P18101: CubeSat Solar Sail/public/
Subsystem Build & Test
Table of Contents
Team Vision for Subsystem Level Build & Test Phase
- Received and assembled frame components from SPEX club
- Sent all parts to the machine shop
- Build all four sensors
- Solder and heat shrink sensor pins to wires
- Tested all sensors to ensure sensors and code works
- Painted all booms with spray paint
- Built all four quadrants of the sails
Received and Assembled Frame ComponentsTwo versions of the machined components were received: a machined version with rough tolerances and some design specifications missing for use in prototyping and a second water jet cut set. This second version had all of the required design components that the machined version lacked and much tighter tolerances. Some of the components were still undergoing revisioning and as such were 3D-printed for use in testing and assembly. These machined components and their integration with 3D-printed components can be seen below.
Test Results Summary
The Test Plans and their results can be viewed in detail here.
Risk and Problem Tracking
Sail QuadrantsIt was discovered in a previous phase that our original target of 32 square meters for the deployed sail was too ambitious given our timeline and budget. This was because in order to fit a larger solar sail, a thinner, much more expensive material would have had to be ordered. As a result, this phase a goal was to complete the four quadrants of a 16 square meter sail. There were several issues that came up with the construction of these quadrants which were overcome. As mentioned in previous phases, the metalized mylar blankets the sail is being constructed from do not have straight edges which could be incorporated into the sail edges. To resolve this, the edges of each blanket had to be trimmed, reducing the size of each sail. This caused the dimensions of each blanket to change, which necessitated unique calculations of the size of the components for each sail quadrant. This was taken into consideration, but as we progressed other issues became apparent. One of these was issues with individual blankets having an incorrectly sized fold. This created an issue when folding, making the packing efficiency lower and increasing the volume of the folded sail. To address this, blankets with this problem were used to make the small corner segments where this would not impact the folding as much in future quadrants. Another issue was with the taping of the sail components. In the first two sail quadrants there were places where the metalized mylar tape was placed over a fold line. This made the packing efficiency lower and resulted in the sail taking up additional volume. This was addressed in the final two sails by cutting the sail material so the tape would not be located on a fold line.
SensorsThe sensors were tested against a base case of the same sensor that functioned properly. However, the purchased sensors did not work originally. The output of the sensors, which should have been a 0 or 1, was always a 0 and would not change values. Originally, this was assumed to be an error in the soldering of the sensors, but after extensive testing and consulting subject matter experts, it was determined that this was not the case. The next solution attempted was to purchase new sensors and test those in case we had received a defective batch. However, after testing the new sensors they did not work either. The supplier was then contacted to discuss the issues with the sensors. Throughout this dialog, it was revealed that the sensors, while nominally identical to the base test we were testing against, would not function using the same code. Although the supplier could not provide a reason for this, they were able to guide us into the proper way of coding them. Using this information, we were able to successfully use the sensors.
Motor TorqueOriginally the motor and geneva cam assembly functioned properly with the motor able to easily turn the cam and attached to the spindle. However, once the booms were attached to the spindle, the motor no longer had enough torque to turn the spindle. The additional force required to turn the spindle with the booms was too much when the pin rotating the cam moved to the point closet to the center of the cam where the required torque was the highest. Two possible solutions to this issue were considered; purchasing a new motor and designing a cam with more, less deep slots. After consideration, a new motor was purchased that can provide 3 times as much torque with the same physical dimensions so no other design changes will be necessary.
Painted BoomsThe painting of the boom markings to track the boom deployment and retraction was originally done with 1" increments. To make this as simple as possible, 1" wide masking tape was purchased to cover the increments that were to remain unpainted. However, the tape was actually 15/16 inches, and to maintain the 1" increments two pieces of tape had to be overlapped. When this was painted with the spray paint and the tape was removed, it was found that paint had leaked underneath the edges where the tape was overlapped and had created a small gap. In addition, the amount of paint applied was too high, causing additional leakage and bleeding through the tape in some locations. To correct this, painting was done with a much thinner coat and only a single piece of tape with the increments being 15/16 inches on the taped sections and 17/16 inches on the untaped ones. This does not cause an issue with the deployment code.
These processes can be seen in the problem tracking document here.
Engineers Week Celebration Presentation
Functional Demo Materials
Plans for next phase
- Test Genva Cam with measuring tape booms and machined inter mount
- Solder H – Bridge and all the other wires to protoboard
- Paint final tape measures
- Do full test deployment with booms, sensors, and small sized sails