•The secret to cool bodies and glazes is a lot of testing.
•The secret to know what to test is material and chemistry knowledge.
•The secret to learning from testing is documentation.
•The place to test, do the chemistry and document is an account at https://insight-live.com
•The place to get the knowledge is https://digitalfire.com
3D Printing Ceramics
It is becoming more practical for potters and ceramic artists or entrepreneurs to take on projects never before possible because of the increasing accessibility of 3D printing of ceramic materials or of materials that can act as piece molds or block molds. Objects themselves can be printed directly by extruding layers of a ceramic paste from a nozzle and by fusing powder particles layer-by-layer. This is an additive process as opposed to subtractive (where material is cut away from a block to create a 3D object). The latter is more practical for making molds of relief designs for pressing the faces of tiles or for ramp pressing plastic clay.
The practicality of additive processes and quality and shapes that can be made are still big limiting factors to the technology. However it is clear that refractories (like shelves, posts, supports and even entire kilns) can also be printed and the smoothness of finish is not nearly as important. By printing a honeycomb structure within the refractory they can be made very light and well insulating and can be made from much more expensive materials than would otherwise be possible. Even things like stainless steel can be printed, this enables making complex molds for use with plastic and dust pressing processes.
Many technologies must be understood and exploited to make this possible in an application. One of the most difficult to surmount is learning the software design and conversion tools. This can be very confusing since hundreds of products are available. However there are standards and printers expect to receive a specific file format: STL. 3D design software can be very expensive but there are open source solutions. A major enabler has been a policy change by AutoDesk, the maker of industry standard tools for many years. They are building their future of mechanical design around Fusion 360 and, as of 2015, are releasing it for free use by education and business earning less than 100,000 per year. This product has exceptional online resources and training and this development could be the most important single factor the puts 3D into prime-time ceramic production.
Another factor is the RepRap international movement to develop open source hardware and software platforms for 3D printing. Reprap printers use standard buy-at-a-hardware-store parts or ones that the printer itself can make. This means that anyone can buy and assemble a printer to learn many details of their mechanics and operation. Once understood a printer of any size could theoretically be constructed. In ceramics the focus is on the print head and how to deliver a thick paste to it, extrude it an a constant rate and be able to turn the flow on and off in an instant. The latter can be a real challenge and one solution is to pressure-feed the paste to a print head having a barrel and auger like a pugmill and control the auger using a stepper motor to feed the actual extruder nozzle.
There is a problem with scaling to a bigger size. The printhead or platform must be able to move on at least two axes. There must be a minimum of mass resisting movement in order to have precise and quick movement. But clay is heavy and if the printhead is full of it it cannot be responsive (e.g. Delta printer designs). Likewise if the item being printed is heavy (and soft) the bed on which it sits cannot move too quick.
A major challenge is making the clay set fast enough so that the next layer applied over it will have a firm base. When plastics are extruded they simply cool and firm up, but clay pastes are soft and fragile. Additions of polymers can help set them after extrusion. Augmenting the water with alcohol (e.g. ethanol is preferred over isopropyl alcohol because it carries off more water as it evaporates) speeds evaporation. Pure ethanol and clay is flammable and the plasticity and dry strength are poor. But 50:50 water:ethanol is not flammable and workability is excellent. However, even though the clay feels cold evaporation does not proceed that quickly, it needs a fan to stiffen up fast enough. Under greater pressure stiffer pastes of lower water content can be delivered. When large objects are printed, fans and the extra time between layer delivery may be enough to enable structural integrity. In some types of 3D printing, support structures of a different material are printed with the item and these are later removed, this could be practical for ceramics also.
An exciting technology is laser fusing of powder, even metal powder. In this way stainless steel can be printed. This enables printing complex molds for dust pressing of tiles.
Notwithstanding all of this, printers designed for clay are appearing on the market (with lots of excitement) and more are coming. You could easily spend $10k but use caution, the laws of physics and common sense apply. Machines have differing priorities. Those that must push clay through a thin tube to a tiny pugmill in the printhead (e.g. the Lutum and WASP machines) will obviously need soft clay and to be reloaded more often and they may not work well with deflocculated bodies (the auger having trouble getting traction). The PotterBot is moving the entire piece constantly (on the x-y axes), and when that piece gets large it becomes rather like a big cube of jello on a plate being jerked around! Obviously the body will need to be stable or it will just collapse. The only solution is often to simply print really slow. This can make the novelty of 3D printing clay wear off pretty fast!
Don’t be stuck with a fancy machine and no clay that works with it. And, you can be sure, the manufacturers are going to follow the revenue model of ink jet printers so brace yourself when you find out the price of cartridges. It might be best to make your own bodies, that is what the pros are doing. You need a propeller mixer to blend the powder and water (more powerful mixers will do this much better). Run the mixer until all air bubbles have surfaced (to de-air it) and all particle surfaces are wetted (this could take 15 minutes). Then pour it on a plaster bat to dewater to the needed consistency. A hand extruder can be used to create the diameter needed to feed the machine (the clay needs to be soft). The character and suitability of the body will be a big part of any success you have, and understanding the recipe and being able to control it will give you a big edge (especially if you want to incorporate alcohol). Click the links below for information on a mixer and plaster table.
Clay suppliers are also producing clays for this. One of the bodies has 40% 80 mesh grog. Its water content is 22% (an unconventional recipe, likely the grog is added to channel water to speed up drying and add plastic stability). A body with 40% grog would normally only need around 18% water for normal modelling so this amount explains the high drying shrinkage they claim (8.5%). Another porcelain has 25% water yet only has a drying shrinkage of 6.5%. This would make it non-plastic and fast drying (plastic throwing porcelains would be around 22% with the same or lower shrinkage).
To make your own body follow the same pattern. Start with and existing recipe for a plastic pottery clay (your body manufacturer may give you the generic recipe of the body you already use). If it is highly plastic it may be able to tolerate a large percentage of fine grog and work well. If you want smooth porcelain then change any ball clay in the recipe to kaolin, that will drop plasticity. Drop out bentonite. If the body is not plastic enough then add some bentonite back in. The recipe will need 20-25% silica so glazes are easy to fit. And it will need enough feldspar to make the body vitreous. For example, the popular 50:25:25 recipe for cone 10 is 50 clay, 25 feldspar and 25 silica.
The computer board on a common RepRap 3D printer
This controls all the stepper motors and the heating element and watches temperature and position sensors. It run open source software that knows how to interpret an STL file. As it reads that file steps the z-axis upward for each slice and then prints that layer by moving the printhead and movable bed for the x and y axes.
The movable printing bed on a common 3D RepRap printer
Objects are printed on a platform that moves along the y-axis. The bed is attached to bushings that run along stainless steel rods. Its position is controlled by a rubber belt that feeds around a pulley in the front and around a gear on a stepper motor at the back. It is heated to prevent printed layers from hardening too rapidly or the piece warping during printing.
The printhead of a common RepRap printer
The assembly consists of stepper motor with its own cooling fan and a heated brass nozzle mounted in a small aluminum block (at the bottom). The nozzle has a heat sensor and its own cooling fan). A plastic filament feeds down through a hole in a laser-cut aluminum spring loaded part. It has an attached roller that forces the filament against a gear fastened to the motor shaft. When the motor steps it pulls in the filament and feeds it down into the heated print head below. The entire head assembly is screwed to a plate that is in turn screwed to bushings that are pulled along the x-axis by a belt controlled by another stepper motor. The computer can thus control the rate of filament feed, the temperature of the nozzle and the x-position of the entire head.
X and Z axis stepper motors on a RepRap printer
In this printer (which is being assembled) the printhead moves along two stainless steel rods (for the x-axis). Its position is controlled by the front top stepper motor (which has a gear through which runs a rubber belt attached to the printhead. The two lower stepper motors with worm gears attached to their shafts control the vertical z-axis position of the printhead assembly. Since the computer controls these motors it can move the head to any position on the x or z axis. Vertical z-movement is slower and more precise since it determines the thickness of each slice to be printed.
Printing a prototype propeller for my Lightnin lab mixer
An example of how handy the ability to print in 3D can be. The worn-out stainless propeller costs $300 to replace. But the size and pitch of the blades is not ideal for the work I do. I can experiment with different configurations and print them in hard plastic, then have a new stainless one printed at shapeways.com when I am ready. These plastic propellers are surprisingly durable and it was easy to print one the with a hole the exact size needed, it fits tight on the shaft and never moves.
Making ceramic tile shapes by 3D printing your own cookie cutters
This was done on an affordable RepRap printer. The red plastic templates were drawn in Illustrator, extruded in Fusion 360 and sliced and printed using Simplify3D (which took about 30 minutes each). The round wooden block was used to press these cookie-cutters into the clay. The plastic wrap made sticking a non issue (and rounds the corners nicely). The clay is a low fire, buff burning talc body (Plainsman L212). Commercial bottled glazes were applied by brushing (in three coats) after bisque. The tiles were fired at cone 03. This is an old classic design that I discovered when researching Damascus tile. The toughest obstacle was learning how to use Fusion 360. It turns out that cookie cutters are a starter project for many 3D software packages, there are lots of videos on making them.
Out Bound Links
In Bound Links
By Tony Hansen