine is discarded as not solid or (if mixing) being inconsistent. The pugmills also need to be cleaned and emptied after use. This is not an issue on a house sized 3d printing site, but is a disadvantage in the workshop, where 12kg of cob discarded at the start and end builds up quickly. Unlike pure clay, cob with straw mixed in can’t be stored wet for months.
The effect of the pugmill’s blade on the straw have not been verified (to my knowledge), and while it may break down clods, again, I’m not sure how effectively mixed the result would be. Emerging Objects appear to use one to feed their printer (visible at 8 seconds on their video, still image below). I was sure I could see a Pan mill in one of their old photos, now taken down. Perhaps I was mistaken. EDIT: Prof Ronald Rael goes through their entire process in detail here: https://youtu.be/tTChT5olDAY?t=889 It’s not a bladed pugmill. It’s a rubberised rotorstator, which is a type of continuous volumetric flow pump, not a mixer.
Pan mills, take even more floor space, and are also fairly cheap and well established (mostly in gold mining). In gold mining use, lots of water are added, the wheels go around slowly, and the finest of particles are squeezed out and flushed out in the water. Note in the image below, one wheel is fixed height, the other is free to bump up and down the follow the material in the bottom. The advantage here is that it can cope with stones in the soil, will break up clods and once the material is mixed, you can fit your extruder loading cylinder under the track and the pan mill will slowly squash and pack material into the tube.
Can they take up less space? In the image below, taken from a French patent, note that the size of the wheels and the circle they turn on is mismatched. This is always true to an extent (it only perfectly lines up for a infinitely thin slice in the image above, for example), but in this case it’s a massive difference. This creates big shear forces across the face of the wheels and the material, hopefully tearing the correct one apart, and letting you grind much finer then simply crushing does. The top left image shows how the wheel axle can move up and down, but I can’t figure out that set of rods and axles between the wheels, or how the mill is loaded or unloaded.
That still takes a lot of space though. Can we improve it?
This gif inspired this entire post: It is basically a larger version of the Indian kitchen implement below, and with a wider wheel designed to crush and smear rather than chop. This is important with chillis, as using fast spinning grinders can cause a burnt flavour. Does heating have an negative impact on the cob mix? Almost certainly not. Does a large flat wheel system that can pack neatly against one wall have potential in the lab? maybe….
Bakefoldprint 
Leave a comment