Hermes CubeSat: Testing the Viability of High Speed Communications on a Picosatellite

Hermes CubeSat: Testing the Viability of High Speed Communications on a Picosatellite Dustin Martin, Riley Pack, Greg Stahl, Jared Russell Colorado Space Grant Consortium dustin.martin@colorado.edu March 29 th, 2009 Abstract Begun nearly three years ago, the Hermes CubeSat project is nearing completion and is currently in the final testing and integration phase. Hermes was designed using the CubeSat form factor, which includes very strenuous size and mass requirements. As a result, the design team faced several challenges when designing the various subsystems to meet these requirements, as well as to satisfy the overall goals of the mission. This paper describes the purpose of the Hermes CubeSat mission, as well as the resulting satellite design that has been developed and assembled. 1. Introduction The Hermes CubeSat Project was started with three primary objectives. In developing and operating the Hermes satellite on orbit, the team will demonstrate that the CubeSat bus that has been developed is operational and that high data rate communications is possible on a small satellite platform. The secondary objective of the Hermes mission is to acquire sensor data that will allow for the characterization of the orbital environment that CubeSats are subjected to during their lifetimes. Additionally, each system of the satellite is to be developed in-house at the Colorado Space Grant Consortium at the University of Colorado by a team composed entirely of students. developing a satellite that meets the requirements set forth by the CubeSat program of the California Polytechnic State University (CalPoly). The requirements that define a CubeSat by the standards set forth by CalPoly include restrictions on size and mass, as well as the external configuration of the satellite. Generally, the satellite must be a ten centimeter cube with a mass under one kilogram. This extreme size constraint requires that each of the subsystems be designed to be extremely compact and lightweight, while also incorporating the functionality required to complete the mission. 2. Hermes Satellite Subsystems The Hermes satellite consists of several different subsystems which together form a complete picosatellite capable of completing the required tasks. The subsystems that make up the generic CubeSat bus that is being developed include the Command and Data Handling, Power, and Primary Communications systems, as well as the satellite structure. These systems were designed with extensibility in mind, as this bus is designed to be the platform on which future CubeSats from the Colorado Space Grant Consortium will be developed. Figure 1. Hermes Project Logo The challenge since the beginning of development has been to achieve each of the goals of the mission while Figure 2. Hermes System Layout The primary payload of Hermes is the High-Speed Communications system, which was designed to satisfy one of the objectives of the mission, to verify that large 2009 COSGC Space Research Symposium Page 1

data throughput communications are possible on a CubeSat. In order to achieve a high quality link to establish high speed communications, an Attitude Determination and Control system was also designed. Attitude control was achieved through the incorporation of a passive attitude control system to allow for adequate pointing accuracy. Together these six subsystems form the Hermes CubeSat. In the following sections, each of these subsystems is described. 2.1. Command and Data Handling The Command and Data Handling (CDH) system for Hermes CubeSat was designed to meet three major criteria: reliability, resilience to external stimuli, and ease of use from the ground. As the brain of the CubeSat, it is very important that the CDH hardware and software be reliable. To accomplish this, a commercial real-time operating system (RTOS) was used as the backbone of the system. The Salvo RTOS by Pumpkin is a lightweight system designed for use on small embedded systems. Because the PIC24H, the chosen microarchitecture, has hardware limitations that make it difficult to use an RTOS, Salvo was a perfect fit for the CDH hardware. It allows for a cooperative multitasking environment, which greatly simplifies the software organization. Each major subsystem (thermal, power, commands, sensors, and communications) has one or more tasks to handle specific functionality for the system, which allows the system to be modularized and easily debugged. Furthermore, to ensure that critical data, such as flight code, is stored safely on the SD cards, an Error Detection and Correction (EDAC) algorithm was implemented. This algorithm duplicates critical data across all three SD cards and encodes it with several EDAC schemes, including multiple Hamming codes. As mentioned above, a majority of the data stored in the EDAC section of the SD cards is flight code. The CDH system is capable of reprogramming both itself and the primary communications system via a tiered bootstrap loader approach. The CDH code is broken into three sections: primary bootloader, secondary bootloader, and main flight code. The primary bootloader checks the secondary and main flight code stored on the internal flash of the PIC24 and reprograms it if any discrepancies are found between the current code and that stored in the EDAC section of the SD cards. Once it has finished, the primary bootloader jumps to the secondary bootloader, which checks the primary bootloader for problems and reprograms it as necessary. Finally, the secondary bootloader jumps to the main flight code, which checks the PCOM AVR against the EDAC code and then continues through its startup routine. This system allows the system to withstand errors caused by radiation in the flight code. The final design goal for the CDH system is a simple interface to the ground software. While the flight software has a large range of functionality, it is important that an operator on the ground can easily access the information they need to assess the state of the satellite. In order to do so, flight software exports several commands that can be sent from the ground to either obtain information, like the current satellite time or a health and status report, or to set parameters, like sensor polling rates. By reducing the exposed interface to specific commands that are simple in nature, the ground interface has been made relatively simple, especially with the ground software based around the InControl system developed by L-3 Communications. 2.2. Power System Figure 3. CDH Board The major feature of the CDH system is its resilience to external stimuli, most notably radiation. While the system is composed entirely of commercial off-the-shelf (COTS) parts, special attention was given to the problem of radiation on-orbit. Surrounding the PIC24 microprocessor are three SD cards for mass storage and three real-time clocks (RTCs). By adding three of each of these vital components, the system is capable of surviving a single failure to each part by a majority vote. The CubeSat's Electrical Power System (EPS) was a significant challenge when creating the satellite. The entire system was designed, built, and tested by students. Even the circuit boards were laid out and routed by students. Most of those students were undergraduates. It was designed with goals of minimization of size, mass, and cost while using entirely COTS parts, and was also made to be modular and extensible. It is currently fully functional. It uses DC-DC buck and boost converters to regulate input and output levels, which are redundant wherever possible. It also interfaces with two separate battery packs it can autonomously charge and discharge, 2009 COSGC Space Research Symposium Page 2

and can turn subsystems on or off by communicating over SMBUS with the CDH system. In its current state the power system could easily be extended to a different CubeSat with different power requirements by doing little more than recalculating some component values to change the levels on the DC-DC converters. A Yaesu VX-3R was chosen for the Hermes mission to modulate and demodulate the transmitted signals for its small size and low power consumption. To meet mass and outgassing restrictions, the casing screen, speaker, microphone and antenna were removed. A monopole antenna constructed with spring steel that was cut to a quarter wavelength of the 437.425 MHz frequency that is being used was connected to the radio in place of the commercial antenna. A simple 1/8 inch mono connector was used to communicate between the VX-3R and the TNC. Figure 4. Power System Board Several challenges arose in the development of the system. One of the earliest challenges faced was the design of the system while meeting the often-conflicting design goals. It needed to be kept small and low-cost while taking into consideration evolving mission requirements and those of the future, all while using COTS parts. This was overcome through research and trade studies. The current configuration was selected largely for extensibility, but also for relative simplicity and the sacrifice of some of that extensibility. When assembling and testing the system, human error and inexperience also created several challenges. Small design changes could create adverse effects elsewhere in the system, due in large part to lack of knowledge of the student making the changes. As has been mentioned, most of the students working on the system were undergraduates, and for an extended period the students with the largest roles were not even electrical engineering majors. For this same reason, troubleshooting was often a difficult process. These obstacles were overcome by using all the resources the University of Colorado and Space Grant made available, from faculty and textbooks to other more experienced students. 2.3. Primary Communications System The primary communications system for Hermes was designed and built almost entirely by the members of the Hermes team. The system is separated into two major parts: the radio, and the terminal node controller (TNC). The TNC was designed and built by the Hermes team, and then integrated with the commercial radio purchased from Yaesu. Figure 5. Primary Communications Board The TNC fulfills the function of providing error free communication between the satellite and the ground station. It consists of three main sections, the Atmel ATMega 168 microcontroller, the MiXed Signal MX604 digital to analog and analog to digital converter, and a simple filtering circuit. The filter circuitry removes received signal frequencies that are outside of the 1200-2400 Hz range. This is done because the signals that are desired to be received are 1200 Hz and 2400 Hz. The MX604 converts the 1200 Hz or 2400 Hz signals to a binary value that the ATMega 168 can then interpret digitally, and also produces a 1200 Hz and 2400 Hz based on a digital input from the microcontroller. To receive data, the 168 samples the digital output of the MX604 at a rate of 1200 Hz and uses each sample as one bit of information. To transmit data, the 168 places the VX-3R into transmit mode using the 1/8 inch mono connector interface and places one bit at a time on the MX604 digital input at a rate of 1200Hz. All data, both received and transmitted is encoded using restricted AX.25 amateur radio data protocol. As a result of the resource limitations of the ATMega 168, only one packet can be buffered for transmission along with one received packet, instead of the specified eight packets in either direction. Communication between the TNC and command and data handling system of the satellite is achieved using a serial peripheral interface (SPI) protocol. Received data is passed to CDH when it is available and requested by CDH. Data is transmitted to the ground station 2009 COSGC Space Research Symposium Page 3

immediately after CDH passes it to the TNC and a link with the ground station has been established. 2.4. High Speed Communications System The High Speed Communications system is the primary payload of the Hermes CubeSat. While the primary communications system serves as the standard communications link, the high speed communications system is an experimental system designed to demonstrate that a very high data rate communications link is possible using a CubeSat. This system, once proven on orbit, would be very valuable to future CubeSats from the Colorado Space Grant Consortium as it would be able to replace the UHF system as the primary communications system. The high speed communications system was designed around the Microhard MHX-2400 S-band modem, which will allow the satellite to communicate at 115,200 baud at ideal conditions, which is significantly larger than the 1200 baud that is possible with the current primary communications system. However, the system utilizes a forward error correction scheme which will allow the link to be closed more reliably while reducing the baud rate in half. Therefore, the high speed communications system operates at 57,600 baud under ideal conditions, which is still significantly faster than the UHF system. This allows future satellites to incorporate the high speed communications system as well as a science payload that requires a high data rate downlink, such as a camera or high speed data acquisition hardware. Thus, in proving the feasibility of the high speed communications system, a larger range of science objectives can be pursued at the Colorado Space Grant Consortium using the CubeSat platform. Figure 6. High Speed Communications Board Due to a lack of availability of the MHX-2400 modem originally selected, the modem that will be flown is the newer MHX-2420 modem loaded with the firmware from the MHX-2400, so it will be identical to the 2400 in all aspects. The MHX-2400 modem is a spread spectrum frequency hopping modem that transmits in the 2.4 to 2.48 GHz range, which lies in the S-band. During operations, the modem on board Hermes will link to an identical MHX-2400 connected to the S-band ground station used by the Colorado Space Grant Consortium. The MHX- 2400 is very well suited to this implementation, as it is capable of performing forward error correction and Doppler shifting internally. Additionally, the modems will handshake automatically, and all the details of the link are transparent to the mission operators. The antenna for the high speed communications system is also a monopole antenna constructed with spring steel that has been tuned to the frequency range that the MHX- 2400 utilizes. The wavelength of the signal generated by this antenna is roughly 12.5 cm at 2.4GHz, which is very close in magnitude to the external dimensions of Hermes, which is roughly 10 cm. Because the satellite structure dimension and the signal wavelength are roughly the same size, the structure of the satellite will effectively shield the antenna s radiation that is directed back towards the satellite. As a result, the high speed communications system is very directional, and imposes a pointing requirement in order to secure a link. 2.5. Attitude Determination and Control System The Attitude Determination and Control System (ADCS) is designed to point the high speed communications antenna in the ideal orientation to achieve the highest quality link. This is necessary because of the high speed communication system's gain pattern, which has an optimal gain when the antenna is oriented 30 degrees from the line of sight between the satellite and the ground station. In order to achieve this level of attitude control, it was determined that a passive system was ideal, as an active control system was found to be too volume-and-mass consuming to be worth the additional pointing accuracy. The design of the passive control system incorporates magnets and hysteresis rods in order to stabilize the satellite in the desired orientation. The magnet is aligned with the antenna axis of the satellite which will align that axis with the Earth s magnetic field as the satellite orbits the Earth. The magnet and hysteresis rod sizing was performed using an attitude model that was created by Auburn University for the Aubie-Sat I project. The result was a 3 inch long AlNiCo Cast-5 magnet oriented along the antenna axis and two one inch long HYMU80 hysteresis rods oriented in the other two orthogonal axes. It was determined that this magnet design would provide the required pointing accuracy and the hysteresis rods would adequately dampen out the rotation of the satellite. 2009 COSGC Space Research Symposium Page 4

2.6. Satellite Structure The structure of the Hermes Cubesat was designed to meet the requirements set forth by CalPoly as well as to adequately support and protect the satellite subsystems. The resulting design in an isogrid satellite structure with four corner rails designed to support Hermes in the Poly Picosatellite Orbital Deployer (PPOD). Each of the subsystems are integrated onto square PCB boards which are then integrated together using threaded rods which pass through each board. The spacing between each subsystem board is fixed using standoffs between the boards that are positioned on each of the threaded rods between the boards. In order to meet the requirements set forth by CalPoly, the antennas have to be restrained such that they will not contact the PPOD when Hermes is inside of the deployer. In order to satisfy this requirement, the antennas were to be wrapped around the outside of the satellite so that they would not extend beyond the range required to fit in the PPOD. To hold the antennas in the wrapped position, Dacron thread was tied around the antennas and an element of the aluminum structure through holes in the solar panels. This restrains the antennas securely enough that they will be able to remain in place during integration with the PPOD and during launch. Figure 9. Hermes with Antennas Stowed Figure 7. The Board Stack In order to integrate the flight electronic hardware into the flight structure, the internal board stack is first integrated, and then the structural panels affixed to the sides of the stack. Once the external structure has been assembled, the six solar panels are attached to each face of Hermes using solar panel clips that were designed and manufactured in-house. The antennas for each of the two communications systems are passed through holes in the aluminum structure and through the solar panels such that the entire antenna length is outside of the satellite. Figure 8. Hermes with Antennas To deploy both the High Speed communications and Primary Communications antennas, the Dacron restraints are burned through using Nichrome burn wire. A current is passed through sections of Nichrome wire, which has a very low resistance and will generate a large enough amount of heat to cut through the Dacron restraints. This deployment system is activated after deployment from the PPOD and a pre-determined delay period. 3. Current Status of the Project As mentioned previously, the satellite is in the final integration and testing phase of development. All of the flight hardware is in-house and assembled, and each of the subsystems has been exhaustively tested individually. Currently testing of the fully integrated board stack is being performed, which will be followed by full structural integration of the satellite. Several long range communications tests have been performed with both of the communications systems, and the results are very positive. Deployment tests have also been completed and show that the deployment system that was designed is very reliable. However, additional communication and deployment tests will take place in the near future to show conclusively that the satellite is fully operational in those respects. Following the additional operational testing and structural integration, a series of environmental tests will be completed in order to show that the satellite will survive launch. The two tests that will be performed are a 2009 COSGC Space Research Symposium Page 5

vibration test and a thermal vacuum test which will be administered at test conditions required by the launch provider. Currently the project is on schedule for completion in late April, 2009, and all indications are that the satellite will be fully capable of satisfying each of the mission goals when launched. 2009 COSGC Space Research Symposium Page 6