# Axial flux motor using PCB winding as electromagnetic coil

[Guide]At the beginning, I just wanted to make a very small drone. But soon realized that in the design, there is a limiting factor, that is the size and weight of the motor. Even a small motor is still a discrete device and needs to be connected to all other electronic components and structural components. So I started to wonder if there is a way to combine these components and reduce some quality.
IEEE SPECTRUM recently reported on the whimsy of an engineer.
Axial flux motor using PCB winding as electromagnetic coil
Each layer of the motor printed circuit board has a set of coils, which are stacked and connected to each other to form a continuous trace.

In the beginning, I just wanted to make a very small drone. But soon realized that in the design, there is a limiting factor, that is the size and weight of the motor. Even a small motor is still a discrete device and needs to be connected to all other electronic components and structural components. So I started to wonder if there is a way to combine these components and reduce some quality.
My inspiration came from how some radio systems use antennas made from copper traces on a printed circuit board (PCB). Can something similar be used to create a strong enough magnetic field to drive a motor? I decided to see if it is possible to use electromagnetic coils made of PCB traces to make an axial flux motor. In an axial flux motor, the electromagnetic coils forming the stator of the motor are installed parallel to the disk-shaped rotor. The permanent magnets are embedded in the disk of the rotor. The stator coil is driven by alternating current to rotate the rotor.
The first challenge is to ensure that I can create enough magnetic flux to turn the rotor. Designing a flat spiral coil trace and letting current flow through it is very simple, but I limited the diameter of my motor to 16 mm so that the diameter of the entire motor is comparable to that of the smallest finished brushless motor. 16 mm means that I can only install a total of 6 coils under the rotor disc, with about 10 turns on each spiral. Ten turns are not enough to generate a large enough magnetic field, but nowadays it is easy to make multilayer PCBs. By printing into stacked coils (with coils on each of the four layers), I can get 40 turns for each coil, enough to turn a rotor.
As the design moved forward, a bigger problem emerged. In order to keep the motor rotating, it is necessary to synchronize the dynamically changing magnetic field between the rotor and the stator. In a typical electric motor driven by alternating current, this synchronization occurs naturally due to the arrangement of the brushes that bridge the stator and rotor. In a brushless motor, what is needed is a control circuit that implements a feedback system.
Left: The completed four-layer printed circuit board. Middle image: Pulses are applied to these coils to drive a 3D printed rotor with embedded permanent magnets.
Right: Although not as powerful as a traditional brushless motor, the PCB is cheaper and lighter.
In a brushless motor drive I made before, I measured the back-EMF as feedback to control the speed. The reason for the back-EMF is that the rotating motor is like a small generator, generating a voltage opposite to the voltage used to drive the motor in the stator coil. Induction of the back electromotive force can provide feedback information about the way the rotor rotates and allow the control circuit to synchronize the coils. But in my PCB motor, the back EMF is too weak to use. To this end, I installed a Hall-effect sensor, which can directly measure the change in the magnetic field to measure the speed of the rotor and its permanent magnets rotating above the sensor. This information is then input into the motor control circuit.
To make the rotor itself, I turned to 3D printing. At first, I made a rotor, which I installed on a separate metal shaft, but then I started printing the snap shaft as an integral part of the rotor. This simplifies the physical components to only the rotor, four permanent magnets, a bearing, and a PCB that provides coils and structural support.
I quickly got my first electric motor. Tests have shown that it can generate a static torque of 0.9 g cm. This was not enough to meet my original goal of manufacturing a motor integrated into a drone, but I realized that this motor could still be used to propel a small and cheap robot wheel along the ground with wheels, so I insisted on researching (motor Usually one of the most expensive parts on the robot). This printed motor can operate at a voltage of 3.5 to 7 volts, although it will heat up significantly at higher voltages. At 5 V, its operating temperature is 70°C, which is still controllable. It draws approximately 250 mA of current.
At present, I have been working hard to increase the torque of the motor (you can follow the research progress I continue to publish on Hackaday https://hackaday.io/project/39494-pcb-motor). By adding a ferrite sheet to the back of the stator coil to contain the coil’s magnetic field lines, I can almost double the torque. I am also working on designing other prototypes with different winding configurations and more stator coils. In addition, I have been trying to use the same technology to build a PCB electric push rod that can drive a 3D printed slider to slide on a row of 12 coils. Also, I am testing a flexible PCB prototype that uses the same printed coil to perform electromagnetic drive. My goal is-even if I still can’t make a small drone that can fly into the sky-start making robots with smaller and simpler mechanical structures than existing robots.
This article was published in the September 2018 issue of IEEE SPECTRUM, entitled “The Printable Motor”.