Image for post
SmartRap Plus 3D Printer

How Would You Build a 3D Printer Operating Without Electricity?

Moving and controlling a 3D printer using fluidics control logic and pneumatic motors.

ollowing up my story on building a non-electrical economy on other planets, I like to talk specifically about 3D printers. For any space colony I think 3D printers will be one of the most important machinery, because it allows people to build a wide variety of tools, parts and constructions without the need of a large industrial base.

A 3D printer normally requires electricity for several parts:

  • Stepper motors moving the XYZ robot arm.
  • Electronics controlling the stepper motors.
  • Heating up the material used for printing. That could be a laser or just a simple metal rod that gets hot when electricity runs through it.

An XYZ robot arm is one in which all the movements happen typically happen along the X, Y and Z axis rather than being based on rotations in some joints as a typical car welding robot.

Image for post
Image for post
A robot with an XYZ robot arm. You see how get sliding movements along all the axis.

It is possible to build 3D printer not based on XYZ robot arms, but this is the simplest solution. If you want to build considerably larger designs then using a more tradtional robot with rotational joints would be useful.

Lets look at possible non-electric replacements.

Pneumatic Stepper Motor

To be able to print anything accurately you need to be able to move your XYZ robot arm accurately to a given position. There are many ways of doing this.

One approach is to use servo motors. A servo contains a encoder, which keeps track of how far the shaft of the motor has rotated. A motor controller inside the servo keeps track of how far the shaft has rotated compared to how far it has been instructed to go. When the shaft gets very close, the controller will slow it down.

Image for post
Image for post
Inside an electrical stepper motor. The coil magnets define the steps of the motor.

However a much simpler and robust way of accomplishing this is to use a stepper motor. That is because stepper motors don’t require this feedback mechanism and complicated electronics to move the shaft to the desired angle. Instead stepper motors are made physically to only move in steps. A full rotation is made up of fixed number of steps. Each pulse given the motor will move it one step. Unless it gets another pulse, the shaft will be stuck in that position. The shaft never stop in between its steps.

Image for post
A Globe Pneumatic stepper motor. All you need to drive it is pressurized air.

While electrical stepper motors are well know, pneumatic stepper motors do in fact exist as the one shown above. This one is from a company named Globe, and each “pulse” will move the axis 3°. If that is too much or little on can just change that with a gear reducer.

This red motor is made of metal, which makes it difficult to make it from a 3D printer if we wanted to make a sort of self replicating 3D printer, or make it on other planets without easy access to metals such as Venus.

Image for post
Image for post
Inside a pneumatic stepper motor. Notice how the steps are defined similar to how coil magnets define steps in an electrical stepper motor.

However you can find examples online of people making air powered motors entirely 3D printed in plastic, such as the green motor below.

Image for post
A 3D printed air motor

There are also examples of 3D printed pneumatic stepper motors, which you can see in this video. These have quite impressive power despite just being made of plastic. This yellow linear stepper motor from the video has 350 Newton of pull strenght, meaning it can lift 35 Kg.

Image for post
Image for post
3D printed linear pneumatic stepper motor in yellow, lifting 35kg of weights.

Motor Driver

In a normal 3D printer you would have a micro-controller controlling a stepper motor through a motor driver. The motor driver is basically an amplifier. A regular electronic circuit doesn’t produce enough electricity to drive a motor, hence we need a way of amplifying this electricity. What amplification really means in this context is that you use a small current to control a much larger one.

Image for post
How fluid amplifiers work

Small jets of fluid can however change the direction of larger jets of fluid which is utilized in fluidics amplifiers. That means we could e.g. use tiny air pressures from a fluidics computing device, which through a fluidics amplifier could control a pneumatic stepper motor.

Image for post
Left: A model of a Fluidic jet-deflection amplifier with no moving parts. Right: Fluidic oscillator made using the amplifier.

Motor Controller

A step up from a driver, is a motor controller. Basically you can think of it as packaging a micro-controller with a motor driver. The point of this is to provide a higher level interface to other circuits. E.g. so that a motor could be controller by a serial interface, analog signal or PWM (pulse width modulation).

Fluidics Micro-controller

A home made 3D printer today would typically use something like and Arduino micro-controller. Instead we could build a micro-controller based on microfluidic chips. Dr. Michael G. Mauk, of Drexel University write that:

Microfluidic chips can be made as laminates of plastic films such as acrylic (PMMA), polycarbonate (PC) and other polymers. These materials are generally low in cost, widely available in thicknesses ranging from 0.1 mm to several millimeters, and easy to machine with a mill or (in the case of PMMA) laser cutter.

But this is the flexibility of fluidics devices that they can be made in almost any material and with almost any method. Yufeng Zhou at Singapore Centre for 3D Printing details some possibilities for 3D printing of fluidics devices:

Additive manufacturing or 3D printing could produce complicated intricate architectures effectively at rela- tively low cost and infrastructure investment

Zhou mentions some possibilities for channels sizes using different 3D printing technologies. The 3D printer hobby users use is usually of the Fused deposit modeling. The smallest channels possible with this method is around 0.1 mm.

Image for post
How melted filament is deposited using Fused Deposit Modeling.

But there are more advance printing techniques mentioned which gives thinner channels and more complex chips.

The problem with reading these modern papers on micro fluidics is that they are almost exclusively focused on lab-on-a-chip type of applications. This means they are actually sending tiny amounts of chemicals, blood or other samples through the these tiny fluidics channels for analysis. So that is not what we are interested in here. For more specific discussion of fluidics applied to general purpose computing you have to look at papers from the 60s and 70s.

Most notable is the FLODAC fluid computer built in 1964 as a proof of concept. It was made up of 250 fluid NOR gates. It had 4 bit word sizes and a memory of 4 words as well as 4 different instructions:

  • Move
  • Add
  • Jump
  • Halt

This was to demonstrate that a program made with the 4 most fundamental instructions for any computer. FLODAC ran at 10 cycles per second. It has however been theorized that clock rates of 10 to 100 kc (kilo cycles per second) is possible. That sounds extremely low compared to an electronic computer which operates at billions of cycles per second. However that isn’t necessarily as limiting as it sounds. The human brain operates at a measly 30 cycles per second. Still the human brain outperforms almost every computer. It has been calculated that the human brain has a processing power of 6 peta flops. That is like six million billion calculations per second. Which compares favorably to the worlds fastest super computer:

the world’s fastest supercomputer is actually about 30 petaflops. Of course, it cost half a month of China’s GDP to build, and requires 24 megawatts to run and cool, which is about the output of a mid-sized solar power station.

The human does roughly the same with just 20 watts.

How does the human brain achieve this while operating at such low frequency? Due to massive parallelism. Fluidics systems could likewise gain processing power from parallelism. Since you can build them in 3D using simple plastics, you can build quite a lot of channels in small area.

Regardless it ought to have enough processing power to drive a 3D printer. Since a 3D printer is mechanical we can’t make it move insanely fast anyway.

Filament Heating

I am assuming here we are using a 3D printer of the Fused deposit modelingtype. That means we got some sort of thread of plastic which needs to be heated up to deposit in layers.

How we do this depends a bit on our limits. If we really wanted to avoid usage of metals to the minimum, we could utilize some form of sun based heating mechanism. Lenses can be made from plastic and can be used to focus sun lights.

In fact German designer Markus Kayser build a solar sinter 3D printer.

Image for post
Markus Kayser in the dessert using his Solar Sinter 3D printer to build things out of sand.

If metals could be used but not electricity then some form of gas flame could perhaps be used. Valves controlling the gas flow could be controlled by a fluidics device. Alternatively we heat up the metal that does the melting of the plastic using focused sun light.

If metals can’t be used some form of graphene may be possible to use instead. Graphene apparently has amazing heat conductivity properties.

Geek dad, living in Oslo, Norway with passion for UX, Julia programming, science, teaching, reading and writing.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store