Electrical
The electrical portion of the ELLIE system includes two major components: COM and DAQ.
COM interfaces with the computer and passes commands, while the DAQ talks to sensors, listens for COM's commands to execute, and collects data.
COM
This portion is responsible for communications between the computer and the system. It is consisted of the following:
ESP32-Wrover (1). Central command and communication unit.
Physical Buttons (3). Used to change the states (check here for details on the state machine)
LED (2)
Since the COM ESP32 and the DAQ ESP32 are connected via WiFi (details here), they both must be mounted high to achieve maximum connectivity. We created a signal tower for the COM side:
DAQ
The ELLIE data acquisition system contains the following:
Printed circuit board (PCB) (1). It keeps all the onboard electronics organized.
ESP32- Wrover (1). Central computing unit.
HX711 breakout boards (4-7). Used to amplify signals and take in data from pressure transducers and load cells.
9V LiPo Battery (1) for controlling the servos that opened and closed the system’s valves. It is directly connected to the PCB.
Servos (2). They are kept separated from the lower voltage signal connections by a large power pour, which prevents the high current from burning traces.
5V relay breakout board (1). Used to set off the system's igniters with a high current burning through very thin wires surrounded by flammable material in the feed system.
Before the PCB board was designed, we tried out various electronic arrangements on breadboards and protoboards:
Sometimes things did not go so well, since solder boards are not as reliable as a PCB. Therefore, we opted for an upgrade. An image of the CAD model of the PCB board follows (credit to Scout Weber, Trevor Zinky, and several other teammates who worked around the clock to design the board):
A voltage divider was used on the HX711 E+ input, since without it, measurements maxed out at around 200 psi. The cause was the programmable gain on the software side. It only allowed twice the minimum gain due to a software bug. Fixing the software would have required changing the Arduino library for the boards, which was much more difficult than implementing the voltage divider.
Additionally, a power pour was implemented on the top in the shape of a yellow rectangle. Its purpose was to shield the delicate traces on that part of the PCB from high current, usually induced by the servos. Then all we have to do is to add a parallel low-resistance path for the current, and nothing will get crispy!
Last updated