The primary objectives of the MarsCAT mission are as follows:
- Explore the interaction between Mars’ atmosphere and the solar wind as defined by the MarsCAT science goals.
- Demonstrate CubeSat interplanetary exploration enabled by the CubeSat Ambipolar Thruster (CAT).
- Support, augment, and possibly extend the mission objectives of NASA’s MAVEN.
The following sections discuss the feasibility of meeting these objectives.
Concept of Operations
MarsCAT consists of several identical 6U CubeSat spacecraft compliant with applicable Cal Poly CubeSat and Poly Picosat Orbital Deployer (PPOD) standards, and with the paper An Advanced Standard for CubeSats, related to 6U and larger CubeSats, and with NASA SMD Guidelines For Heliophysics EM-1 CubeSat Proposers, .
The authors have determined that the best option for the MarsCAT spacecraft is to launch with a major Mars mission and remain stowed onboard until reaching Mars. They will jettison from the major mission at the edge of Mars’ gravitational sphere of influence and use either the RFT or Pod Carrier propulsion to achieve capture and Mars orbital insertion (MOI). Exhaustive trades revealed three compelling reasons that make direct from Earth (DFE) mission plans infeasible. First, even with the RFT propulsion system, the amount of propellant required for independent transit to Mars by each spacecraft, followed by MOI to a relevant science orbit, would be prohibitive given the constraints of a 6U form factor. Second, it was deemed that the risk of CubdeSats navigating a transit trajectory to Mars independently and reliably arriving within a reasonable time window for multipoint science observations was too high. Third, mitigating the risk of additional radiation exposure on an independent transit (without the shielding mass of the carrier) was infeasible within the available 6U mass constraint. For these reasons, MarsCAT proposes to launch with a major mission.
Each MarsCAT spacecraft is equipped with the Radio Frequency Thruster (RFT), a 3-axis attitude determination and control system capable of attitude solutions in interplanetary space and near Mars, and an embedded radio navigation system utilizing the navigation capabilities of the Mars Relay’s CCSDS protocol. Upon ejection from Mars 2020 at Mars’ gravitational sphere of influence (500,000 km altitude), the RFT performs a finite continuous maneuver to transition from MOI to a highly elliptical, stable orbit of 200 x 7000 km and 60° +/ 5° inclination. The planes of the two MarsCAT orbits will be separated by approximately 15° in right ascension of the ascending node (RAAN).
Each MarsCAT spacecraft carries a science payload consisting of two three-axis magnetometers, a Faraday cup, double-Langmuir probes, and a dual-frequency crosslink radio occultation experiment embedded within the spacecraft communication system. Communications with Earth will be over UHF via the Mars Relay Electra package on MAVEN and MRO. MarsCAT operations are planned for one year following Mars orbit insertion.
During the disposal phase, any remaining fuel will be used to raise the periapsis as high as possible. This requires very little fuel because of the eccentricity of the orbit and changes of 100 km can be accomplished with only ~10 m/s delta-v expended at apoapsis. Preliminary analysis shows MarsCAT complies with the COSPAR Planetary Protection Policy,  and NASA Procedural Requirement NPR 8020.12D, Planetary Protection for Robotic Extraterrestrial Spacecraft, . The planetary protection requirement is a disposal orbit with 0.99 certainty of no impact with Mars within 20 years and 0.95 certainty within 50 years.
Figure 1: MarsCAT Concept of Operations
Launch & Mars Transit
Two launch and Mars transfer scenarios were analyzed: 1) Deploy from EM-1 or equivalent and perform transfer maneuvers using the RFT; 2) Launch on major Mars mission and deploy when nearing Mars and use RFT to slow for MOI. Analysis showed option 1 (launching on EM-1) would not be feasible to complete mission objectives; the delta-v required from Earth escape velocity to Mars transfer would leave insufficient propellant to lower into an operational orbit. Starting from Earth escape (EM-1 release) and assuming a best-case scenario that the 6Us were released in the ideal direction and no pointing adjustments were necessary, 3.38 km/s of delta-v would be required for each spacecraft for trans-Mars injection. [A lower delta-v of 2.98 km/s could be possible, however, this is only achieved by spreading the maneuver over multiple orbits of the sun, in essence becoming like a Hohmann transfer spread over many orbits. Each additional orbit adds >1 year to the voyage, and it would take many orbits to approach that best case delta-v, which is not practical and very risky given both that there are multiple spacecraft involved and the extended exposure to the radiation environment.] Assuming 6 kg of iodine propellant (50% of the total mass) and an Isp of 690s, each spacecraft would have 4.78 km/s of delta-v capacity, so that 1.40 km/s would then be left to capture into a Mars orbit. The calculations show that this is insufficient to achieve any orbit around Mars. At a minimum, about 1.8 km/s of delta-v would be required to achieve a non-circular orbit around Mars, however, the capture orbit would be too highly eccentric, with an apoapsis near the edge of Mars’ sphere of influence and very little time spent at low altitude for the science.
Based on these trades, the team concluded that the only feasible option to meet the science and end of life requirements was to go with option 2, and piggyback on a major Mars mission to a Mars transfer orbit and deploy when nearing Mars and use RFT to slow for MOI or rely on the Pod Carrier to provide the MOI burn. The balance of this document assumes option 2 as an element of the concept of operations.
Science Instruments Overview
The structure and dynamics of the plasma and magnetic field environment around Mars is determined by interaction of solar wind and the Martian ionosphere and crustal magnetic field. Therefore, five instruments are needed to determine the local solar wind and Mars ionospheric properties and to address all of the science requirements: a magnetometer, two plasma experiments, an energetic radiation detector, sand a remote sensing radio experiment. The investigators were selected based on their strong expertise and experience in instrumentation for space missions. The instruments share a common spacecraft bus interface, spacecraft DPU, and ground system operations.
The Induction Magnetometer (iMAG) to be flown on the two MarsCAT spacecraft is a new magnetometer design, but derives heritage from a long line of space-qualified fluxgate magnetometers built by the Institute of Geophysics and Planetary Physics at UCLA.
Important assets of the iMAG magnetometer are its inherent linearity (with nonlinearities significantly smaller than 1 part in 104) and the reliability and stability of the gains and offsets. This instrument, which is a completely different design from the traditional fluxgate magnetometer design, is inherently rad hard, low mass, low power and temperature insensitive. The electronics functional block diagram is shown in Figure 2. There is no microprocessor in this simple design.
The iMAGS magnetometer continues the digitization and miniaturization path of the UCLA NASA ST5 and MMS fluxgate magnetometer designs. The current circuit design realizes significant reductions in power, mass and size. Additional mass savings have also been achieved by the use of extremely small sensors and the elimination of the boom and sensor boom cable. Figure 2 shows an engineering version of the iMAG sensor.
The current generation of fluxgate magnetometers being flown on missions such as ACE, Wind, Cluster II are too bulky (several kg) and power hungry (several W) to be flown on CubeSat class missions due to the ultra-constrained resource environment of micro-satellites. They are also too expensive. Even more modern small magnetometers, such as built for the NASA ST-5 and MMS missions are large compared to what is proposed here (Table 2).
To address the science requirements, our goal is to reduce the mass and power of DC magnetometers by an order of magnitude over those currently flown while approaching the precision, noise-level, linearity, and stability of past fluxgate magnetometers. An innovative approach enabled by low-cost magnetometers is the elimination of large booms, by placing multiple sensors in and on the bus in combination with a careful magnetic cleanliness program.
← Scroll Table →
Mission Name Ranges (nT) Cadence (Hz) Sensor Mass (g) Elect Mass (g) Circuit Area in2 Power (W) ST5 60,000 16 75 650 35 0.5 Proposed New Mag 60,000 Up to 147 15 100 1 0.1
Table 2: Comparison between iMAGS and ST-5 magnetometer.
The UM new digital induction magnetometer takes advantage of mobile phone magnetometer sensor development to reduce the mass, power consumption, and increase the radiation tolerance of the fluxgate magnetometer. The instrument does not use an A/D converter making it much more radiation tolerant than traditional fluxgate magnetometer designs. The commercial magneto-inductive magnetometer from PNI will be used with custom built sensors to increase the sensitivity and a custom housing to provide thermal and radiation protection. In the new induction magnetometer, the magnetic field is measured by counting the time between flips of the magnetic induction of the circuit which is dependent on the strength of the applied DC field. As seen in the functional diagram in Figure 2 top, the magnetometer is a LR circuit with a Schmitt Trigger for counting pulses.
Figure 2: (left) Block diagram of PNI Induction 3-axis UM Magnetometer. (right) Sensor assembly size compared to penny.
In Figure 2, HE is the external magnetic field parallel to the coil. The total field that the sensor experiences is due to the external field and the field generated by the circuit (H = kI +HE, where k is a property of the sensor, and I is the current through the circuit). The Schmitt trigger causes the current through the circuit to oscillate as the voltage bases a set “trigger” value. The time between oscillations or trigger flips is dependent on the strength of the external field and therefore the DC field can be measured by simply counting the number of flips or triggers.This design is inherently radiation tolerant due to the elimination of the need for any form of ADC. Power consumption and mass are lowered by a reduction in the number of amplifiers required, as well as the elimination of the ADC. In fact the entire three-axis magnetometer can be implemented on one 1 cm x 2 cm printed circuit board.
The size and mass of the new magnetometer instrument is extremely modest compared to traditional fluxgate magnetometers and the design significantly reduces the number of components in the overall analog circuitry. The prototype vector iMAGs magnetometer weighs less than 10 gm including the custom built sensors, but not including the housing.
One hidden cost of fluxgate magnetometers is the need for a boom. This drives up mission complexity and adds significant mass due to meters of cabling and the boom itself. By driving down the resource needs and cost of the magnetometer, a new approach can be incorporated in MarsCAT that eliminates the need for a boom,. This approach places several magnetometers inside and on the bus to be able to identify spacecraft magnetic signals in the data so that the external field can be recovered with processing and careful magnetic cleanliness and characterization prior to launch. With CubeSats (even 6U structures), testing the fully integrated system is possible within the UM magnetic testing facilities. Such a characterization and testing was done with the 3U RAX mission.
The development and testing of the new sensor is on-going (performance testing, environmental testing – including radiation and thermal vacuum testing) and a high-altitude balloon flight in May 2015. The system is being proposed to fly on Earth orbiting CubeSat missions years prior to the Mars launch to raise the TRL, provide heritage and reduce the risk on both the sensor and boom-less design.
Thermal Ion Analyzer
An instrument to measure the major thermal ion constituent densities, temperature and velocity is patterned after successful such sensors that have flown in many earth orbiting missions. Of greatest relevance to the proposed concept is a dual function ion velocity meter (IVM) that measures both the ion temperature and the bulk ion drift vector and occupies a small volumetric envelope compatible with a CubeSat. The SORTIE mission, which is scheduled to launch in 2017, utilizes this technology in a low earth orbit CubeSat mission to be deployed from the Space Station. Figure 3 shows an isometric view of the sensor, which has an unrestricted view approximately along the satellite velocity vector.
Figure 3: Thermal ion analyzer
It utilizes a square entrance aperture and a segmented collector, to measure the ion flux on each segment as a function of a retarding potential that is applied internally to a planar grid to control the energy of the ions that have access to the collector. The sensor aperture and internal electronics are electrically isolated from the spacecraft power system and the floating aperture potential provides an independent ground reference for the sensor potentials that is close to the plasma potential and independent of the spacecraft ground. In Mars orbit the ions flow supersonically into the sensor and the collector current measured as a function of retarding voltage may be fit to extract the ion temperature and the ion velocity along the sensor look direction. The major constituent ion densities are resolved by the mass dependence of the ram energy. The current distribution among the segmented collectors provides the arrival angle of the plasma with respect to the plasma look direction, thus permitting a derivation of the full plasma flow vector [Hanson and Heelis, 1975; Heelis and Hanson, 1998].
Double-Langmuir probes (DLPs) measure the ambient ion density, ion flux, and electron temperature, and electron density,. These probes are not to be confused with “double probes,” which is a common term in the space physics community to refer to two boom-mounted probes for making DC-AC electric field measurements. Each probe of the DLP is biased with an AC voltage with respect to the other probe, so that ion or electron loss area problems are mitigated from normal single Langmuir probe problems onboard spacecraft. This means that the entire spacecraft body could be made of insulating material and the DLP would still function nominally. This advantage of DLPs is important for use on CubeSats, since the exposed conducting surface area can be somewhat small. For a single Langmuir probe to provide accurate results, the spacecraft conducting surface area needs to be sufficiently large as an ion loss area when the single Langmuir probe is biased to electron saturation so that the probe itself doesn’t alter the ambient space plasma potential that is being measured. For atomic oxygen, this ratio of electron loss area at the probe tip to ion loss area at the spacecraft surface is ~200:1, and ~250:1 for NO+ ,57,. The DLP however, is a floating system and electrically isolated from the spacecraft ground/skin. Ion and electron saturation currents that are drawn to the DLP probe tips are effectively drawn between each probe tip, and the electron saturation current drawn in one probe is limited by the ion saturation current drawn in the other probe. A voltage sweep is required for DLPs, as is a transimpedance (current to voltage) amplifier for measuring the current between the two probes of the DLP. The anticipated ion current collected by each probe is on the order of nA/cm2, well within the detection limits and signal to noise ratios for modern transimpedance amplifier designs.
DLPs have not been extensively used in the early decades of spaceflight owing to the low ion saturation current densities in orbit and the required current detection limits of the then existing electronics. The modern transimpedance amplifier circuits for this application are non-trivial, but are achievable within the form factor and power limitations of a CubeSat. Up until now, the majority of probes flown into space have been used to measure AC or DC electric fields, though the use of swept single Langmuir probes (SLPs) has been steadily growing57, . Until the advent of automatic gain control circuits, the single Langmuir probe analysis ignored the ion saturation current portion of the I-V trace, Fig. 3. Photoemission from the spacecraft body is the other major complicating factor. In some cases, the photoemission current can be of a similar current density and can swamp the ion current, or it can lead to measurements of the electron temperature that are too high. The solution requires that the double probes be mounted on extension booms that are far enough away from the spacecraft body. The two probes also need to be separated by several/many Debye lengths from each other so that the ion sheaths do not overlap. However, in the relatively high plasma density regions of low Mars orbit, these issues can be mitigated with careful design. This may limit the effective measurement range of a CubeSat mounted double probe to the interior of the ionosphere.
Figure 4: Current-voltage characteristic trace: a) “normal” lab based single Langmuir probe (SLP). b) lab based double Langmuir probe (DLP). Figure adapted from Stenzel, .
Once an I-V characteristic is generated, the plasma parameters can be found from fitting a hyperbolic tangent function. Fig. 4 shows a representative I-V trace from a laboratory based “normal” single Langmuir probe, a), and a laboratory based DLP, b). DLPs are swept in voltage to find the current-voltage characteristic. The power and volume requirements will be similar to the Langmuir probes on DICE. The probes will deploy on a pair of rigid booms in a “V” configuration to open the path between the probes. Plasms density can then be inferred from observation of the plasma frequency.
Dual-Frequency Crosslink Radio Occultation Experiment
Figure 5: Differential pseudorange delays due to a given plasma density of 1e11 m-3 along signal raypath. Differential delay between 390 and 450 MHz is 2.2 ns/TECU compared to 1 ns/TECU for typical measurements made at Earth using GPS. Solid and dashed red lines show selected frequencies are well above critical frequency for 1e11 m-3 plasma density and also for one order magnitude higher at 1e12 m-3 .
Radio tomography of the Earth’s ionosphere has enabled near-global observations of the structure and dynamics of key ionospheric regions and processes. Radio tomography of the Earth’s magnetosphere has also been proposed to help place multipoint in situ observations into global context. MarsCAT will use its telemetry transceivers to measure electron column density data between the two spacecraft, indicative of kilometer-scale density structure. A 10 s data cadence will provide 1000s of ray-paths per orbit from a variety of viewing geometries including occultation geometries near Mars’ periapsis.
The radio science technique of measuring the integrated column electron density (or Total Electron Content (TEC)) is accomplished by having a spacecraft transmit coherently phased pairs of radio frequencies to a receiver. The dispersion of the two frequencies through the Martian ionosphere separates their arrival time at the receiver. The differential delay between the two signals gives an accurate measure of the TEC. At Earth, this can be achieved using dual-frequency GPS signals. However at Mars, the experiment will use the telemetry transceivers on both MarsCAT spacecraft, each will transmit and receive simultaneously at 390 and 450 MHz, frequencies selected to maximize dispersion and reduce the time synchronization burden on the receiver oscillator and microprocessor while also remaining with the frequency band of the Mars Relay CCSDS protocol. The two spacecraft clocks will be synchronized using the microprocessor to align a pseudorandom code sequence, similar to how low-precision GPS receivers maintain accurate time.
We will explore including low power hybrid pixel radiation imaging detectors from the CERN-based Medipix2 Collaboration [Llopart et al., 2007; Kroupa et al., 2015] on MarsCAT during the concept study. These detectors can measure energy and arrival directions of electrons and ions from 3 keV to GeV energies.
Figure 6: ISS Medipix2 (Timepix) dosimeter
The Timepix-based radiation imaging devices from the CERN-based Medipix Collaborations take data in gated time “frames,” after which the raw 14-bit ADC digitized data from all 65,536 of the 55 micro-m pixels in each device are shifted out, and zero-suppressed during the process, so the final data frame will have a variable size. NASA has selected this technology for the next generation of radiation monitors on the new Orion MPCV. Versions of this technology in units the size of USB memory sticks, which are powered and controlled via normal computer USB ports have been functioning successfully on the ISS for over 4 years. These devices are planar with an area of 2 cm2, and can measure the energy and track-length in the sensors of thicknesses up to 1 mm. Analysis of the track images allows particle-by-particle charge and energy estimates as well as relative track orientations. The detection efficiency for penetrating charged particles is 100%. Estimates of incident x-ray and gamma-ray fluences are also possible. Multiple individual orthogonal detectors can enhance the 3-D characterization of the incident radiation field. Recent versions of this technology using data-driven readout rather then frame-based have demonstrated the capability to function in fluxes as high as 107 protons/cm2s without data loss.
Science Mission Profile
Once released from the carrier spacecraft into a Mars transfer orbit and using its RFT engine with nominal parameters of 1.1 mN of thrust and approximately 6 kg of iodine propellant at 690 s of Isp, each CubeSat inserts into a highly elliptical Martian orbit to perform the science mission and finally raise the periapis for a stable disposal orbit. The orbit inclination is 60° +/ 5° to focus on the concentration of crustal magnetic fields in the southern hemisphere centered around 60° south latitude and 180° longitude. A science mission lifetime of 1 Earth year is planned.
Approaching Mars on an impulsive Hohmann trajectory via a major mission carrier will give the spacecrafts an intercept velocity of 2.67 km/s. While still far from Mars, the trajectories will be adjusted to insert into the desired inclination and periapsis at minimal delta-v cost. The minimum delta-v for Mars capture to an orbit of 200 x 500,000 km would be 2.40 km/s, costing 3.64 kg of propellant (assuming an initial wet mass of 12 kg). The periapsis altitude has a negligible effect on the delta-v cost, it is lowering the apoapsis that is costly. This capture maneuver is a continuous burn over 4.7 months so as to be captured as quickly as possible.
Once captured, the apoapsis is lowered to decrease the orbital period and increase the total time the spacecraft spends near periapsis, where the science mission is to be completed. The closest orbit achievable will be 200 x 7,000 km which will require an additional 2.3 months, a total of 7 months, and 2.33 kg of additional propellant for the 2.17 km/s maneuver. The science mission orbit has a period of 4.9 hours, though higher apoapsides would be acceptable if less propellant mass than expected is available. If the carrier spacecraft or the pod carrier provides the MOI burn, all of the RFT delta-v will be available for mission cnfiguration maneuvers
A highly elliptical orbit offers great flexibility for the science mission. With small burns at apoapsis, the periapsis can be raised or lowered between 80 and 200 km at the cost of only a few 10s of m/s. This allows for deep dives and other altitude changes that could greatly improve the quality of the science product without significantly affecting the propellant budget.
Finally, after the science mission has been completed (nominally 1 Earth year or possibly after any further science extensions given that the spacecraft are still healthy), the MarsCATs will transfer into a disposal orbit, raising the periapsis to 250km. This orbit will be stable for 100 years using worst-case-scenario drag area and drag coefficients. The amount of propellant to be held in reserve beyond the science just for this purpose is TBD; however, preliminary analysis indicates that the mission periapsis seems very stable so the reserve amount of propellant needed to raise to the disposal orbit is on the order of 10s of grams.
Flight System Capabilities
Multiple identical 6U structures are to be custom built based on lightweight skeletonized structures designed for previous successful CubeSat mission at Texas A&M and the University of Michigan. Provision will be made for deployable solar arrays as well as two deployable booms for the Langmuir probes. The total allowable spacecraft mass will be 14 kg, the internal volume will be no greater than 7000 cc, the overall total dimensions will be no larger than 239.2 mm by 105.6 mm by 365 mm, and the spacecraft will be compatible with the Planetary Systems Corporation (PSC) Canisterized Satellite DispenserTM.
GNC and Data Transfer. The MarsCAT CubeSats will not have access to communication with Earth until ejection from the transporting spacecraft at roughly a distance of 500,000 km from Mars. At this point there are two options for guidance, navigation and control (GNC). One option is to communicate directly to Earth (DTE) using the NASA Deep Space Network (DSN). The other option is to communicate with the existing Mars orbiters (e.g., MAVEN or MRO).
Communicating DTE would be done using one of the DSN 34 meter dishes (e.g, the 34 meter beam-waveguide dish) at either S band or X band. Navigation (Doppler and ranging) would require a deep-space transponder. NASA JPL has developed the Iris CubeSat-compatible deep space transponder for INSPIRE, the first CubeSat to voyage to deep space. Iris is 0.4 U in size, 0.4 kg in mass, consumes 12.8 W, and interoperates with NASA’s Deep Space Network (DSN) on X-Band frequencies (7.2 GHz uplink, 8.4 GHz downlink) for command, telemetry, and navigation . The recent solicitation indicates that this system should be available for these mssions as Government Furnished Equipment. Furthermore, since the MarsCAT CubeSats will only have low-gain antennas (LGAs), the data rate using DTE communication will be limited to a few bps at this range, based on published Iris data. This data rate will allow for limited GNC. It will not allow for data transfer from MarsCAT to Earth at a rate that is sufficient to accommodate the science experiments. Therefore, communication with the Mars orbiters is necessary even if the Iris transponder is incorporated into MarsCAT. Therefore, the decision is to forgo the use of a deep space transponder and rely only on communication with the orbiters.
Communication with the existing Mars orbiters will be done via the UHF Electra system onboard the orbiters, which is supporterd by IRIS. The Electra system is a software defined transceiver radio system operating in the 390-450 MHz band that uses the CCSDS Proximity-1 protocol for communication. The capability to use the Electra system for GNC has already been demonstrated, though not for CubeSats. Therefore, this choice presents a low-moderate risk.
Link Budget and Data Transfer: Based on the bit error rate (BER) performance of Electra, a link budget analysis determines the communication data rate between the MarsCAT CubeSat and the Electra system on the orbiters. The results, based on published Electra data, are shown in Table 3. The link budget has assumed a full duplex mode (8.5 W transmit) using a UHF transmit CubeSat antenna with a gain of 6 dBi and a UHF low-gain nearly omnidirectional receive antenna (e.g., a quadrifilar helix) on the orbiter with a gain of 0 dBi. Three different BER were assumed. The range calculations for data rates at 8000 bps and above were based directly on the published Electra data while the range calculation for data rates below this were based on extrapolation, assuming that the data rate is proportional to the SNR.
Table 3 also shows that early navigation using a data rate of a few bps can be established at a distance of about 500,000 km from the orbiters, which is the planned range for initial communication.
A moderate data rate (50 kbps) will require the CubeSat to be within approximately 3,000 km of a relay orbiter. This data rate will be used for relaying data back to Earth from the science experiments on the MarsCAT, once MarsCAT is close to Mars in final orbit.
← Scroll Table →
Data Rate (bps) Range (km) BER = 10-3 BER = 10-4 BER = 10-5 1,024,000 780 620 550 512,000 1,400 1,100 870 128,000 2,800 2,200 1,900 32,000 5,500 4,400 3,900 8,000 11,000 8,800 7,800 2,000 22,000 17,600 15,500 500 44,000 35,200 31,000 125 88,000 70,400 62,000 31 176,000 140,800 124,000 8 352,000 281,600 248,000 4 704,000 563,200 496,000
Table 3: Electra range calculation for various BER.
It is estimated that the amount of data that needs to be transmitted back to Earth from the science experiments is on the order of 50 MB per week per CubeSat. If we assume an average range of about 3,000 km during the communication link, the data transfer rate will be roughly 50 kbps (0.0061 MBs). It will thus take approximately 140 minutes per CubeSat to relay the data to the orbiter if the transfer is scheduled for once per week. If the data is transferred twice weekly, the data transfer time will be approximately 70 minutes per CubeSat.
The data transfer from the orbiter to Earth will be faster than that from MarsCAT to the orbiter. As an example, the MAVEN satellite transmits to Earth twice per week, for five-hour periods. The data transfer rate from MAVEN to Earth via the X-band high-gain antenna (2.1 meter dish) on MAVEN is approximately 550 kbps (0.0671 MBs). Thus, it will take approximately 25 minutes to download the stored data from MAVEN to the DSN each week. If the data is transferred twice weekly, the data transfer time will be approximately 13 minutes.
MarsCAT Antennas. Each of the several 6U MarsCAT CubeSats will have a UHF antenna system onboard, which will be dual-use. The antenna system will be used for GNC and data transfer, and also for the total electron count (TEC) science experiment. The dual-use design using the same UHF bands is possible since the GNC/data transfer and the TEC experiment will not be done simultaneously. This allows for an optimum antenna system that makes the best possible use of the limited space resources on the CubeSat. Each antenna system will consist of two antennas, one on the front side of the CubeSat and one on the back side. This will allow the CubeSat to communicate with the orbiters or with the other CubeSat (for the TEC measurement) using whichever antenna has the highest received power, enabling reliable communications without having to reorient the CubeSats. This will also provide a redundancy, in case of a failure of one of the antennas. The antenna system will consist of a pair of planar dual-band UHF microstrip antennas covering the 390-405 MHz transmit and 435-450 MHz receive bands (used for both GNC and data transfer).
One of the novel aspects of the antennas is that they will be designed based on microstrip antenna technology, so that that the antennas will be planar and low profile, and can therefore be integrated directly onto the CubeSat frame, enhancing mission reliability and performance, as there will be no risk issues associated with mechanical antenna deployment. This is a significant risk factor when using conventional whip antennas on CubeSats.
At the University of Houston, the Small Satellite Laboratory recently, under NASA contract, began improving antenna design for CubeSats. Since the antenna is instrumental for communications, a failure of the antenna can have disastrous consequences for the CubeSat mission. Conventional whip antennas rely on a mechanical deployment mechanism, which is subject to jamming or breaking during deployment. One of the important goals of the SSL CubeSat research is to replace these off-the-shelf whip antennas with customized low-profile antennas that are integrated into the frame of the CubeSat body. The low-profile designs do not protrude from the CubeSat surface by more than 0.5 cm, and are mechanically durable. There are two options for the planar dual-band UHF antenna (subsolar and transparent), and these are described below. The best candidate will be selected after examining the performance (radiation pattern, impedance match, bandwidth) and final weight of each option.
Subsolar Antennas: One type of planar low-profile antenna being developed is the subsolar antenna. This consists of a microstrip antenna that goes across the entire face of a 3U CubeSat. The upper face of the microstrip antenna is metal, so that the solar panels can be placed on top of the antenna without degrading the antenna performance (Montaño et al. 2014). The antenna radiates from its edges. The radiating edges of the microstrip antenna are aligned with the edges of the CubeSat, and hence are not blocked by the solar panels. Such antennas have been successfully developed at the UHF 434 MHz ISM band. Fabricated prototypes are shown in Fig. 7. These designs will be extended to dual-band performance, to cover both the transmit and receive Electra bands.
The resulting antennas are low profile and integrated directly onto the CubeSat frame, giving much better reliability and compactness in the design that what can be obtained using conventional antennas that protrude from the CubeSat. This new approach to CubeSat antenna design is expected to greatly increase mission reliability.
Figure 7: (left) Three low-profile CubeSat Integrated Microstrip Antennas developed by Small Satellite Laboratory (before SMA connectors). (right) Prototype low-profile transparent meshed microstrip patch antenna fabricated on quartz substrate.
Transparent Antennas. Transparent microstrip antennas have been developed at the SSL for operation at 2.4 GHz. These antennas are fabricated from a meshed metal conductor (silver epoxy) that is deposited on the surface of a transparent quartz substrate. The meshed silver epoxy is used to construct both the microstrip patch antenna surface as well as the ground plane surface below the patch. Figure 5 shows one fabricated prototype.
These transparent antennas are designed to be placed on top of the solar panels on the CubeSat, so that they are functional without blocking a significant portion of the light from reaching the solar panels. A typical transparency is about 75%.
These designs will be extended to cover UHF Electra bands, allowing for both communication and navigation with the Electra system on the orbiters as well as the TEC experiment.
Using these antennas, there is a mild loss of gain due to the 25% blockage, corresponding to a loss of approximately 1.2 dB. However, there may be advantages to the transparent design in terms of the pattern or bandwidth. These antennas are not constrained to radiate only from the edges, as there is no concern about components being placed on top of the antenna, so the entire antenna surface can be used as the radiating surface, allowing for more flexible design.
A careful comparison between the two antenna types will be made to determine the best candidate for the MarCAT mission. This trade study includes a comparison of weight, reliability, radiation pattern, gain, impedance match and bandwidth.
Thermal and Radiation
The main thermal concerns are for the batteries and fuel in transit. Unsteady finite-element thermal modeling will be accomplished, and passive thermal control will be applied via judicious component placement, radiators, and thermal blankets. Once in Mars orbit, if active heating is required, simple patch heaters will be used.
At Mars, solar energetic protons (SEP) are the primary source of radiation. The on-board computer selected is rad-hardened (see Command & Data Handling, below). As well, various of the other components, many commercially available and CubeSat compatible, will be rad tolerant or hardened.
Electrical Power System (EPS)
Table 4: Power and energy budget.
Overview. The MarsCAT power system consists of a 7.2 V bus regulated to 5 V and 3.3 V supplies. Primary power comes from a deployable solar array coupled with body-mounted cells. 185 SpectroLab XTJ triple-junction solar cells with 29.5% beginning-of-life collection efficiency will be used. Solar arrays are sized for performance after three years in the vicinity of Mars. MarsCAT is launched fully charged to provide all of the necessary start-up power required to deploy the solar arrays and fire the RFT. Testing will confirm the batteries’ ability to hold charge during Mars transit. In addition, the team will investigate the prospect of a power line through the PPOD for battery maintenance.
While MarsCAT is stowed, the power system is inhibited via two redundant separation switches contacting the rear of the deployer. Upon deployment the plunger switches are released, and the power system turns on the flight computer to begin its boot cycle.
Power Generation Degradation. Assuming coverslide degradation at ~15% per 1e+16 1-MeV equivalent electrons (though at Mars, solar energetic protons (SEP) are the primary source of radiation), the CREME-96 short term (solar event) predicts an equivalent 1-MeV fluence of 6.64e+15 for one year at 1.5 AU, or approximately 10% degradation per year for solar events similar to the October 1989 event. The long term proton fluence given by the ESP-PSYCHIC model suggests degradation on the order of an additional 0.02% per year.
Battery Sizing. Battery packs will be built from eight 3.6 V Panasonic Li-ion batteries with specific energy density of 243 W hr kg-1. Two batteries will be in series with 4 sets in parallel.
Regulation and Control. The power board also provides fault protection that can be reset via the Command and Control board.
Power and Energy Budget. The power and energy budgets are shown in Table 4.
The MarsCAT spacecraft will be propelled using the Radio Frequency Thruster (RFT), a permanent magnet helicon plasma thruster specifically designed for CubeSats. With an Isp of 690s and a thrust to power ratio of 73 mN/kW, RFT originated at the University of Michigan and is now being developed by PhaseFour, Inc., which has secured a flight to demonstrate this technology in 2017 or 2018.
The MarsCAT mission will build off of the success of the upcoming RFT demonstration mission. With lessons learned the thruster will be improved to ensure mission success. The RFT demonstration mission will use xenon as a propellant but to achieve the necessary delta-v for MarsCAT without taking up too much volume in the spacecraft iodine must be used. RFT’s electrodeless design is uniquely positioned to utilize iodine, which stores as a solid, minimizing propellant volume. Iodine, however, has never been demonstrated as a propellant and the TRL of the propellant management system is quite low (~2). Since iodine propellant is an enabling technology for this mission, so part of the MarsCAT budget will be dedicated to developing a fully functional and flight ready iodine propellant management system.
Guidance & Navigation
The attitude determination and control system (ADCS) will be composed of two subsystems: the attitude estimation subsystem and the attitude control subsystem. ADCS comprises three reaction wheels (Sinclair Interplanetary RW3-0.06-28), two horizon sensors, one star tracker (Sinclair Interplanetary ST-16RT), photodiodes on the solar array for coarse sun sensing, and rate gyros all coupled with algorithms and small burns by RFT. An Accion Max-1 thruster system will be used to desaturate the wheels. Refinements on accuracy and knowledge estimates will occur once the award is in place, but it is apparent from our investigations thus far that CubeSat compatible systems are available to meet mission performance needs.
Sensor data will be fused together using a multiplicative extended Kalamn filter (MEFK), a well know and robust attitude estimation algorithm. The MEKF will have various modes whereby different sensors will be used should some not be available or fail prematurely. The attitude estimate computed using the EKF will be passed to the attitude control system. A quaternion-based filtered PID+ control law will realize both coarse and fine pointing. Rather than using a constant gain PID control law, filtered PID control plus feedforward control given the desired spacecraft attitude will be used to realize faster slewing, reduce overshoot, and a faster settling time. Filtered PID+ control is a well established, robust, control technique.
Command & Data Handling (C&DH)
For mitigation of radiation effects, the on-board computer will be the Space Micro rad-hard Proton 200kTM Lite FMB model for a 2-5 year mission. Advertised features include radiation hardening utilizing Space Micro’s patented radiation mitigation technologies; 200 grams mass; low power for CubeSat missions (1.5W standard); optimized processing speed of 900 MFLOPS; DSP processor; Memory of 512 Mbyte SDRAM w/EDAC, 1 Mbyte EEPROM to 8 Mbyte (option), 8 Gb RH Flash (option); Operating System and Software Support: TI DSP/BIOS RTOS (option), TI Code Composer Studio (option), JTAG debugging support; and CubeSat PPOD compatible.
Based on the success and established framework of the AggieSat Lab at Texas A&M, the architecture for flight software will be distributed across different, multi-threaded processes. The processing workload will be split-up among the subsystems: C&DH, ADCS, COMM, EPS, and each of scientific payloads including RFT. Each process has its own defined hardware connections to different ports on the flight computer, and then communicates to each other.
The version control for flight code uses backups on a subversion server on a main Sun server with the AggieSat Lab at Texas A&M. Current working versions of the software are kept on the uncontrolled portion of the server. As for test plan, some milestones include automated system checkout routine (system health) and running through automated scripts in Day-In-Life-Testing.
Previous: Mission Concept