The mechanical subsystem is composed of the thermal and structural subsystems. In the former, it was determined that no passive or thermal control system is necessary for the ADCS. This is because the steady-state temperature of the system was expected to be 0°C.

The structural subsystem is composed of reaction wheels, motor housings, mounts for the reaction wheels, and the CubeSat skeleton. As the electrical subsystem and dynamics subsystem designs evolved, mechanical subsystem components experienced multiple design iterations to accommodate new design requirements.

Magnetorquer mounts were designed and meant to be integrated in the ADCS for detumbling purposes, but were omitted due to limited funds. Originally the mount was designed where the caps would be placed in the center of the rod (Figure 1). The design was revised by moving the caps to the end of the rod for accessibility, and to ensure the coils are rigid. To maximize available space, a square-based magnetorquer design was created (Figure 2).

Reaction wheels were designed and manufactured. A flywheel design was initially created to minimize overall system mass. The alternative disc-based design was focused around on obtaining more angular momentum; to obtain more control authority. FEA results showed that both reaction wheel designs would not fail at operating conditions, although the initial design exhibited greater stress concentrations (Figure 2). Because more control authority was deemed more beneficial for the ADCS, the reaction wheel's size was increased and its frontal lip was removed (Figure 3). FEA results that no failures would occur and that there was no consequence in increasing the reaction wheel's size. The only limiting factor was the space available within the CubeSat.

For ADCSs that incorporate four reaction wheels, the most common reaction wheel configurations are the skew, pyramidal, and tetrahedral configurations. The skew configuration was initially considered to be ideal because it is the most space-efficient and most straightforward to implement a control algorithm. 

Through further literature review, it was determined that the tetrahedral configuration is the overall better choice (Figure 2). Although the control algorithm is more complicated, this configuration is more optimal because the load is  evenly distributed across each reaction wheel. Motor housings were created to isolate potential motor failure incidents. The motor mount configuration was revised to support the attachment of these housings (Figure 3). Originally, the motor mounts were created where the ADCS would slide within a 1U CubeSat. The mounts were redesigned with the consideration that the ADCS would be its its own standalone 0.5U unit (Figure 4 - 5).

The electrical subsystem is primarily composed of the following:

• The Arduino Nano IoT 33 is used to control other electrical components and for data processing.

• The ADIS16460 is the primary inertial measurement unit (IMU) to measure the CubeSat's angular velocity and acceleration relative to a particular axis.

• The LMSM6D3  is a secondary inertial measurement unit.

• The INA219 current sensors are used to monitor and manage component power usage.

• Originally, the system was composed of the Faulhaber 1509 Brushless DC (BLDC) motors to drive the reaction wheels. The downside is that each one would require an external motor driver when there were already volumetric limitations. Four Faulhaber 2610 motors were later selected because they have an integrated motor driver.

• The TPS7B3601DCQR power management integrated circuit (PMIC) is used to regulate 3.3 V to the microcontroller units (MCUs).

Essential instruments for attitude control involved the gyroscope and the accelerometer. Both of these instruments were calibrated to improve system accuracy and efficiency. 

Initially, the gyroscope outputted non-zero values when idle (Figure 1). To calibrate the gyroscope, 2,000 data points with a 0.02 second time delay were obtained. These data points were averaged and subtracted from the measurements of the gyroscope for each axis (Figure 2).

Based on the accelerometer data sheet, measurements should output 1G when idle. The same gyroscope calibration process was applied to the accelerometer.

Using the accelerometer and gyroscope, the pitch and roll angles of the CubeSat were obtained (Figure 6). This involved using gyroscope integration, trigonometric methods, and a Kalman filter. Due to monetary and time constraints, a magnometer couldn't be integrated to obtain the final yaw angle.

At SJSU, freshmen work with a senior team to gain experience within the aeronautics or astronautics realm of aerospace. In which the freshmen teams gain an opportunity to apply their CAD or programming knowledge. Taking initiative, I was primarily responsible for advising the mechanical and software freshmen teams. To guide these teams, I held an introductory meeting discussing TechEdSat's background, objectives, and projects the teams could do. For each project, I discussed the importance, background, objectives, and requirements on an elementary level. Additionally, conceptual and technical resources for each project were provided. Throughout the duration of TechEdSat, I answered the questions of the freshmen teams and guided them whenever they were stuck.

[1] "General Environmental Verification Standard (GEVS)," NASA, Greenbelt, 2021.

[2] "WORKMANSHIP STANDARD FOR CRIMPING, INTERCONNECTING CABLES, HARNESSES, AND WIRING," NASA, District of Columbia, 2016.

[3] "Generic Standard on Printed Board Design," IPC, Northbrook, 2003.