The four subsystems of this project are mechanical, power regulation, amplification and digitization, and data collection and distribution. The mechanical subsystem is responsible for the overall structure of the satellite as well as the interfaces between the avionics stack, the radiation tolerant computer stack, and the solar cell experiment. The other three subsystems are responsible for regulating power from the satellite to the solar cell experiment, amplifying and digitizing radiation data, and logging data for telemetry distribution.
Exploded view of RadSat
RadSat is a 3U CubeSat. One U is a cube approximately 10cm on a side, this puts the RadSat mission at a 10x10x30cm rectangular prism. This form factor was determined by CubeSat standards and through the use of a commercially purchased chassis.
The interior components of the satellite are split into two major sub-systems, the avionics and the payload. The avionic sub-system consists of all components required to make the satellite functional. This includes managing power, handling communications, and attitude control. The design of these components have been reused from a previous CubeSat mission developed by the Space Science and Engineering Laboratory located at MSU. By reusing components the design and build time of RadSat has been drastically reduced. The reliability of the system has also been increased because of these components proven heritage.
The avionics stack is able to communicate directly with the payloads through mechanical and electrical interfaces allowing it to supply power and receive data. The payloads for this mission consist of a Radiation Tolerant Computer System (RTCS) and a Solar-Cell Experiment designed to detect ionizing radiation. These two systems follow the already established design of the avionics stack to ensure compatibility between the sub-systems.
Schematic of Power Regulator
The radiation sensor board will have connection to an unregulated 8.7V source from the satellite battery, and this must be regulated to a usable voltage level. The schematic in figure to the right shows the design that will be implemented to provide a 3.3V source.
The LTC3642 was chosen for several reasons. The first and most compelling reason being the chip has flight heritage with MSU's Space Science and Engineering Laboratory (SSEL). Having successfully flown on multiple satellite missions with no issues speaks to the reliability of the device, especially in the harsh environment of space. A 3.3V regulator is needed to power the PIC microcontroller on the sensor board, and as 3.3V is quite common, this will also be used to power all other devices on the sensor board. The LTC3642 has an input voltage ranging from 4.5V to 45V and a max output current of 50mA, both of these parameters are well respected in the design.
Taking into consideration the current draws of both the amplifier and PIC, and accounting for the power efficiency of the regulator, the total power draw can be calculated. At max power draw the system is not expected to exceed even half of the power budget, only using 49 mW of the allotted 100 mW. The max current draw expected is about 14 mA, well below the LTC3642 maximum current output of 50 mA.
This regulator boasts high efficiencies, which is key due to the sensor board's limited power budget of 100 mW. With an input voltage of 8.7V and an expected current draw of 14mA from the sensor board, a power efficiency of about 93% is expected from the LTC3642.
This is our first prototype for strike detection and digitization; it uses a comparator made from a LM358 op-amp.
The transient pulse caused by incoming radiation is of short duration and low magnitude. In order to reliably detect radiation strikes, this output needs to be conditioned and amplified. The first part of this chain is ac-coupling so that only the transient part of the pulse is passed to the amplifier. This cuts out DC components from light continuously striking the cell. The next step is to amplify the pulse to a more readable voltage; because of slew rate limitations this may require several stages. Some pulse stretching will probably occur during amplification. There are pulse stretching ICs that should be able to stretch the signal more if necessary. Once the pulse is of appropriate length and magnitude it needs to be digitized so that it can be read in by the FPGA. The project requirements only require that the strikes are detected so a simple comparator can be used to digitize the pulse into one bit.
In the prototype on the right, a LM358 operational amplifier is being used as a comparator with a high-pass filter preceding it. The laser pulses that are stimulating this setup are on the order of ms so additional amplification is not necessary. When a faster laser apparatus is used we will have to modify this design.
PIC Critical Subroutines
The function of the computer subsystem is data collection and distribution. When the digitized output of the amplifier subsystem is high, an interrupt service routine is triggered to increment a running count of radiation strikes. Additionally, when the avionics sends a data request, another interrupt routine loads the strike count into a First-In-First-Out (FIFO) buffer before sending the count via UART to the avionics stack to include in the telemetry packet. The figure to the right is a high level depiction of the two major routines needed to count radiation strikes and prepare them to be sent to the avionics stack.