Geopolymer Mini Extruder

We are 4 students in mechanics and materials. We have a one year project that consists in finding formulations of local materials that would be printable and usable for construction. For the printable aspect, we decided to make a mini extruder in order to test the extrudability of our different formulations. We needed a little recipient that allows us to test a little quantity of formulation so that we don't waste too much material during our tests.

This project can be improved, or adapted to the use you want :

• We use it for earth and geopolymer, but this can be adapted for clay or other materials,
• We used a 300 ml syringe, but you can do more or less,
• We use it fixed, and we move the container which collects the extruded rod, but you can adapt it to put the mini extruder on a mechanical arm in order to print the shape that you want!

Be aware that as this is a school project, we had some constraints to respect in order to be in adequation with our coursework. Here is the list of the main constraints we had :

• use 3D printing
• use a milling machine
• use an arduino controller
• limited time
• limited budget --> we mainly used stuff that we already had at the school

Of course you are free to adapt the step we describe below.

Before we start to describe how we made our mini-extruder, we want to quickly explain what are the forces involved when you extrude some material. You can skip this part if you are not interested in this aspect of the project. Everything that is written below is based on an article called Use of ram extruder as a combined rheo-tribometer to study the behaviour of high yield stress fluids at low strain rate, by Arnoud Perrot and all (2012). We won't go in to much details here but the article does so if you want to now more about the theory we invite you to check it out !

Basically what is important to have in mind is that we need to push a high yield stress fluid through the syringe. To do that we need to composate two forces. The first one is related to the friction of the fluid on the walls of the syringe. The second one is due to the reduction of section in the nozzle die. You can visualize these contributions on the attached image. K_w is the denoted as the wall friction stress. It is related to F_fr, the wall friction force, by some equation that we won't describe here. This force is developed along the extruder body in the first zone. F_pl is the shaping force which occures in the second and the third zone. This force is the sum of three contributions :

• The force required for bulk deformation --> plastic work needed to reduce the billet diameter
• The force required to overcome the friction of the fluid on the dead zone which is a zone where the fluid remains blocked around the die entry. It is known to act like a conical entry die. You can see the dead zone on the attached image
• The firction force on the exit orifice walls

Kowing that, using the article and the thesis of Julien Archez (Formulation de composites à base de liants basse température type géopolymère à base d’argilites et de différents renforts : réalisation d’une pièce par fabrication additive, 2020) we can get an estimation of the force we will need to reach in order to extrude our material. All maths done we estimated we would need to push around 400N. We can now explain how built our extruder.

You will need :
- SBR16 rails in order to guide the piston in translation

- A syringe of 300 ml (or the quantity you like)

- A 3D printer, and PLA (we have used approximatively 1241 g of PLA)

- A stepper motor (reference : 34HS31-5504S), a driver (DM860 in our case), an Arduino Card (Arduino Uno) and a computer

- A threaded rod (we used an M8)

- Wood in order to make a support

-A microswitch (we chose a GT032)

- Motivation !

The aim of our mini extruder is to push the piston of the syringe with the motor. So the piston needs to be connected with the rail's pads. The piece that will do that connection must be connected to the motor with a threaded rod, in order to push the piston. This piece will be called "mobile part". To do the worm system, we entered by force a nut in the mobile part in which the screw will turn. With this technique, if the motor and the rod are fixed, when the rod turns, the mobile part will move, guided by the rail's pads and will cause the translation of the piston.

For the nut's entering in force, if you want to do your own conception of the pieces, we advice you to put exactly the dimension of the screw on you CAO. The nut will enter in the hole with a hammer and will not move in the hole, and that's what we are looking for.

We also need two other fixed parts that will maintain the fixed part of the syringe. We add holes in it in order for the worm to be guided (and help the alignment of the extruder).

The stl file of the CAO we made are uploaded.

Hey ! Remember the 400N force we talked about in the introduction ? Well, that's a lot of force and the fixed-part-nozzle will have to resist all of it... So before we print the part (which will take around one entire day) we wanted to be sure that it was not going to break. As we did the CAD using CATIAV5, we used the finite element module to simulate a 1.5kN force acting on the part (securiy coeff of 3.75). So if the part resist to this 1.5kN force, it should resist to the extrusion of our material. Using a Von-Mises criteria, we found (with a quick convergence analysis) the max stress was around 12 MPa which is below the elastic stength limit of the PLA we used (29 MPa, the value was found on the datasheet of the seller). You can see an image of the finite element model on the second image. Therefore, theorically, the part should not break. We will see in the future how it resist to the extrusions !

The mini extruder must be vertical so that the extrusion use the help of gravity to work. For that, you have to find, or do it with a milling machine, a wood support that would hold your mini extruder vertical.The stress on your support is not really hard but you must take into account the fact that there is a 2 kilos engine above your support. Thus, the "feets" of your support should be long enough so everything will not fall down on the weight of the engine.

Moreover, the part where all componants are fixed should be a little bit upper than the ground so the extrusion

Taking that into account, we chose the dimensions as on the image below.

For the motorization we use a stepper motor. This type of motor have an important torque and can be turned by a precise angle. This is extremely important for our project because we need to control the process of extraction precisely.

But first, what do you have to know about stepper motor?

The stepper motor is known for its property of converting a train of input pulses (typically square waves) into a precisely defined increment in the shaft’s rotational position. Each pulse rotates the shaft through a fixed angle. Stepper motors effectively have multiple "toothed" electromagnets arranged as a stator around a central rotor, a gear-shaped piece of iron. The electromagnets are energized by an external driver for our project. To make the motor shaft turn, first, one electromagnet is given power, which magnetically attracts the gear's teeth. When the gear's teeth are aligned to the first electromagnet, they are slightly offset from the next electromagnet. This means that when the next electromagnet is turned on and the first is turned off, the gear rotates slightly to align with the next one. From there the process is repeated. Each of those rotations is called a "step", with an integer number of steps making a full rotation. In that way, the motor can be turned by a precise angle. If you want to see a quick video which explain what is a stepper motor with some animations here is a link you can check stepper motor video.

The driver is connected to different components: the motor, the Arduino to control the system and a power supply component to energy the system. The driver circuit give us countless opportunities to improve the system. If you want to learn more about drivers for bi-polar stepper motor here is a link you can check (sorry it's in french) driver

Firstly it gives an important tension to the motor which can not be given by the Arduino: 20 V in our case and the intensity can be chosen using to the 3 first boutons of the driver ( look at the picture to see them ). We choose 6,5A which give us the torque we need to push the syringe. Be aware that every stepper motor has an intensity limit and you souldn't go higher than this limit otherwise you might damage your motor.

Secondly with the 4 last boutons of the driver you can control the number of step you need to do a full rotation of your motor. It can be extremely useful if you need to be precise or to increase/decrease the rotation speed of the motor.

Finally with the born ENA you can activate or deactivate your motor, with the born your make one step of the full rotation, and with DIR you control motor’s sense of rotation.

The last point give us the different born we used to link the driver with the Arduino. Refer to the picture to see the connection between the driver and the Arduino. We've decided to use the serial monitor in order to control the speed and the rotational direction of the stepper motor. Everything is described in the code which you can download.

The aim of this step is to adapt the extrusion head of the syringe. In fact, a usual nozzle of a syringe is too little to extrude the geopolymer or whatever paste you want to extrude. So we heated a cutter blade in order to make the plastic softer before cutting it, and we remove the little nozzle. It was not possible to make a clean cut with the cutter, so we finished with a lime. Then, we put the new extrusion head that with printed. For that, made a system that allows us to test different types of extrusion head : we stick the extrusion head support on the syringe and we screw the extrusion hand on the support. The one on the picture has a diameter of 1 cm.

The stl files of the support of the extrusion head and the extrusion head uploaded.

We realized a bit late that we might some issues with the conception we had. Indeed, in our CAD, the threaded rod is in compression while pushing the piston. This might cause the bulkling of the rod if the load is to heavy. To confirm our intuition, we deed some quick maths to get an estimation of the critical load of the rod. Assuming the threaded rod acts like a beam which is fixed at one extrimity (the stepper motor one) and that we have a concentrated force of F acting on the other extremity, we can calculate the critical load (F_cr) using an Euler-Bernoulli model. You can see the expression of the critical load on the first image and here is a link which explains how the calculation is made calculation of the critical load. On the second image, you can see the variation of this critical load with the free length of the beam (which in our case is the length between the the stepper motor and the mobile part). This length is around 50cm in our initial conception which corresponds to F_cr = 240N... It is not enougth ! Even if this modelisation of our threaded rod is fairely simple, we can see that we might have a problem if we indeed reach 400N or more during an extrusion.

So we needed to solve this problem. We considered two solutions :

• If we put the motor below the extrusion head, and that we pull the piston rather than pushing it, there is no need of a nut to take bak the effort. Given the configuration of the object, we cannot do that, but we could put the motor lower and on the side so that it doesn't disturb the extrusion, and have a system of gear.
• In order to avoid that for the worm blaze in compression while pushing the piston, we put a nut and a washer behind the fixed middle part in order to make the lower part in traction rather than the upper part in compression (cf schema uploaded). For that, you need to screw the nut on the worm before fixing the middle fixed part. Then, you put everything in its place, including the motor coupled to the worm, and before turning the motor on, you preload a little the worm by tightening the nut against the middle fixed part.

We could not do the first solution (even if it might be more effective) because we had time constraints for our projects. But we invite you to try it and share your solutions with us ! So we did the second one.

This step is the most exciting because we see our project taking shape! But it is not the easiest.

You need to put a nut in the hole made for that purpose on the mobile part, screw the mobile part to the rail pads, screw the worm in the nut, and pass it in the holes of the 2 other fixed parts, fix the motor, put the syringe on the two fixed parts, and the piston on the mobile part.

For the coupler that lies the motor shaft and the threaded rod we used an OpenScad file that we adapted in order to have the right diameters for the shaft and worm. We also tried to make one ourselves, but this wasn't the right idea : it goes faster to adapted one already done.

We decided to add a microswitch in order to stop the stroke of the piston. Thanks to this, we no longer have to worry about stopping the motor. Moreover the piston will now stop at the exact same place every time we extrude some material. This means we have a better control on the quantity we extrude.

We decided to use a microswitch GT032. It is a very cheap one and it is compatible with the Arduino UNO. The set-up and an example of code are given in the data-sheet which we have uploaded so you can check it if you are interested.

We have integrated the microswitch to our system and we have modified the code. You can download it in the step 4.

The principle of this microswitch is that when the switch is not pushed, the output and the 5V are connected so we read the "HIGH" value with the code. When the switch is pushed the output and the ground are connected so we read the "LOW" value with the code.

What we have decided to do in our code is : if the switch is pushed, the piston goes little bit upward and stops.