Back

Clay 3D Printer

Since knowing what 3D printers were I have been a huge fan of clay 3D printers. The possibility to shape what is essentially stone into anything you want in a few minutes is just bizarre.

I have done my fair share of printing with clay, but I had some shortcomings in my knowledge on the mechanical part.

Even though I have always wanted to create my own 3D printer I did not posses the knowledge and know how to do so. luckily for me I had a 20 week internship at the 3D printing lab of NHL Stenden Leeuwarden, where I learned a huge amount about 3D printers. Even more lucky for me is that my supervisor gave me a project to build my own clay 3D printer halfway through my internship!

Process

I started with the frame of a Velleman K8200, a Eurotec20A compressor, screw, a tube that fits around it with a nozzle end, and a stepper motor. My task was to incorporate these parts into a design, order the remaining parts, design/print them, then adjust and upload the firmware for the functionality of the clay 3D printer.

My approach was to work backwards from the nozzle to the compressor, so I started with a design for the printhead. The printhead is the part where I already have most of the components; I had to find a way to efficiently attach these components to each other. I also had to immediately look for a way to allow the clay to run into the printhead.

My printhead is inspired by a design from Jonathan Keep, an artist and ceramist. The principle is that the 'body' of the printhead is attached to the stepper motor with vertical screws. The body has a hole with the outer diameter of the screw attached to the stepper motor; the clay runs through the grooves of the screw. On the side of the body at the height of the beginning of the screw, there is another hole; this hole is a connection between the printhead and a tube through which the clay is fed to the printhead, from now on I refer to this tube and its components as the 'feed'.

Eurotec 20A

Velleman K200

Starting equipment

Tube mount

Feed mount

Feed connected

Before I mounted the housing on the stepper motor, I made a hole in the housing with the same diameter as the feed. The housing is secured by a clamping system that I will use more often in the design of the clay 3D printer. The clamping system contains two holes for the vertical thread; this thread also fits into two of the 4 holes in the stepper motor.

Then, I added another clamp, this time at the height of the hole I had made earlier. In the clamp itself, there is also a hole with a threaded screw where a tube connector fits, which I ordered along with a tube of 6mm outer diameter and 4mm inner diameter. By using the 3D printed clamping systems, the screw is still able to move, and I can immediately create an opening for the feed. For now, the printhead is finished, and I will move on to the container for the clay and the delivery of the air pressure. An adjustment that can be made later to the printhead or feed related to the delivery of air pressure. If I want the clay to be able to 'retract' during printing, I will have to let out the built-up air pressure somewhere, for example, through a valve that can open and close. For this retraction, I can of course reverse the screw, but between the end of the screw and the nozzle hole, there is a small space where clay resides that is normally pushed through by the air pressure; if the air pressure is not removed, this clay will come out of the nozzle at times when I do not want it to.

If I print in 'vase mode/spiral mode,' this is not a problem because the clay goes out of the nozzle in a constant flow from beginning to end. But if I want to print complex angles, some form of retraction will be crucial; although this is a problem for which I can only find a concrete solution during testing.

Compressor with connector

Container

The container is inspired by a pneumatic Boilie gun, which is used to make dough balls with air pressure. There is an opening at the top for air pressure and an opening at the bottom for the dough, in this case, clay. For both openings, I again use the tube connectors. These tube connectors have a capacity of up to 14 bar, which makes them more than suitable for use as connectors for the air pressure supply.

The air pressure pushes a disc located at the top downwards, which in turn pushes the clay downwards.

For the compressor, I printed a connector that ensures the ordered tube connector fits into the compressor. In this way, I can use the same tubes and connectors for both air pressure delivery and the feed, which is very convenient for replacing parts. On the compressor itself, there is a regulator that I set between 4-6 bar, depending on how much air pressure I need.

Now, all parts still need to be attached to the frame of the printer. The container is not necessarily fixed but rests on a platform that slides into the profile. Since the container extends above the printer, the top would normally come into contact with the upper profile; by creating distance between the container and the profile, this is prevented, and the printer can safely move up and down. This is a good moment to mention that the profile, which holds the container, printhead, rapid drying element, and side-layer scraping, only moves up and down. The bed moves while the head remains stationary. The fact that the printhead is fixed allows for more weight to be placed on it without causing too much strain on the printhead. Also, I can use the entire length of the profile because the printhead does not need to move from one side to the other.

I secured the printhead with a clamping system that slides into the profile. The right side is reserved for the future rapid drying element and side-layer scraping. I also removed the heater bed and attached a plasterboard on top of the frame.

All parts mounted on frame

Filling tube with clay

I use G&S 254 STONEWARE WHITE because this clay should be workable for the screw and not too sticky. I basically press the clay in the inner tube and work form outside to inside to get rid of any air bubbles. The inner tube gets slid into the container where its locked in place by the two caps.

With the current design, I want to finish the printer, but of course, it will not work directly without fine-tuning.

The biggest issue I faced was with the connections of the threads. For the compressor, I printed a coupling to attach my connectors, which let a small amount of air through, making it difficult for the compressor to build up pressure. The same was true for the lid of the top of the container, where most of the air escaped. This meant that the clay was not delivered to the screw quickly enough or even at all.

Especially with 3D prints, it is difficult to get these parts accurate enough to be airtight. So, I wrapped PTFE tape around the screws; multiple layers of it so that the tape is pressed down when the lid is screwed on. I did this at the connection of the compressor and with the rest of my connectors as well.

Now, the clay is efficiently delivered to the screw, which is crucial. If the screw does not always receive new clay, the flow coming out of the nozzle will also not be consistent, at least not with the nozzle I am using. Between the screw and the end of the nozzle is space where the clay comes together, which could compensate for any shortcomings in clay, but it is not perfect.

Auto Homing working

First print with decent settings

Air pressure working

Extrusion working

First print with perceivable second layer

I took a basic firmware to which I made adjustments with the specifications of my printer. I spent two afternoons trying to print, making a firmware adjustment, re-uploading, and reprinting until I got the desired results. The most important settings I had to adjust were the steps per mm for the stepper motors, which were far from what was needed. For example, the extruder motor was initially set at 690 steps per mm, but 2650 was required for my purposes. Of course, printing with a material very different from plastic filament, so settings such as material thickness, print speed, nozzle diameter, and layer height/width also had to be adjusted.

Eventually, I became very satisfied with what comes out of the nozzle and how the printer moves, but I am very dissatisfied with what happens once the clay comes out of the nozzle. The printer lays down a good line, but the clay does not stick well to the plasterboard and does stick to itself. As a result, the clay already laid down is taken along by the clay coming out of the nozzle. By adjusting layer height, you have that at a small layer height the first layer is flattened and generally comes out nicely, but the second layer will be extruded too much. Now, while writing this report, I've come up with the idea to keep the first layer as is and give the second layer ±80% extrusion. Unfortunately, I am typing this on my last day of internship and my school projects have started again.

This is also a good moment to confess that I have not progressed further than this. I have set up a functioning machine that still needs a lot of calibration in its printing settings, maybe even a different type of clay would help.

Despite not being able to deliver a final product, I have gained the knowledge and experience to be able to do this in the future with more time. I have researched and tried out many things that I have not mentioned in this report simply to find out how it works. As a result, I know exactly how my machine is put together and how I can anticipate on it.