Engineers may be able to integrate sensors into rotational mechanism in order to make smart hinges. Or, gears inside motors that tell mechanics how fast the motor is rotating. 3D printing has been used by MIT engineers to make it easy to incorporate sensors into these types.
Although 3D printing has made it possible to quickly fabricate rotational mechanisms, it is still difficult to integrate sensors into designs. Sensors are usually embedded manually because of the complexity of rotating parts. This is typically done after the device has been manufactured.
It is not an easy task to manually integrate sensors. If you embed them in a device, wires can get tangled or block their rotations. However, mounting external sensors will increase the size and limit the motion of the mechanism.
Instead, the new system that MIT researchers created allows a maker of sensors to 3D-print sensors directly into a moving part of a mechanism using conductive 3D printed filament. This allows devices to sense their rotation speed and direction.
Their system, called MechSense allows a maker to produce rotational mechanisms with integrated sensors using a multimaterial 3D printer in one pass. These printers can be used to produce multiple devices using multiple materials.
Researchers created a plugin for SolidWorks, a computer-aided design program, that automatically integrates sensors in a model of the mechanism. This can then be sent to the 3D printer for fabrication.
MechSense allows engineers to quickly prototype devices that have rotating parts like motors or turbines. They can also incorporate sensing directly into their designs. It could be particularly useful for creating tangible user interfaces to augmented reality environments. In these environments, sensing is essential for tracking user movements and interaction with objects.
“Most of our research in our laboratory involves creating fabrication methods for factories and specialized institutions, then making them accessible to the public. 3D printing can be a very affordable tool. How do we give the average maker the tools to develop these kinds of interactive mechanisms. Marwa AlAlawi is a graduate student in mechanical engineering and the lead author of a paper about MechSense.
AlAlawi’s former postdoc in MIT Computer Science and Artificial Intelligence Laboratory CSAIL (now assistant professor at Aarhus Univ.) and Stefanie Mueller (senior author), an associate professor in MIT departments of Electrical Engineering and Computer Science and Mechanical Engineering and a member CSAIL. Also, other collaborators from Accenture Labs and others at MIT. The research will be presented at ACM CHI Conference on Human Factors in Computing Systems.
The researchers used capacitive sensors to incorporate sensors into a rotating mechanism that wouldn’t disrupt its movement.
A capacitor is made up of two plates of conductive material with an insulating material sandwiched in between. A capacitive sensor detects changes in the electric field between plates by measuring the area between them. This information could be used to calculate speed for example.
“Capacitive sensing does not require that there be contact between the opposing conductive plates. You can monitor any changes in the sensor by using capacitive sensors. AlAlawi states that they took advantage of this feature for their sensor design.
Rotational mechanisms usually consist of a rotating element that is located above, below or near a stationary element. For example, a gear spinning on a static shaft over a flat surface. The rotating element is the spinning gear, and the flat surface below it is the stationary element.
The MechSense sensor has three patches made of conductive materials that are printed on the stationary plate. Each patch can be separated from its neighbours by nonconductive material. The rotating plate also includes a fourth patch of conductive material. It has the same area as all three.
The device spins and the rotating capacitor patch overlaps each stationary plate patch in turn as the floating capacitor. Each patch detects the change in capacitance as the overlap between each stationary patch and the rotating patch changes (from fully covered to half covered to completely uncovered).
The floating capacitor does not connect to any circuitry. This means that wires will not get tangled together with rotating components.
Instead, the stationary patches are wired up to electronics using software developed by researchers to convert raw sensor data to estimations of angular location, direction of rotation and speed.
The researchers created a SolidWorks extension to simplify the process of sensor integration for users. The maker specifies the stationary and rotating parts of the mechanism and the center of rotation. After that, the system adds sensor patches automatically to the model.
It doesn’t alter the design in any way. AlAlawi states that it just replaces a part of the device with another material, in this instance conductive material.
They used the system to prototype several devices including a smart lamp for desks that changes its color and brightness depending upon how the user rotates it. A planetary gearbox was also developed, similar to those used in robotic arms.
They also performed technical experiments to improve their sensor design as they prototyped. They discovered that the error rate in sensor data was increasing as the patch sizes were reduced.
We want electronic devices that are smaller in size but still have the ability to perform well, so we need devices that are small enough to be recyclable. “If we keep the same approach, but maybe use a different material or manufacturing method, I believe we can scale down while still accumulating less error using this same geometry,” she states.
AlAlawi, along with her collaborators, will be testing various materials and exploring how to increase the resistance of their sensor design against external noise. They also plan to develop printable sensors to suit other types of moving components.
Accenture Labs funded part of this research.