- Email: [email protected]
Munin: A Student Nanosatellite for Space Weather Information
MUNIN: A STUDENT NANOSATELLITE FOR SPACE WEATHER INFORMATION Olle Norberg\ W. Puccio\ J. 01sen\ S. Barabash\ L. Andersson\ J. D. Winningham^ U. Jonsson^,and M. Eriksson^
^Swedish Institute of Space Physics, Box 812, SE'981 28 Kiruna, Sweden ^Southwest Research Institute, Drawer 28510, San Antonio, Texas 78284, USA ^Lulea University of Technology, SE-971 87 Lulea, Sweden
ABSTRACT The Munin satellite (Figure 1) is set to become the first of a new type of monitoring spacecraft. Using modem technology, this very small satellite (mass « 5 kg) will have all the necessary ftmctions needed to support its specific scientific mission: monitoring of the auroral activity on both the northern and southern hemispheres. With Munin we can now usher in a fleet of monitor spacecraft to cost-effectively provide global monitoring, and at the same time have a large degree of student involvement on different levels. Figure 1. The Munin nanosatellite. Munin has three scientific instruments for monitoring of auroral activity: • MEDUSA, a combined electron and ion spectrometer with continuos coverage of all pitch angles. Covers the energy range 10 eV -18 keV. • DINA, measures high energy ions and neutral particles at pitch angles 0° and 90°. Covers the energy range 30- 1200 keV. • HiSCC, a high sensitivity CCD camera for visible and infrared wavelengths. The camera has a field-ofview of 50°, and a resolution of 340 x 240 pixels. The payload has a total mass of 1.6 kg, and consumes 2.4 W, if power permits it will be operated continuously. The satellite is cubic with solar cells on all six sides. The solar cells provides Munin with a power of 6.0 W. The average power consumption is 4.7 W, peak consumption (when the transmitter is on) is 9.1 W. A
— 352 —
Li-Ion battery will provide the peak power and also keep the satellite powered during eclipses. The structure is made from aluminum and has a mass of 1.9 kg. The total mass of the Munin spacecraft, including the separation system, is 6.0 kg. A separation system holding the spacecraft secured to the launcher using three hooks has been designed. Munin conmiunicates on the UHF band, the up-link frequency is 449.95 MHz and down-link frequency is 400.55 MHz. The down-link is designed for a bit rate of either 9600 or 19200 bps. The modem is implemented in software in a digital signal processor (DSP), which also handles command handling and telemetry formatting. Another DSP takes care of instrument operation and data compression. The attitude will be controlled by a magnet aligning Munin with the local magnetic field. Soft-magnetic hysteresis rods will dissipate oscillation energy. Munin will use an existing ground station located at the Swedish Institute of Space Physics, Kiruna, Sweden. The ground station was buih in 1994 and has been successfiilly used for the two Swedish satellites Freja and Astrid. PROJECT OBJECTIVES Auroral Research and Forecasting The Munin satellite is set to become the first of a new type of monitoring spacecraft. Using modem technology, this very small satellite (mass « 5 kg) will have all the necessary functions needed to support a specific scientific mission such as the one we will describe in this document. With Munin we can usher in a fleet of monitor spacecraft to cost-effectively provide global monitoring, and at the same time have a large degree of student involvement on different levels. Today auroral research is a wide field covering topics such as the acceleration mechanisms of precipitating particles, the physics behind the triggering of substorms, and forecasting of magnetic storms to mitigate the damage that "space weather" can cause in e.g. power distribution lines. These research areas all need data about the conditions in space. Both the solar wind conditions and the resulting disturbances on the Earth such as magnetic and electric field fluctuations, particle populations, and auroral activity are needed in order to deduce the physical mechanisms behind these phenomena. The northern and southern auroral ovals are always present, day and night. From the Earth they can be seen by the naked eye at latitudes between 60 and 80 degrees north and south. The aurora is the visible manifestation of the particle precipitation along the Earth s magnetic field lines at high latitudes. Since the aurora can be viewed only during relatively cloud-free nights, it is not an optimal source of information about the result of magnetospheric activity. The best tool to gather this information was (very early in the history of space flight) found to be satellites in polar orbits. There have been a number of satellites of this type over the years, with new discoveries about auroral physics made by every one. Up until now, the data from such satellites have first been analyzed by the Principal Investigator of the instrument in question, then the results of the investigation have been published in scientific papers. With Munin, we propose to use a new paradigm of data dissemination, which we think will enhance the output from auroral research. We will make the data from the instruments on Munin available in real time to everyone interested, by using the Intemet and WWW-based services. Munin will provide measurements of the electron and ion particle distributions above the auroral ovals, the fluxes of energetic particles (ions and neutral particles), as well as images of the aurora taken by a CCD camera in visible wavelengths. As the data comes down to a ground station located in Kiruna, Sweden, the data will be processed such that it can be used by scientists all over the worid for auroral research. The satellite will store data from the pass
— 353
over the southern auroral oval (over Antarctica) in its memory which will then be downloaded when it is over Kiruna, whereas the datafromthe northern oval will be down-linked in near real time. The fact that the data is "fresh" will enhance the possibilities of using it for forecasting of auroral activity. The shape of the ovals, and the offset of their centers relative to the geographic poles, means, for example, that it is possible to use measurements over Scandinavia to predict the auroral activity in north America or Siberia at a later time. We intend to use the Munin measurements to present the present status of the auroral ovals, and let users of the data interpolate the position and size of the auroral oval a few hours into the future, in order to forecast auroral activity. Public and Educational Outreach The Munin team has striven to implement this project in order to promote the public interest in science, technology, space research, and international cooperation. During the last decades the public interest in these questions has declined, and we are of the opinion that a small, focused, project like Munin can catch the interest of the public. We intend to pursue this objective by visibility in newspapers, magazines, and by the Munin WWW-server, http://muiiin.irf.se. The complexity of today's space projects makes it extremely difficult for engineering students to grasp how the complete system works, or how it was designed. This results in more and more engineers becoming experts in just a small part of a project, with very limited insight in the overall system functions. With Munin, we intend to give students a possibility to be not only engaged in the mission, but also to gain a genuine knowledge about all parts of a space mission. The Swedish Institute of Space Physics is engaged in two undergraduate space engineering programs, a 3year program given by Umea University, and a 4.5 year program given by Lulea University of Technology. Students from these programs are already contributing to the project by doing their degree projects on parts of the satellite. The intention is also to have students involved in the operation of the satellite, for example in the form of lab-courses involving hands-on satellite operations. Orbit Considerations In order to fulfill the scientific objectives of the Munin project, the satellite should be injected into a polar orbit with an inclination of at least 63 degrees and an altitude between 400 and 2000 km. A circular orbit at approximately 1000 - 1400 km and an orbit plane at around noon-midnight local time is optimal, since this orbit takes Munin both through the active night-side of the auroral ovals, and the polar cusps. PAYLOAD DESCRIPTION Miniaturized Electrostatic DUal-tophat Spherical Analyzer fMEDUSA^ The primary instrument on Munin is MEDUSA (Miniaturized Electrostatic DUal-tophat Spherical Analyzer), a combined electron and ion spectrometer. The instrument sensor is provided by the Southwest Research Institute, San Antonio, Texas. Electrons and ions with energies up to 18 keV/q will be measured simultaneously, with a maximum time resolution of 16 energy sweeps per second for electrons, and 8 seconds for ions. Particles are measured in 16 sectors in the plane of acceptance, which is aligned with the Earth's magnetic field. See Table 1 for a sunmiary of the MEDUSA instrument characteristics.
354 —
Electrons are responsible for the major part of the energy input creating the aurora, and therefore the most important species to measure. The precipitating electrons energize (heat) ionospheric ions in the auroral region, these heated ions are then accelerated along the magnetic field lines and leave the heating region. This creates an outflow of atmospheric ions, primarily hydrogen and oxygen. With Munin we intend to measure both the electrons responsible for the aurora and ion heating, and the heated and accelerated ions. During certain circumstances, precipitation of ions also occur in the auroral region. The MEDUSA sensor will be mounted on Munin in such a way that each of the 16 sectors in the acceptance plane always looks at particles arriving with a certain angle to the magnetic field lines, the "pitch angle". This is possible due to the simple, yet very efficient, attitude control chosen for Munin. A permanent magnet will align one of Munin's axes along the local magnetic field line, this axis will be in the aperture plane. Particles enter the spectrometer aperture at any angle in the 10° wide plane of incidence (see Figure 2), electrons and ions are then deflected into their respective spectrometer unit by a spherical electrostatic analyzer (deflection plates). The particles hit a microchannel plate (MCP) after being filtered in energy in the electrostatic analyzer, the hits are counted by preamplifiers connected to registers, and the number of hits per sample interval are thenfiirtherprocessed by the data processing unit (DPU). The MEDUSA sensor represents a completely new development in terms of highly compact sensor design. MEDUSA will be used on the Swedish microsatellite Astrid-2.
Table 1. The MEDUSA Instrument Characteristics Instrument Characteristics 0.60 kg Sensor mass LOW Sensor power 10eV/q-18keV/q Energy range (for both electrons and ions) 25% Energy resolution (AE/E) 360°x 10° Acceptance angles 4 x lO'^cm^sr Geometric factor per sector
-0113mm AMPS
HT^.JII^
55y
ION SIDE
ELECTRON SIDE
i.
4X SUPPORT VANES
-^
':TX • •
HIGH VOLTAGE POWER SUPPLY FOR MCP & DEFLECTION PLATE
AMPS
-|^li!
Figure 2. Cross Section of the MEDUSA Sensor Unit.
— 355 —
Detector of Ions and Neutral Atoms (DINA) The meiin scientific objective of the instrument is to extend the measurements of the precipitating and mirroring ions to the higher energy range 30 - 1200 keV, complementing the MEDUSA experiment. Apart from the main scientific objective, the DINA instrument will also be able to measure energetic neutral atoms (ENAs). ENAs are produced via the exchange of charge between singly charged magnetospheric ions and atoms of the upper atmosphere/exosphere. ENAs can be found in almost any space environment and are suitable agents to remotely probe ion populations of plasmas at a distance, socalled ENA imaging. In magnetospheric physics, general attention has so far concentrated on the ENAs generated within the high altitude ring current, because they can be used in global magnetospheric imaging fi*om high altitude (> 20000 km ) spacecraft (Williams et ai, 1992). However, in the auroral region where the ring current / radiation belt particles plunge into the dense upper atmosphere/exosphere, the charge exchange process is much more eflFective and ENAs emissions are much more intense. As was shown theoretically by Roelof(\991\ ENAs fi'om low altitudes are emitted fi-om a very thin area near the exobase and the ENA generation region is, essentially, two-dimensional. The ENA camera PIPPI onboard the microsatellite Astrid (Norberg, 1995) has performed ENA measurements from a 1000 km polar orbit and demonstrated the potential of low altitude ENA imaging (Barabash et al, 1997; Brandt et al., 1997). DINA will continue studies of the low-altitude ENAs concentrating on measurements close to the generation region. The results fi-om DINA, looking at the ring current fi'om below at low latitudes and close to the ring current at high latitudes, will undoubtedly help the interpretation of the observations to be performed by IMAGE, the NASA mission to image the ring current from above. Another interesting topic to be addressed is ENA albedo. The O"^ ions precipitating onto the upper atmosphere during major geomagnetic storms transform their energy and momentum into atmospheric heating and result in escaping fluxes of fast neutral atoms via charge exchange and elastic scattering (Ishimoto et al, 1992). The total neutral flux escaping due to these processes could be significant for the atmospheric evolution during the life-time of the Earth (Torr et ai, 1974). However, no direct measurements have been performed to prove the correctness of the developed models. The DINA simultaneous measurements of the precipitating ions and ENA flux would provide the necessary inputs to evaluate the existing models of mass and energy transport in the ring current-atmosphere interaction. DINA will also investigate the low-altitude equatorial ring current formed by stripped ENAs from the main radiation belt (see Voss et al, 1995, and references therein) by measuring both the trapped ions and parent ENAs. During geomagnetic storms the ENA flux produced in the low-altitude ring current could be sufficient for detection, although for quiet conditions the flux will not be sufficient {Bishop, 1996). In summary, the scientific objectives of DINA are: 1. Measurements of the ion flux in the energy range 30 -1200 keV. 2. Measurements of the ENA flux from the exobase in the energy range 30 - 400 keV. 3. Measurements of the outflowing ENA flux (ENA albedo) from the precipitation region in the energy range 20 - 280 keV. 5. Mass resolving measurement of particles in the energy range 100 -1200 keV. 6. Studies of the low-altitude ring current. The instrument consists of two sensors with an aperture opening of 10° x 38° each (see Table 2). The sensor DINA-0 will detect particles with 0 degree pitch-angle, and DINA-90 will detect particles with 90 degree pitch-angle. Over the northern hemisphere DINA-0 provides measurements of the precipitating — 356 —
ions. While the spacecraft moves over the polar cap, DINA-90 will make measurements of ENA flux from the exobase in one local time sector. Over the southern hemisphere DINA-90 will measure the exobase ENA flux from a different local time sector due to the expected very slow spin of the spacecraft along the magnetically aligned axis. DINA-0 will be pointing toward the Earth over the southern hemisphere, and is aimed to detect outflowing ENA in the precipitation region. Information about the input ion flux can be obtained from the DINA-90 measurements because the energetic ion distributions can be considered approximately isotropic (Lyons, 1987). The instrument performs alternative measurements of ions and energetic atoms by turning on and off the high voltage on the electrostatic deflection system. Electrons with energies below 400 keV are always swept away by permanently magnetized broom magnets (the simulated transmittance for 400 keV electrons is 14%). Mass identification is performed by AE/E detectors. For the energy range 3 0 - 1 0 0 keV, the front detector (2^m Si detector) will provide the integral flux of all masses. For energies above 100 keV, coincidence and anti-coincidence logic provide measurements of hydrogen and helium (coincidence between the front and back detectors) and particles with A > 4, mainly oxygen (anti-coincidence between the front and back detectors).
Table 2. The DIN A Instrument Characteristics Instrument Characteristics Mass 0.9 kg Power 0.5 W Energy range (protons or H atoms) 30 -1200 keV Masses to resolve A = 1, 4, A > 4 for E > 100 keV Particles to measure ions, neutrals Aperture per sensor 10' x 38° 3.1 x 10'^ cm^ sr Geometric factor (per detector) Deflector cut-off energy 400 keV Broom magnets transmittance « 2% at for 100 keV electrons « 14% for 400 keV electrons
High Sensitivitv CCD Camera (HiSCC^ One of the best global space weather measurement tools is a camera imaging the auroral oval. A camera onboard a satellite supports both onboard instruments and ground facilities. Over the northern hemisphere HiSCC will take images of the night side auroral oval. This will give information about the activity level of the magnetosphere. HiSCC will support MEDUSA with identification of the spacecraft's footprint in the auroral oval, and any intensifications in the oval. Over the southern hemisphere HiSCC will operate as a star imager, enabling attitude determination based on star positions in each image. Attitude knowledge is crucial for interpretation of the measurements made by the MEDUSA and DIN A instruments, as well as a major interest for the satellite operation group since the spacecraft behavior in space can be studied. This means that HiSCC can be used to verify the performance of the spacecraft attitude control system. The camera will be the main instrument to provide the mission with pictures to the public. HiSCC data will help to educate the public about what space weather is, and the easy access to the data on the Web will increase the public interest in understanding the science of space plasmas in general and specifically the aurora.
357
The camera field of view is 50° with a highest spatial resolution of 3 x 3 km (nadir looking from an altitude of 1000 km). When the satellite is at 60° latitude the ground coverage ranges from 65° latitude and over the pole. At 75° the camera is almost looking in the nadir direction, covering 10° in both latitude and longitude. In order to avoid blurry pictures the exposure time will be below one second. Approximately 25 pictures will be taken during each pass over the northern auroral oval. The data will be compressed by the DPU before being down-linked to the ground station. The CCD camera is a modified Quickcam from Connectix Corporation. It is made for connecting to a PC parallel port; for Munin this has been modified to a special interface to the DPU. A new housing and lens system assembly has been designed. We have removed the built-in infrared filter to increase the light sensitivity. The CCD chip used in the camera is a Texas TC-255 with a quantum efficiency of up to 60% in the wavelength range 500-800 nm. It can produce pictures with a resolution of 320x 240 pixels, in 64 shades of gray. The camera can operate with different exposure times, the shortest time being 1/1000 second. See Table 3 for a summary of the HiSCC instrument characteristics.
Table 3. The HiSCC Instrument Characteristics Mass Power Camera Optics Spectral passband Detector Resolution Quantum Efficiency Exposure time
Instrument Characteristics 0.10 kg 0.3 W focusing lens, f 1.9, field-of-view 50° 450 - 850 nm TC-255 CCD chip, 50,000 e" full well capacity 0.2°x 0.2°, 320 X 240 pixels/3.2 X 2.4 mm 20 - 60 % minimum 10 ms, operationally 0.5 - 1 s
SATELLITE SYSTEM DESCRIPTION Satellite Structure The combined electron and ion spectrometer MEDUSA has a requirement of a 360° field-ofview in a plane, the satellite was thus designed as a natural extension of the MEDUSA instrument. The satellite can be considered to be an experiment with its own power and telemetry resources. The use of magnetic attitude stabilization means that Munin will rotate twice per orbit. All faces of the satellite will thus need to be covered with solar cells to maximize the energy input. To satisfy the power requirement of approximately 6 W it was decided to design Munin as a cube with the sides 210 mm long. The inner structure consists of the bottom platform, four support struts, the battery Figure 3. The interior of Munin as seen from the side of the satellite which will face towards the platform, the MEDUSA platform, and the top Earth over the northern pole (the camera platform (Figure 3). The four side panels are side). 358
screwed together mto a shell which can be integrated on Munin after all the units inside the satellite have been assembled and tested. There are aperture openings in some of these panels for the HiSCC and DINA instruments, as well as a mounting hole for the radio antenna. The structure is made of anodized alummum. Separation Syf^t^m
The low mass and the small size of the Munin satellite gives us the opportunity to design a very simple yet reliable separation system, see Figure 4. Munin will be clamped down to the launcher interface plate at three points. The satellite is equipped with three steel wires fitted with cylindrical end blocks, these blocks are held down by grappling hooks. Clamping will be provided for by a spring-tensioned Kevlar line. Three helical springs will be used for pushing Munin away from the launcher with a speed of about 0.6 m/s. A small launcher activated pyro-guillotine will be Figure 4. The Munin Separation System. With used to initiate the separation by cutting the compressed springs the total height of Kevlar line. The mass of the separation system is system is 40 mm. 350 grams. The separation mechanism will be attached to the launcher with three bolts. The separation system will include a separation switch, to indicate the separation of Munin in the launcher telemetry. Inside Munin there is also a separation switch which is in the off position when Munin is located on the separation system. At separation this switch will power-on the satellite. Power Sy>stem Each side of the satellite is covered with solar cells. We are using standard monocrystalline silicon cells, with an efficiency of 15%. These cells are cut to a size of 20 x 40 mm and are glued directly onto the satellite body with a 0.5 mm cover glass on top of each cell. Each cell interconnection is done with two leads for safety reasons. Each side of Munin will have one string with 40 cells serially connected, which will give about 6W power, and 15-20V. This power will be sufficient to run the payload instalments continuously, and the radio transmitter for 10-15 minutes per orbit. See Table 4 for the power budget. Munin will use a Lithium-Ion battery. These batteries can store more energy per unit mass and volume than Nickel-Cadmium, Nickel-Metal-Hydride or any other commonly used battery types. Lithium-Ion batteries do not have any memory effect like Nickel-Cadmium, and a very low self-discharge rate. The battery used will have three cells stacked in series to achieve a maximum bus voltage of 12.6 Volt and a nominal voltage of 10.8 Volt, v^th a capacity of 4050 mAh resulting in an energy density of 44 Wh with a mass of only 560g (including mounting parts). The depth of discharge (DOD) will typically be around 9%, this will ensure the battery survives the huge number of charging and discharging cycles it will undergo during the life-time of the satellite. The practical charging of a Lithium-Ion battery is separated into two phases, first the battery should be charged with a constant current and when the voltage over the cell reaches a certain point, constant voltage charging should be used instead. The temperature range for discharging the batteries should be kept between -10 and +50 degrees Celsius and charging between 0 and +45 degrees Celsius. A computer simulation of the thermal control, as well as 359
a solar simulation test, was performed as a M.Sc. thesis work during 1997. This work found that the battery needs a 1. 2 W heater mounted inside the battery pack to keep the battery above the minimum level.
Table 4. Power Budget Consumer MEDUSA (ion & electron analyzer) DINA (high energy particle detector) with electronics HiSCC (CCD camera) DPU (data processing unit) Charging electronics and DC/DC converters Radio transceiver (transmitting / receive only) Modem TOTAL (transmitting / receive only)
Power (mW) 1000 500 300 600 50 6500 /180 200 9150 / 2830
Radio Transceiver and Modem The conununication will be done with a modified commercial Tekk-1000 radio and a dedicated 320C50 digital signal processor (DSP) acting as a modem to digitally generate and demodulate the frequency shift keying (FSK) signals carrying the data to and from the Munin satellite (see Table 5). This will enable the DSP to control the bandwidth and speed of communication with the software implemented filters and demodulation techniques. The software based modem was designed as a M.Sc. thesis work. The spacecraft will transmit on 400.550 MHz with a maximum output power of 2.2 W; this will be sufficient for a bit rate of 9600 bps based on our experience with the Freja and Astrid satellites. During favorable conditions also 19200 bps should be possible. Munin will receive on 449.950 MHz using a narrow band 600 bps FSK since only a small amount of data needs to be uploaded to the spacecraft. The up-link will mainly be used for commands to the DPU for different operating modes and schedules, including a possibility to upload patches of the existing onboard software. The modem DSP also has a 12-bit AD converter used for monitoring of satellite status, such as temperatures, battery and sun panel voltages and currents etc. There are also standard fimctions such as a real-time clock, watchdog etc. which are mostly realized in Actel PLD (programmable logic devices). The antenna used both to transmit and receive consists of a one quarter wavelength (18 cm) whip located on one of the solar panel sides, pointing along the local magnetic field line due to the magnetic attitude stabilization scheme. The transceiver will run on a regulated 8 VoU supply requiring 20 mA in receive mode (default) and 0.8 A while transmitting with an output power of 2.2 W to the antenna. Doppler shift occurring due to the satellite motion relative to the ground will be corrected for by the ground station equipment.
Table 5. Specifications for the Tekk-1000 Transceiver. Specifications for the Tekk-1000 Transceiver Receiver Frequency 449.950 MHz Transmitter Frequency 400.550 Mhz Operating Temperature -30 to +60C
360 —
Operating Voltage Receiver noise Dimensions Mass
floor
7.5 to 12.0 V -118 dBm 85 x 52 x 21 mm 200 g
Data Processing Unit The DPU (data processing unit) handles the payload functions and data compression. The DPU used on Munin is the engineering model of the DPU for the MEDUSA instrument on Astrid-2. The heart of the DPU is a Texas Instruments TMS320C50 digital signal processor, which has 20 kb of internal memory and runs at a clock frequency of 40 MHz. We have equipped it with an additional 16kb of external RAM (used as program memory), 64k word EPROM (for flight software), and 32 k word EEPROM (for various data and software patches). The PROMS are only used (and powered) during the boot process and when patching of the onboard software is needed. For onboard data storage, 2 Mb of RAM are used for instrument data compression and buffering (FIFO). The data compression software in the DPU is based on a NASA-developed algorithm (CCSDS, 1994). We have successfully used this data compression scheme on previous missions we have been involved with. Attitude Control System (ACS) To enable continuos coverage of all pitch angles for the particle instrument MEDUSA, and to ensure correct pointing of the two DINA apertures, the spacecraft needs to be aligned with one axis along the local magnetic field. This requirement is realized by the use of a passive magnetic stabilization system. Due to the fact that Munin is a small satellite with a mass less than 5 kg, the use of this particular stabilization technique is ideal. Magnets have been used many times before, e.g. in the AMSAT microsatellites. Magnets are simple and cost-, mass- and volume effective. Soft-magnetic hysteresis rods are used for dissipation of oscillation energy. Mass budget The goal has been to keep the mass around 5 kg, this will be achieved by using light weight materials (mostly aluminum) for structural parts and a high degree of integration between the payload instruments and the satellite itself By using programmable logic devices (PLD) in the electronics we will achieve highly compact and low mass systems. To handle the high degree of integration most printed circuit boards will have six layers. Table 6 gives the mass break down for the different parts of Munin. Table 6. Mass budget Part Satellite structure MEDUSA (ion & electron analyzer) DINA (high energy particle detector) with electronics HiSCC (CCD camera) DPU (data processing unit) Battery (including holder) Charging electronics and DC/DC converters Solar cells (including cover glasses) Radio transceiver — 361 —
Mass (g) 1950 588 900 100 500 560 110 300 200
Modem Antenna + cable Separation system Magnet (attitude stabilization) Soft-magnetic hysteresis rods TOTAL
120 50 500 25 100 6003
Qn-board Data Handling The telemetry consists of data packets of fixed size (512 bytes). Every packet will start with four synchronization bytes followed by four housekeeping bytes. The remainder of the packets will contain datafi-omthe experiments, keeping the "non-science" data at a low level of only about 2% of the total telemetry. Only instrument data from the latest orbit will be sent down in the telemetry, i.e. after each telemetry session the on-board memory will be cleared and new data will be collected. In order to transmit as much relevant data as possible within the allocated bit rates, the data to be transmitted has to be reduced in some way. One solution is to use thresholds for measured parameters, and trigger data recording only when the threshold is exceeded. Our solution on Munin is to use data reduction by temporal integration, combined with loss less data compression. A data compression algorithm. Universal Source Encoding for Space (CCSDS, 1994), has been implemented and tested in the DPU. This algorithm will be used for all instrument data. There is also a mode in which only uncompressed satellite housekeeping data is sent down in the telemetry. Commands up-linked from the ground station, decoded by the modem, and considered valid by the DPU are executed either immediately or at a prescribed time. If the command is "time-tagged" it will be placed in a command queue and executed when its time-tag corresponds to the satellite clock. Ground Station Munin will use an existing ground station located at the Swedish Institute of Space Physics, Kiruna, Sweden. The ground station was built in 1994 and has been successftiUy used for the two Swedish satellites Freja (for student lab courses on satellite control) and Astrid (as a secondary ground station). The hardware used in the ground station is mainly off-the-shelf equipment, except for the modem. The ground station consists of three computers, one dedicated to Doppler shift control and satellite tracking, one for the actual satellite control, and one for Web services and presentation of received data. Operations The ground station will be designed to work unsupervised. The Munin team has gathered knowledge in constructing this type of automatic ground station during the Freja and Astrid missions. Since the satellite operations and safety fimctions to a large degree are implemented in on-board autonomy, the ground station will during normal operations only receive telemetry data. If something unexpected occurs in the ground station, e.g. there is no telemetry from the satellite during a scheduled pass, the ground station will automatically alert the satellite operator on duty by the use of a mobile pager. This scheme was successfiilly used during the Astrid mission in 1995, and will also be used for the Astrid-2 mission. MEDUSA and DINA data from the passes over the northern and southern auroral region will be included in the telemetry data. HiSCC images will be mainly taken over the northern auroral region, but some — 362 —
images will be taken over the southern hemisphere (where HiSCC looks out into space) for use in attitude determination, by comparing star positions in the image with star catalogues. PROJECT MANAGEMENT The Munin project is managed by young scientists and engineers at the Swedish Institute of Space Physics. Two Swedish universities and an U.S. research establishment are also involved in the project. Public Relations The scientific purpose of the Munin satellite is to collect data on the auroral activity. Such data are vital to prediction of space weather, but the information must be made public in order to be useful outside the Munin project team. To enable quick publication, we will use the World Wide Web (WWW), where everybody can retrieve the information they need. The idea is to use real-time uploading, from the downlink-computer to Internet, so that the information will be accessible as soon as it is downloaded from the satellite. The data will be free for everybody to use, regardless of institution, company or nationality, and will be available as soon as it is received at the ground station. We are planning to make the received data available in several ways: • • •
as simple graphs and plots, in a format that illustrates the auroral activity to the layman. as highly processed images (for scientific use, education or illustration) as uncompressed scientific data, forfiirtherprocessing by the end user
Schedule and Milestones The Munin project was conceived in September 1996, and work on the project has been on-going since then despite the fact that initially there was no guarantee that the satellite would be launched. The reason for this is that we see this project as an important contribution to science and the education of space engineers, and we have always been optimistic about the possibilities of finding a piggyback launch for such a small spacecraft as Munin. Munin is scheduled for a launch during 1999 as a secondary payload on a NASA mission using a Delta-II launch vehicle. SUMMARY Munin will fimction as a test bed for possible technologies and methods to be used in the miniaturization of spacecraft and instrumentation. Within the next decade there will be a need for fleets of small, autonomous spacecraft for various purposes. One objective for such a fleet will be to resolve the spatialtemporal ambiguity in satellite measurements. Another will be to form a fleet of spacecraft for global measurements of plasma parameters, as input for global modeling. Whatever the objective is, it will be necessary to keep the recurrent costs low when producing large number of identical spacecraft, here the use of conmiercial, off-the-shelf, equipment is essential. The miniaturization of spacecraft and instruments is necessary to enable deployment of large number of satellites. One issue to address in the fixture is the possibility to de-orbit satellites after they have ceased to ftmction. Small, simple satellites are also ideal for use in educational environments, in which a Munin-class satellite can be designed, built, launched, and operated during the time of a undergraduate or graduate student career. ACKNOWLEDGMENTS
— 363 —
Munin is financed by grants from the Swedish National Space Board. The project is managed by the Swedish histitute of Space Physics in Kiruna, Sweden, where the spacecraft and the instruments DINA and HiSCC are buih. The MEDUSA instrument is provided by the Southwest Research Institute, San Antonio, Texas, with fimds from the U.S. Office of Naval Research grant NOOO14-98-1-0175. Space Engineering students at Umea University, Sweden, and students at LuleS University of Technology, Sweden, are involved in all phases of the project. The DINA instrument will be developed, and the data will be processed and analyzed, in close cooperation with our Co-Investigators Dr. K. C. Hsieh and Dr. C. Curtis of University of Arizona, Tucson, Arizon, USA, and Dr. E. C. Roelof, JHU / APL, Laurel, Maryland, USA. REFERENCES Barabash, S., P. C:son Brandt, O. Norberg, R. Lundin, E. C. Roelof, C. Chase, B. Mauk, Energetic neutral atom imaging by the Astrid microsatellite. Adv. Space Res., Vol. 20, pp. 1055-1060,1997. Bishop, J., Multiple charge exchange and ionization collisions within the ring current-geocoronaplasmasphere system: Generation of a secondary ring current on inner L shells, J. Geophys. Res., 101, pp. 17,325-17,336, 1996. CCSDS, Source Coding for Data Compression, Consultative Committee for Space Data Systems, CCSDS 11 l.O-W-2, Washington, D.C., March 1994. Brandt, P. C:son, S. Barabash, O. Norberg, R. Lundin, E.C. Roelof, C.J. Chase, B.H. Mauk, and M. Thomsen, ENA imaging from the Swedish micro satellite Astrid during the magnetic storm of 8 February, 1995, Adv. Space Res., Vol. 20, No. 4/5, pp. 1061-1066, 1997. Ishimoto, M., G. J. Romick, and C.-I. Meng, Energy distribution of energetic O"*" precipitation into the atmosphere, J. Geophys. Res., 97, pp. 8619-8629, 1992. Lyons, L. R., Processes associated with the plasma sheet boundary layer, Physica Scripta, 8, pp. 103-110, 1987. Norberg, O., S. Barabash, I. Sandahl, R. Lundin, H. Lauche, H. Koskinen, P. C:son Brandt, E. Roelof, L. Andersson, U. Eklund, H. Borg, J. Gimholt, K. Lundin, J. Ryno, and S. Olsen, The Microsatellite Astrid, Proceedings of the 12th ESA Symp. on European Rocket and Balloon Programmes and Related Research, Lillehammer, Norway, 29 May - 1 June, 1995. Roelof, E. C , ENA Emissions from Nearly Mirroring Magnetospheric Ions Interacting with the Exobase, Adv. Space Res., Vol. 20, No. 3, pp. 361-366, 1997 Torr, M. R., J. C. G. Walker, and D. G. Torr, Escape of fast oxygen from the atmosphere during geomagnetic storm, J. Geophys. Res., 79, pp. 5267-5271, 1974. Williams, D. J., E. C. Roelof, and D. G. Mitchell, Global magnetospheric imaging. Rev. Geophys., 30, 183-208,1992. Voss, H. D., J. Mobilia, H. L. Collin, and W. L. Imhof, Satellite observations and instrumentation for measuring energetic neutral atoms. Optical Engineering, 32, pp. 3083-3089, 1993.
364
^Swedish Institute of Space Physics, Box 812, SE'981 28 Kiruna, Sweden ^Southwest Research Institute, Drawer 28510, San Antonio, Texas 78284, USA ^Lulea University of Technology, SE-971 87 Lulea, Sweden
ABSTRACT The Munin satellite (Figure 1) is set to become the first of a new type of monitoring spacecraft. Using modem technology, this very small satellite (mass « 5 kg) will have all the necessary ftmctions needed to support its specific scientific mission: monitoring of the auroral activity on both the northern and southern hemispheres. With Munin we can now usher in a fleet of monitor spacecraft to cost-effectively provide global monitoring, and at the same time have a large degree of student involvement on different levels. Figure 1. The Munin nanosatellite. Munin has three scientific instruments for monitoring of auroral activity: • MEDUSA, a combined electron and ion spectrometer with continuos coverage of all pitch angles. Covers the energy range 10 eV -18 keV. • DINA, measures high energy ions and neutral particles at pitch angles 0° and 90°. Covers the energy range 30- 1200 keV. • HiSCC, a high sensitivity CCD camera for visible and infrared wavelengths. The camera has a field-ofview of 50°, and a resolution of 340 x 240 pixels. The payload has a total mass of 1.6 kg, and consumes 2.4 W, if power permits it will be operated continuously. The satellite is cubic with solar cells on all six sides. The solar cells provides Munin with a power of 6.0 W. The average power consumption is 4.7 W, peak consumption (when the transmitter is on) is 9.1 W. A
— 352 —
Li-Ion battery will provide the peak power and also keep the satellite powered during eclipses. The structure is made from aluminum and has a mass of 1.9 kg. The total mass of the Munin spacecraft, including the separation system, is 6.0 kg. A separation system holding the spacecraft secured to the launcher using three hooks has been designed. Munin conmiunicates on the UHF band, the up-link frequency is 449.95 MHz and down-link frequency is 400.55 MHz. The down-link is designed for a bit rate of either 9600 or 19200 bps. The modem is implemented in software in a digital signal processor (DSP), which also handles command handling and telemetry formatting. Another DSP takes care of instrument operation and data compression. The attitude will be controlled by a magnet aligning Munin with the local magnetic field. Soft-magnetic hysteresis rods will dissipate oscillation energy. Munin will use an existing ground station located at the Swedish Institute of Space Physics, Kiruna, Sweden. The ground station was buih in 1994 and has been successfiilly used for the two Swedish satellites Freja and Astrid. PROJECT OBJECTIVES Auroral Research and Forecasting The Munin satellite is set to become the first of a new type of monitoring spacecraft. Using modem technology, this very small satellite (mass « 5 kg) will have all the necessary functions needed to support a specific scientific mission such as the one we will describe in this document. With Munin we can usher in a fleet of monitor spacecraft to cost-effectively provide global monitoring, and at the same time have a large degree of student involvement on different levels. Today auroral research is a wide field covering topics such as the acceleration mechanisms of precipitating particles, the physics behind the triggering of substorms, and forecasting of magnetic storms to mitigate the damage that "space weather" can cause in e.g. power distribution lines. These research areas all need data about the conditions in space. Both the solar wind conditions and the resulting disturbances on the Earth such as magnetic and electric field fluctuations, particle populations, and auroral activity are needed in order to deduce the physical mechanisms behind these phenomena. The northern and southern auroral ovals are always present, day and night. From the Earth they can be seen by the naked eye at latitudes between 60 and 80 degrees north and south. The aurora is the visible manifestation of the particle precipitation along the Earth s magnetic field lines at high latitudes. Since the aurora can be viewed only during relatively cloud-free nights, it is not an optimal source of information about the result of magnetospheric activity. The best tool to gather this information was (very early in the history of space flight) found to be satellites in polar orbits. There have been a number of satellites of this type over the years, with new discoveries about auroral physics made by every one. Up until now, the data from such satellites have first been analyzed by the Principal Investigator of the instrument in question, then the results of the investigation have been published in scientific papers. With Munin, we propose to use a new paradigm of data dissemination, which we think will enhance the output from auroral research. We will make the data from the instruments on Munin available in real time to everyone interested, by using the Intemet and WWW-based services. Munin will provide measurements of the electron and ion particle distributions above the auroral ovals, the fluxes of energetic particles (ions and neutral particles), as well as images of the aurora taken by a CCD camera in visible wavelengths. As the data comes down to a ground station located in Kiruna, Sweden, the data will be processed such that it can be used by scientists all over the worid for auroral research. The satellite will store data from the pass
— 353
over the southern auroral oval (over Antarctica) in its memory which will then be downloaded when it is over Kiruna, whereas the datafromthe northern oval will be down-linked in near real time. The fact that the data is "fresh" will enhance the possibilities of using it for forecasting of auroral activity. The shape of the ovals, and the offset of their centers relative to the geographic poles, means, for example, that it is possible to use measurements over Scandinavia to predict the auroral activity in north America or Siberia at a later time. We intend to use the Munin measurements to present the present status of the auroral ovals, and let users of the data interpolate the position and size of the auroral oval a few hours into the future, in order to forecast auroral activity. Public and Educational Outreach The Munin team has striven to implement this project in order to promote the public interest in science, technology, space research, and international cooperation. During the last decades the public interest in these questions has declined, and we are of the opinion that a small, focused, project like Munin can catch the interest of the public. We intend to pursue this objective by visibility in newspapers, magazines, and by the Munin WWW-server, http://muiiin.irf.se. The complexity of today's space projects makes it extremely difficult for engineering students to grasp how the complete system works, or how it was designed. This results in more and more engineers becoming experts in just a small part of a project, with very limited insight in the overall system functions. With Munin, we intend to give students a possibility to be not only engaged in the mission, but also to gain a genuine knowledge about all parts of a space mission. The Swedish Institute of Space Physics is engaged in two undergraduate space engineering programs, a 3year program given by Umea University, and a 4.5 year program given by Lulea University of Technology. Students from these programs are already contributing to the project by doing their degree projects on parts of the satellite. The intention is also to have students involved in the operation of the satellite, for example in the form of lab-courses involving hands-on satellite operations. Orbit Considerations In order to fulfill the scientific objectives of the Munin project, the satellite should be injected into a polar orbit with an inclination of at least 63 degrees and an altitude between 400 and 2000 km. A circular orbit at approximately 1000 - 1400 km and an orbit plane at around noon-midnight local time is optimal, since this orbit takes Munin both through the active night-side of the auroral ovals, and the polar cusps. PAYLOAD DESCRIPTION Miniaturized Electrostatic DUal-tophat Spherical Analyzer fMEDUSA^ The primary instrument on Munin is MEDUSA (Miniaturized Electrostatic DUal-tophat Spherical Analyzer), a combined electron and ion spectrometer. The instrument sensor is provided by the Southwest Research Institute, San Antonio, Texas. Electrons and ions with energies up to 18 keV/q will be measured simultaneously, with a maximum time resolution of 16 energy sweeps per second for electrons, and 8 seconds for ions. Particles are measured in 16 sectors in the plane of acceptance, which is aligned with the Earth's magnetic field. See Table 1 for a sunmiary of the MEDUSA instrument characteristics.
354 —
Electrons are responsible for the major part of the energy input creating the aurora, and therefore the most important species to measure. The precipitating electrons energize (heat) ionospheric ions in the auroral region, these heated ions are then accelerated along the magnetic field lines and leave the heating region. This creates an outflow of atmospheric ions, primarily hydrogen and oxygen. With Munin we intend to measure both the electrons responsible for the aurora and ion heating, and the heated and accelerated ions. During certain circumstances, precipitation of ions also occur in the auroral region. The MEDUSA sensor will be mounted on Munin in such a way that each of the 16 sectors in the acceptance plane always looks at particles arriving with a certain angle to the magnetic field lines, the "pitch angle". This is possible due to the simple, yet very efficient, attitude control chosen for Munin. A permanent magnet will align one of Munin's axes along the local magnetic field line, this axis will be in the aperture plane. Particles enter the spectrometer aperture at any angle in the 10° wide plane of incidence (see Figure 2), electrons and ions are then deflected into their respective spectrometer unit by a spherical electrostatic analyzer (deflection plates). The particles hit a microchannel plate (MCP) after being filtered in energy in the electrostatic analyzer, the hits are counted by preamplifiers connected to registers, and the number of hits per sample interval are thenfiirtherprocessed by the data processing unit (DPU). The MEDUSA sensor represents a completely new development in terms of highly compact sensor design. MEDUSA will be used on the Swedish microsatellite Astrid-2.
Table 1. The MEDUSA Instrument Characteristics Instrument Characteristics 0.60 kg Sensor mass LOW Sensor power 10eV/q-18keV/q Energy range (for both electrons and ions) 25% Energy resolution (AE/E) 360°x 10° Acceptance angles 4 x lO'^cm^sr Geometric factor per sector
-0113mm AMPS
HT^.JII^
55y
ION SIDE
ELECTRON SIDE
i.
4X SUPPORT VANES
-^
':TX • •
HIGH VOLTAGE POWER SUPPLY FOR MCP & DEFLECTION PLATE
AMPS
-|^li!
Figure 2. Cross Section of the MEDUSA Sensor Unit.
— 355 —
Detector of Ions and Neutral Atoms (DINA) The meiin scientific objective of the instrument is to extend the measurements of the precipitating and mirroring ions to the higher energy range 30 - 1200 keV, complementing the MEDUSA experiment. Apart from the main scientific objective, the DINA instrument will also be able to measure energetic neutral atoms (ENAs). ENAs are produced via the exchange of charge between singly charged magnetospheric ions and atoms of the upper atmosphere/exosphere. ENAs can be found in almost any space environment and are suitable agents to remotely probe ion populations of plasmas at a distance, socalled ENA imaging. In magnetospheric physics, general attention has so far concentrated on the ENAs generated within the high altitude ring current, because they can be used in global magnetospheric imaging fi*om high altitude (> 20000 km ) spacecraft (Williams et ai, 1992). However, in the auroral region where the ring current / radiation belt particles plunge into the dense upper atmosphere/exosphere, the charge exchange process is much more eflFective and ENAs emissions are much more intense. As was shown theoretically by Roelof(\991\ ENAs fi'om low altitudes are emitted fi-om a very thin area near the exobase and the ENA generation region is, essentially, two-dimensional. The ENA camera PIPPI onboard the microsatellite Astrid (Norberg, 1995) has performed ENA measurements from a 1000 km polar orbit and demonstrated the potential of low altitude ENA imaging (Barabash et al, 1997; Brandt et al., 1997). DINA will continue studies of the low-altitude ENAs concentrating on measurements close to the generation region. The results fi-om DINA, looking at the ring current fi'om below at low latitudes and close to the ring current at high latitudes, will undoubtedly help the interpretation of the observations to be performed by IMAGE, the NASA mission to image the ring current from above. Another interesting topic to be addressed is ENA albedo. The O"^ ions precipitating onto the upper atmosphere during major geomagnetic storms transform their energy and momentum into atmospheric heating and result in escaping fluxes of fast neutral atoms via charge exchange and elastic scattering (Ishimoto et al, 1992). The total neutral flux escaping due to these processes could be significant for the atmospheric evolution during the life-time of the Earth (Torr et ai, 1974). However, no direct measurements have been performed to prove the correctness of the developed models. The DINA simultaneous measurements of the precipitating ions and ENA flux would provide the necessary inputs to evaluate the existing models of mass and energy transport in the ring current-atmosphere interaction. DINA will also investigate the low-altitude equatorial ring current formed by stripped ENAs from the main radiation belt (see Voss et al, 1995, and references therein) by measuring both the trapped ions and parent ENAs. During geomagnetic storms the ENA flux produced in the low-altitude ring current could be sufficient for detection, although for quiet conditions the flux will not be sufficient {Bishop, 1996). In summary, the scientific objectives of DINA are: 1. Measurements of the ion flux in the energy range 30 -1200 keV. 2. Measurements of the ENA flux from the exobase in the energy range 30 - 400 keV. 3. Measurements of the outflowing ENA flux (ENA albedo) from the precipitation region in the energy range 20 - 280 keV. 5. Mass resolving measurement of particles in the energy range 100 -1200 keV. 6. Studies of the low-altitude ring current. The instrument consists of two sensors with an aperture opening of 10° x 38° each (see Table 2). The sensor DINA-0 will detect particles with 0 degree pitch-angle, and DINA-90 will detect particles with 90 degree pitch-angle. Over the northern hemisphere DINA-0 provides measurements of the precipitating — 356 —
ions. While the spacecraft moves over the polar cap, DINA-90 will make measurements of ENA flux from the exobase in one local time sector. Over the southern hemisphere DINA-90 will measure the exobase ENA flux from a different local time sector due to the expected very slow spin of the spacecraft along the magnetically aligned axis. DINA-0 will be pointing toward the Earth over the southern hemisphere, and is aimed to detect outflowing ENA in the precipitation region. Information about the input ion flux can be obtained from the DINA-90 measurements because the energetic ion distributions can be considered approximately isotropic (Lyons, 1987). The instrument performs alternative measurements of ions and energetic atoms by turning on and off the high voltage on the electrostatic deflection system. Electrons with energies below 400 keV are always swept away by permanently magnetized broom magnets (the simulated transmittance for 400 keV electrons is 14%). Mass identification is performed by AE/E detectors. For the energy range 3 0 - 1 0 0 keV, the front detector (2^m Si detector) will provide the integral flux of all masses. For energies above 100 keV, coincidence and anti-coincidence logic provide measurements of hydrogen and helium (coincidence between the front and back detectors) and particles with A > 4, mainly oxygen (anti-coincidence between the front and back detectors).
Table 2. The DIN A Instrument Characteristics Instrument Characteristics Mass 0.9 kg Power 0.5 W Energy range (protons or H atoms) 30 -1200 keV Masses to resolve A = 1, 4, A > 4 for E > 100 keV Particles to measure ions, neutrals Aperture per sensor 10' x 38° 3.1 x 10'^ cm^ sr Geometric factor (per detector) Deflector cut-off energy 400 keV Broom magnets transmittance « 2% at for 100 keV electrons « 14% for 400 keV electrons
High Sensitivitv CCD Camera (HiSCC^ One of the best global space weather measurement tools is a camera imaging the auroral oval. A camera onboard a satellite supports both onboard instruments and ground facilities. Over the northern hemisphere HiSCC will take images of the night side auroral oval. This will give information about the activity level of the magnetosphere. HiSCC will support MEDUSA with identification of the spacecraft's footprint in the auroral oval, and any intensifications in the oval. Over the southern hemisphere HiSCC will operate as a star imager, enabling attitude determination based on star positions in each image. Attitude knowledge is crucial for interpretation of the measurements made by the MEDUSA and DIN A instruments, as well as a major interest for the satellite operation group since the spacecraft behavior in space can be studied. This means that HiSCC can be used to verify the performance of the spacecraft attitude control system. The camera will be the main instrument to provide the mission with pictures to the public. HiSCC data will help to educate the public about what space weather is, and the easy access to the data on the Web will increase the public interest in understanding the science of space plasmas in general and specifically the aurora.
357
The camera field of view is 50° with a highest spatial resolution of 3 x 3 km (nadir looking from an altitude of 1000 km). When the satellite is at 60° latitude the ground coverage ranges from 65° latitude and over the pole. At 75° the camera is almost looking in the nadir direction, covering 10° in both latitude and longitude. In order to avoid blurry pictures the exposure time will be below one second. Approximately 25 pictures will be taken during each pass over the northern auroral oval. The data will be compressed by the DPU before being down-linked to the ground station. The CCD camera is a modified Quickcam from Connectix Corporation. It is made for connecting to a PC parallel port; for Munin this has been modified to a special interface to the DPU. A new housing and lens system assembly has been designed. We have removed the built-in infrared filter to increase the light sensitivity. The CCD chip used in the camera is a Texas TC-255 with a quantum efficiency of up to 60% in the wavelength range 500-800 nm. It can produce pictures with a resolution of 320x 240 pixels, in 64 shades of gray. The camera can operate with different exposure times, the shortest time being 1/1000 second. See Table 3 for a summary of the HiSCC instrument characteristics.
Table 3. The HiSCC Instrument Characteristics Mass Power Camera Optics Spectral passband Detector Resolution Quantum Efficiency Exposure time
Instrument Characteristics 0.10 kg 0.3 W focusing lens, f 1.9, field-of-view 50° 450 - 850 nm TC-255 CCD chip, 50,000 e" full well capacity 0.2°x 0.2°, 320 X 240 pixels/3.2 X 2.4 mm 20 - 60 % minimum 10 ms, operationally 0.5 - 1 s
SATELLITE SYSTEM DESCRIPTION Satellite Structure The combined electron and ion spectrometer MEDUSA has a requirement of a 360° field-ofview in a plane, the satellite was thus designed as a natural extension of the MEDUSA instrument. The satellite can be considered to be an experiment with its own power and telemetry resources. The use of magnetic attitude stabilization means that Munin will rotate twice per orbit. All faces of the satellite will thus need to be covered with solar cells to maximize the energy input. To satisfy the power requirement of approximately 6 W it was decided to design Munin as a cube with the sides 210 mm long. The inner structure consists of the bottom platform, four support struts, the battery Figure 3. The interior of Munin as seen from the side of the satellite which will face towards the platform, the MEDUSA platform, and the top Earth over the northern pole (the camera platform (Figure 3). The four side panels are side). 358
screwed together mto a shell which can be integrated on Munin after all the units inside the satellite have been assembled and tested. There are aperture openings in some of these panels for the HiSCC and DINA instruments, as well as a mounting hole for the radio antenna. The structure is made of anodized alummum. Separation Syf^t^m
The low mass and the small size of the Munin satellite gives us the opportunity to design a very simple yet reliable separation system, see Figure 4. Munin will be clamped down to the launcher interface plate at three points. The satellite is equipped with three steel wires fitted with cylindrical end blocks, these blocks are held down by grappling hooks. Clamping will be provided for by a spring-tensioned Kevlar line. Three helical springs will be used for pushing Munin away from the launcher with a speed of about 0.6 m/s. A small launcher activated pyro-guillotine will be Figure 4. The Munin Separation System. With used to initiate the separation by cutting the compressed springs the total height of Kevlar line. The mass of the separation system is system is 40 mm. 350 grams. The separation mechanism will be attached to the launcher with three bolts. The separation system will include a separation switch, to indicate the separation of Munin in the launcher telemetry. Inside Munin there is also a separation switch which is in the off position when Munin is located on the separation system. At separation this switch will power-on the satellite. Power Sy>stem Each side of the satellite is covered with solar cells. We are using standard monocrystalline silicon cells, with an efficiency of 15%. These cells are cut to a size of 20 x 40 mm and are glued directly onto the satellite body with a 0.5 mm cover glass on top of each cell. Each cell interconnection is done with two leads for safety reasons. Each side of Munin will have one string with 40 cells serially connected, which will give about 6W power, and 15-20V. This power will be sufficient to run the payload instalments continuously, and the radio transmitter for 10-15 minutes per orbit. See Table 4 for the power budget. Munin will use a Lithium-Ion battery. These batteries can store more energy per unit mass and volume than Nickel-Cadmium, Nickel-Metal-Hydride or any other commonly used battery types. Lithium-Ion batteries do not have any memory effect like Nickel-Cadmium, and a very low self-discharge rate. The battery used will have three cells stacked in series to achieve a maximum bus voltage of 12.6 Volt and a nominal voltage of 10.8 Volt, v^th a capacity of 4050 mAh resulting in an energy density of 44 Wh with a mass of only 560g (including mounting parts). The depth of discharge (DOD) will typically be around 9%, this will ensure the battery survives the huge number of charging and discharging cycles it will undergo during the life-time of the satellite. The practical charging of a Lithium-Ion battery is separated into two phases, first the battery should be charged with a constant current and when the voltage over the cell reaches a certain point, constant voltage charging should be used instead. The temperature range for discharging the batteries should be kept between -10 and +50 degrees Celsius and charging between 0 and +45 degrees Celsius. A computer simulation of the thermal control, as well as 359
a solar simulation test, was performed as a M.Sc. thesis work during 1997. This work found that the battery needs a 1. 2 W heater mounted inside the battery pack to keep the battery above the minimum level.
Table 4. Power Budget Consumer MEDUSA (ion & electron analyzer) DINA (high energy particle detector) with electronics HiSCC (CCD camera) DPU (data processing unit) Charging electronics and DC/DC converters Radio transceiver (transmitting / receive only) Modem TOTAL (transmitting / receive only)
Power (mW) 1000 500 300 600 50 6500 /180 200 9150 / 2830
Radio Transceiver and Modem The conununication will be done with a modified commercial Tekk-1000 radio and a dedicated 320C50 digital signal processor (DSP) acting as a modem to digitally generate and demodulate the frequency shift keying (FSK) signals carrying the data to and from the Munin satellite (see Table 5). This will enable the DSP to control the bandwidth and speed of communication with the software implemented filters and demodulation techniques. The software based modem was designed as a M.Sc. thesis work. The spacecraft will transmit on 400.550 MHz with a maximum output power of 2.2 W; this will be sufficient for a bit rate of 9600 bps based on our experience with the Freja and Astrid satellites. During favorable conditions also 19200 bps should be possible. Munin will receive on 449.950 MHz using a narrow band 600 bps FSK since only a small amount of data needs to be uploaded to the spacecraft. The up-link will mainly be used for commands to the DPU for different operating modes and schedules, including a possibility to upload patches of the existing onboard software. The modem DSP also has a 12-bit AD converter used for monitoring of satellite status, such as temperatures, battery and sun panel voltages and currents etc. There are also standard fimctions such as a real-time clock, watchdog etc. which are mostly realized in Actel PLD (programmable logic devices). The antenna used both to transmit and receive consists of a one quarter wavelength (18 cm) whip located on one of the solar panel sides, pointing along the local magnetic field line due to the magnetic attitude stabilization scheme. The transceiver will run on a regulated 8 VoU supply requiring 20 mA in receive mode (default) and 0.8 A while transmitting with an output power of 2.2 W to the antenna. Doppler shift occurring due to the satellite motion relative to the ground will be corrected for by the ground station equipment.
Table 5. Specifications for the Tekk-1000 Transceiver. Specifications for the Tekk-1000 Transceiver Receiver Frequency 449.950 MHz Transmitter Frequency 400.550 Mhz Operating Temperature -30 to +60C
360 —
Operating Voltage Receiver noise Dimensions Mass
floor
7.5 to 12.0 V -118 dBm 85 x 52 x 21 mm 200 g
Data Processing Unit The DPU (data processing unit) handles the payload functions and data compression. The DPU used on Munin is the engineering model of the DPU for the MEDUSA instrument on Astrid-2. The heart of the DPU is a Texas Instruments TMS320C50 digital signal processor, which has 20 kb of internal memory and runs at a clock frequency of 40 MHz. We have equipped it with an additional 16kb of external RAM (used as program memory), 64k word EPROM (for flight software), and 32 k word EEPROM (for various data and software patches). The PROMS are only used (and powered) during the boot process and when patching of the onboard software is needed. For onboard data storage, 2 Mb of RAM are used for instrument data compression and buffering (FIFO). The data compression software in the DPU is based on a NASA-developed algorithm (CCSDS, 1994). We have successfully used this data compression scheme on previous missions we have been involved with. Attitude Control System (ACS) To enable continuos coverage of all pitch angles for the particle instrument MEDUSA, and to ensure correct pointing of the two DINA apertures, the spacecraft needs to be aligned with one axis along the local magnetic field. This requirement is realized by the use of a passive magnetic stabilization system. Due to the fact that Munin is a small satellite with a mass less than 5 kg, the use of this particular stabilization technique is ideal. Magnets have been used many times before, e.g. in the AMSAT microsatellites. Magnets are simple and cost-, mass- and volume effective. Soft-magnetic hysteresis rods are used for dissipation of oscillation energy. Mass budget The goal has been to keep the mass around 5 kg, this will be achieved by using light weight materials (mostly aluminum) for structural parts and a high degree of integration between the payload instruments and the satellite itself By using programmable logic devices (PLD) in the electronics we will achieve highly compact and low mass systems. To handle the high degree of integration most printed circuit boards will have six layers. Table 6 gives the mass break down for the different parts of Munin. Table 6. Mass budget Part Satellite structure MEDUSA (ion & electron analyzer) DINA (high energy particle detector) with electronics HiSCC (CCD camera) DPU (data processing unit) Battery (including holder) Charging electronics and DC/DC converters Solar cells (including cover glasses) Radio transceiver — 361 —
Mass (g) 1950 588 900 100 500 560 110 300 200
Modem Antenna + cable Separation system Magnet (attitude stabilization) Soft-magnetic hysteresis rods TOTAL
120 50 500 25 100 6003
Qn-board Data Handling The telemetry consists of data packets of fixed size (512 bytes). Every packet will start with four synchronization bytes followed by four housekeeping bytes. The remainder of the packets will contain datafi-omthe experiments, keeping the "non-science" data at a low level of only about 2% of the total telemetry. Only instrument data from the latest orbit will be sent down in the telemetry, i.e. after each telemetry session the on-board memory will be cleared and new data will be collected. In order to transmit as much relevant data as possible within the allocated bit rates, the data to be transmitted has to be reduced in some way. One solution is to use thresholds for measured parameters, and trigger data recording only when the threshold is exceeded. Our solution on Munin is to use data reduction by temporal integration, combined with loss less data compression. A data compression algorithm. Universal Source Encoding for Space (CCSDS, 1994), has been implemented and tested in the DPU. This algorithm will be used for all instrument data. There is also a mode in which only uncompressed satellite housekeeping data is sent down in the telemetry. Commands up-linked from the ground station, decoded by the modem, and considered valid by the DPU are executed either immediately or at a prescribed time. If the command is "time-tagged" it will be placed in a command queue and executed when its time-tag corresponds to the satellite clock. Ground Station Munin will use an existing ground station located at the Swedish Institute of Space Physics, Kiruna, Sweden. The ground station was built in 1994 and has been successftiUy used for the two Swedish satellites Freja (for student lab courses on satellite control) and Astrid (as a secondary ground station). The hardware used in the ground station is mainly off-the-shelf equipment, except for the modem. The ground station consists of three computers, one dedicated to Doppler shift control and satellite tracking, one for the actual satellite control, and one for Web services and presentation of received data. Operations The ground station will be designed to work unsupervised. The Munin team has gathered knowledge in constructing this type of automatic ground station during the Freja and Astrid missions. Since the satellite operations and safety fimctions to a large degree are implemented in on-board autonomy, the ground station will during normal operations only receive telemetry data. If something unexpected occurs in the ground station, e.g. there is no telemetry from the satellite during a scheduled pass, the ground station will automatically alert the satellite operator on duty by the use of a mobile pager. This scheme was successfiilly used during the Astrid mission in 1995, and will also be used for the Astrid-2 mission. MEDUSA and DINA data from the passes over the northern and southern auroral region will be included in the telemetry data. HiSCC images will be mainly taken over the northern auroral region, but some — 362 —
images will be taken over the southern hemisphere (where HiSCC looks out into space) for use in attitude determination, by comparing star positions in the image with star catalogues. PROJECT MANAGEMENT The Munin project is managed by young scientists and engineers at the Swedish Institute of Space Physics. Two Swedish universities and an U.S. research establishment are also involved in the project. Public Relations The scientific purpose of the Munin satellite is to collect data on the auroral activity. Such data are vital to prediction of space weather, but the information must be made public in order to be useful outside the Munin project team. To enable quick publication, we will use the World Wide Web (WWW), where everybody can retrieve the information they need. The idea is to use real-time uploading, from the downlink-computer to Internet, so that the information will be accessible as soon as it is downloaded from the satellite. The data will be free for everybody to use, regardless of institution, company or nationality, and will be available as soon as it is received at the ground station. We are planning to make the received data available in several ways: • • •
as simple graphs and plots, in a format that illustrates the auroral activity to the layman. as highly processed images (for scientific use, education or illustration) as uncompressed scientific data, forfiirtherprocessing by the end user
Schedule and Milestones The Munin project was conceived in September 1996, and work on the project has been on-going since then despite the fact that initially there was no guarantee that the satellite would be launched. The reason for this is that we see this project as an important contribution to science and the education of space engineers, and we have always been optimistic about the possibilities of finding a piggyback launch for such a small spacecraft as Munin. Munin is scheduled for a launch during 1999 as a secondary payload on a NASA mission using a Delta-II launch vehicle. SUMMARY Munin will fimction as a test bed for possible technologies and methods to be used in the miniaturization of spacecraft and instrumentation. Within the next decade there will be a need for fleets of small, autonomous spacecraft for various purposes. One objective for such a fleet will be to resolve the spatialtemporal ambiguity in satellite measurements. Another will be to form a fleet of spacecraft for global measurements of plasma parameters, as input for global modeling. Whatever the objective is, it will be necessary to keep the recurrent costs low when producing large number of identical spacecraft, here the use of conmiercial, off-the-shelf, equipment is essential. The miniaturization of spacecraft and instruments is necessary to enable deployment of large number of satellites. One issue to address in the fixture is the possibility to de-orbit satellites after they have ceased to ftmction. Small, simple satellites are also ideal for use in educational environments, in which a Munin-class satellite can be designed, built, launched, and operated during the time of a undergraduate or graduate student career. ACKNOWLEDGMENTS
— 363 —
Munin is financed by grants from the Swedish National Space Board. The project is managed by the Swedish histitute of Space Physics in Kiruna, Sweden, where the spacecraft and the instruments DINA and HiSCC are buih. The MEDUSA instrument is provided by the Southwest Research Institute, San Antonio, Texas, with fimds from the U.S. Office of Naval Research grant NOOO14-98-1-0175. Space Engineering students at Umea University, Sweden, and students at LuleS University of Technology, Sweden, are involved in all phases of the project. The DINA instrument will be developed, and the data will be processed and analyzed, in close cooperation with our Co-Investigators Dr. K. C. Hsieh and Dr. C. Curtis of University of Arizona, Tucson, Arizon, USA, and Dr. E. C. Roelof, JHU / APL, Laurel, Maryland, USA. REFERENCES Barabash, S., P. C:son Brandt, O. Norberg, R. Lundin, E. C. Roelof, C. Chase, B. Mauk, Energetic neutral atom imaging by the Astrid microsatellite. Adv. Space Res., Vol. 20, pp. 1055-1060,1997. Bishop, J., Multiple charge exchange and ionization collisions within the ring current-geocoronaplasmasphere system: Generation of a secondary ring current on inner L shells, J. Geophys. Res., 101, pp. 17,325-17,336, 1996. CCSDS, Source Coding for Data Compression, Consultative Committee for Space Data Systems, CCSDS 11 l.O-W-2, Washington, D.C., March 1994. Brandt, P. C:son, S. Barabash, O. Norberg, R. Lundin, E.C. Roelof, C.J. Chase, B.H. Mauk, and M. Thomsen, ENA imaging from the Swedish micro satellite Astrid during the magnetic storm of 8 February, 1995, Adv. Space Res., Vol. 20, No. 4/5, pp. 1061-1066, 1997. Ishimoto, M., G. J. Romick, and C.-I. Meng, Energy distribution of energetic O"*" precipitation into the atmosphere, J. Geophys. Res., 97, pp. 8619-8629, 1992. Lyons, L. R., Processes associated with the plasma sheet boundary layer, Physica Scripta, 8, pp. 103-110, 1987. Norberg, O., S. Barabash, I. Sandahl, R. Lundin, H. Lauche, H. Koskinen, P. C:son Brandt, E. Roelof, L. Andersson, U. Eklund, H. Borg, J. Gimholt, K. Lundin, J. Ryno, and S. Olsen, The Microsatellite Astrid, Proceedings of the 12th ESA Symp. on European Rocket and Balloon Programmes and Related Research, Lillehammer, Norway, 29 May - 1 June, 1995. Roelof, E. C , ENA Emissions from Nearly Mirroring Magnetospheric Ions Interacting with the Exobase, Adv. Space Res., Vol. 20, No. 3, pp. 361-366, 1997 Torr, M. R., J. C. G. Walker, and D. G. Torr, Escape of fast oxygen from the atmosphere during geomagnetic storm, J. Geophys. Res., 79, pp. 5267-5271, 1974. Williams, D. J., E. C. Roelof, and D. G. Mitchell, Global magnetospheric imaging. Rev. Geophys., 30, 183-208,1992. Voss, H. D., J. Mobilia, H. L. Collin, and W. L. Imhof, Satellite observations and instrumentation for measuring energetic neutral atoms. Optical Engineering, 32, pp. 3083-3089, 1993.
364