Steve Kemble & Bob Parkinson




Дата канвертавання22.04.2016
Памер53.74 Kb.
IAC-03-Q-3.b.08
Mars Sample Return as a Micromission

Steve Kemble & Bob Parkinson


Astrium Ltd

Gunnels Wood Road

Stevenage SG1 2AS

bobparkinson@ntlworld.com


Abstract

The return of ~200 gm of surface/sub-surface material would provide a major improvement in our understanding of Mars. With such a small return payload, sample mass is not a performance driver, and a micro-spacecraft approach would reduce mission size and cost. The possibility of such a mission is investigated using a Soyuz-Fregat launcher. Key issues are the size of the Mars Ascent Vehicle, and the V required for entry into Mars orbit and return to Earth. Optimized transfers have been developed to minimise V for the return. The benefits of solar-electric propulsion (SEP) depend on trade-offs in the system design, and high specific impulse does not necessarily yield the greatest vehicle payload. New operational technologies could be tested by adapting the vehicle design for a precursor Lunar Sample Return (e.g. from the polar Aitken Basin region), providing additional scientific return.




  1. INTRODUCTION

Mars Sample Return is seen as a key mission in the exploration of Mars. Past studies have produced complex and expensive missions demanding a heavy lift launch, or even multiple launches. But the 1970 Russian Luna 16 provided a useful scientific return from just 100 gm of regolith. On Mars, aeolian transport of surface fines means that any sample can be expected to contain grains from a wide – perhaps even planet-wide – area. The Beagle 2 “Mole” provides a means of obtaining subsurface material with the minimum of equipment. As a consequence, neither the size of the returned sample nor the equipment required to retrieve it represent limiting factors in sizing the mission.


As a possible follow-on to Beagle 2, a study has been made of performing Mars Sample Return using a Soyuz-Fregat launch. By deliberately limiting spacecraft size in a micro-spacecraft mission approach, costs might be limited by focussing on a single, scientifically important objective. The study was intended to identify key technology objectives required to define a possible mission rather than to solve all problems. Key issues are the Vs required for entry into Mars orbit and later return to Earth, and the sizing of the Mars Ascent Vehicle.


  1. MISSION STRATEGY




    1. Mission Outline

While Soyuz-Fregat represents an established, low cost, medium class launcher, progressive improvements to its launch capability are expected. Russian data presented as a part of the Bepi-Colombo collaboration identified modifications that would allow Soyuz-Fregat to inject 1334 kg or 1564 kg onto a Mars transfer trajectory. For the purposes of this study the 1334 kg initial mass was taken as the spacecraft objective, with the 1564 kg capability providing the system margin.


The sample return mission assumed that, as with Mars Express/Beagle 2, the launched assembly would include both an Orbiter and a Lander, journeying together to Mars, but separating during the hyperbolic approach. The Lander would make a direct entry into the Martian atmosphere and the Orbiter use on-board propulsion to enter a parking orbit about the planet. The Lander would provide entry protection and descent and landing systems around an Ascent Vehicle that – after having been loaded with the sample of Martian fines – would return to a low altitude (350 km) orbit about Mars. The Orbiter would then rendezvous with the Ascent Vehicle, transfer the sample container, and at the appropriate time boost itself onto an Earth return trajectory. The sample would be returned to Earth using a small, direct entry capsule carried by the Orbiter.


    1. Transfer Strategy

The design of the sample return mission aims to maximise payload and sample return mass with a very limited stay on the Martian surface. The stay duration is short, typically 10 days. The first problem to solve is an optimised transfer from Earth to Mars, followed by an almost immediate optimised return. The baseline propulsion option system is chemical, and manoeuvres are near impulsive.


For an Earth to Mars transfer, the V required depends on the relative planetary positions at injection into the inter-planetary transfer orbit. A minimum energy case exists where the transfer orbit has an aphelion close to Mars orbital radius and a perihelion at Earth radius. This optimum relationship of the planetary positions arises at a frequency given by the synodic period. For Earth and Mars, this period is 785 days. This defines the interval between minimum energy transfers. The geometrical repeat period is 7 synodic periods, at 15 years.
Exactly the same situation exists regarding optimum transfers from Mars to Earth. However, the epochs of the optimum return opportunities do not correspond to those of the optimum arrival. Compromise is therefore required to accommodate the short stay duration. An optimum solution must derive an Earth departure epoch, Mars arrival epoch and Earth return epoch that allow minimum mission V. These solutions were obtained to form the basis of the mission design. Typical characteristics of such a transfer are:



Earth Escape velocity

3.2 km/sec

Mars approach velocity

2.8 km/sec

Outward Transfer duration

220 days

Mars Escape velocity

5.8 km/sec

Earth approach velocity

4.8 km/sec

Return transfer duration

520 days

In fact, the outward journey parameters are nearly optimal for a one-way transfer. The example is based on a 2003 or 2018 launch, with a near minimum outward leg requirement for any launch epoch. The transfer duration of the return is longer than those that optimum ‘single leg’ transfers, as are the escape and approach velocities. Similar total V results can be obtained for a range of launch epochs, with a change in the balance between outward and return legs.


A more efficient return trajectory can be found by allowing a longer transfer involving approximately one and a half heliocentric revolutions. However, the return transfer now takes over 700 days. Alternatively, shorter return transfers can be found, but at the expense of V. Return trip durations of less than 400 days involve a passage inside 1 AU (in fact typically 0.8AU).
The problem is illustrated in Fig.1. The sum of Vinfnities (Mars departure plus Earth arrival), for a range of Mars departure epochs and trip times are shown for short duration stays. Both the optimal, 500 day trip solution and shorter return trips can be seen. It is generally of more importance to reduce the Mars departure Vinfinity and accept a penalty on the Earth approach speed, if aeroassisted capture options are used at Earth.

Fig.1: Return leg total Vinfinity sensitivity to stay time and trip duration


Fig.2 compares the locally optimal, direct return route with the locally 1.5 revolution return route. Total Vinfinity requirements are shown being the sum of Vinfinities for the outward and return legs. The global minimum is seen to lie with the 1.5 rev return route for short stay times. As the stay time extends to 400 days, the situation is reversed

Fig.2: Total mission Vinfinity requirements vs stay time for direct and 1.5 rev return types




    1. Orbiter Options

The orbiter is required to implement a large V, injecting to and departing Mars operational orbits. Staging offers the possibility of substantial net mass savings. This could be done using two spacecraft buses, each with its own propulsion. The first stage would insert the second stage and Mars descent composite into Martian orbit, and the second stage used to return to Earth after rendezvous with the Mars ascent vehicle.


Alternatively, greater net mass saving can be achieved using a single spacecraft bus with a separable tank system. This removes the need for additional thruster units, but retains the advantage of jettison of redundant tank mass. Such a system requires a more complex separation system but improves performance.
A further option is to use Solar Electric Propulsion (SEP) to implement the transfer, Transfer times of typically 300 to 400 days each way can be found with limited stay times at Mars. SEP can be used for orbit insertion and escape at Mars in addition to the interplanetary transfer. The high specific impulse of the low thrust system ensures low fuel loads, but a key item is the mass required for the thrusters and power generation. This later consideration is particularly relevant for solar electric systems at Mars.
The use of low thrust opens the possibilities for other interesting transfer techniques. If longer transfer durations can be accepted, use of Earth gravity assist can reduce total mission Vs by >2 km/sec. Transfer times increase by over a year on the outward leg


    1. Earth Entry Vehicle

For a minimal mission, direct entry return to Earth is necessary. The Japanese Muses-C return capsule, performing a very similar mission, has a mass of 25 kg [AWST May 19 2003, p. 40]. For the Mars Sample Return a simpler (but slightly heavier at 40 kg) Earth Entry Vehicle has been conceived (Fig. 3) with a low entry ballistic coefficient limiting the terminal descent through the Earth’s atmosphere to low speed (~22 m/s) without the use of a parachute, relying on crushable protection to cushion the sample at impact.


Fig. 3: Passive Earth Entry Vehicle Concept





  1. LANDER/ASCENT VEHICLE




    1. Mars Entry, Descent & Landing

The Mars entry heat shield (see Fig.4) has an identical ballistic coefficient (152 kg/m2) and geometry (30 sweep angle, 0.417 m nose radius) to Beagle 2. With an entry mass of 187 kg, the heat shield diameter is 1.480 m. The leeward aeroshell cone has to be proportionately taller than Beagle 2 to accommodate the ascent vehicle, but will still be entirely within the wake region of the entry heat shield. After entry into the Martian atmosphere, the heat shield and entry aeroshell separate to allow deployment of the parachutes. For the MSR a cluster of 3 parachutes is proposed to improve packaging within the entry capsule. Deployment of the main parachutes takes place at a descent speed of 89 m/s. The terminal descent speed, as for Beagle 2, will be about 16 m/s.


Surface impact attenuation assumes the use of deflating airbags, as opposed to the “bouncing” airbag system used on Beagle 2. Release of the system from the parachutes occurs at ~50 m altitude, with air-bag sizing and impact velocity designed to avoid the vehicle toppling even at the maximum expected lateral drift velocity.
The descent assembly carries equipment for the following surface operations:

Fig. 4: Stowage of the Ascent Vehicle within the entry capsule.




      • A “Mole” for collecting the sub-surface sample

      • A robot arm to deploy and retrieve the “Mole”, and to load the collected sample into the Ascent Vehicle

      • A camera

      • A UHF transceiver for communication with the Orbiter

      • An inclinometer, to establish the local vertical with respect to the ascent vehicle co-ordinates.

All but the last are directly derived from Beagle 2. The robot arm has sufficient reach to extend past the deflated airbags and obtain a sample from a clear Martian surface.


Power for surface operations is provided by LiSOCl2 primary batteries, sized to provide power to operate the Lander for 4 sols after touch-down. This is judged sufficient to collect the sample. Use of deployable solar arrays as in Beagle 2 was ruled out due to packaging and deployment problems.


    1. Ascent Vehicle

The two-stage Ascent Vehicle uses a simple bi-propellant propulsion system for the booster stage, and a monopropellant (N2H4) orbital stage. The task of the Ascent Vehicle is to propel itself into a low Martian orbit, and to maintain itself there as a co-operating target for Orbiter rendezvous. Mass performance is critical, since the vehicle must achieve a V of about 3850 m/s, with a launch mass of about 91 kg.


Because the Ascent Vehicle launches from the Martian surface, the booster stage is a single-burn, pressure fed system without the need for propellant retention devices. The booster stage uses a fixed 500 N liquid engine and 4 x 22 N RCS steering thrusters. The small size of the ascent vehicle makes the resulting vehicle quite agile, with adequate control for the ascent burn.
The tiny upper stage (~18 kg at separation) uses 4x10 N thrusters offset at 7 to perform the orbital insertion. The stage then deploys a ~30 w solar array and maintains a 3-axis stabilized, sun-pointing attitude until the Orbiter makes its rendezvous and retrieves the sample container for return to Earth.
The sensitivity of the mission to Ascent Vehicle inert mass means that the design must avoid “fixed mass” items such as bolted joints, connectors and individual equipment boxes, which at this small size become dominant features in the mass budget.


    1. Ascent Vehicle/Orbiter Operations

For Lander surface operations, the Orbiter will act as a communications relay to Earth. This drives mission design immediately after arrival. With a low thrust LAE the Orbiter will not be able to enter a low Mars orbit efficiently with a single burn. The first burn will achieve a 400 km x 33753 km altitude orbit with a second burn one Martian day (sol) later, placing it in a 400 x 4000 km orbit with a 3.52 hour period. Communications with the Lander cannot begin until the start of sol 2. The objective then is to determine the sampling site during sol 2, with Mole operations extending into sol 3, ending with the stowage of the sample within the Ascent Vehicle and launch to Mars orbit. Sol 4 provides a margin for surface operations.


Once the Ascent Vehicle has achieved orbit, the Orbiter must accurately determine the orbital elements of each vehicle to carry out the rendezvous and capture. A variety of possibilities exist for this, including use of the star trackers with Mars occulation to determine the orbits of each vehicle, active laser range-finding by the Orbiter, or even direct optical acquisition. Fortunately time is not a critical constraint at this stage. By starting in an initially elliptical orbit and descending to the 350 km circular orbit of the Ascent Vehicle the Orbiter can minimize the effects of differences in the orbital planes of the two vehicles by performing combined burn manoeuvres.


    1. Mass Budget

Table 1 shows an initial mass budget for the mission. The present design lies within the projected capability of the Soyuz-Fregat launcher. However, the system is sensitive to a number of design assumptions. Because the Lander separates from the Orbiter before Mars entry, while mass growth factors in the Lander are significant, mass growth in the Lander does not consequentially impact the Orbiter design. Elsewhere, mass growth factors are less significant than specific impulse or V requirements.







Mass (kg)

Mass at Launch




1240

Lander at Separation




187

Entry & Surface Element

97




Ascent Vehicle Dry Mass

28




Propellant Loaded

62




Orbiter at Separation




1053

Arrival Module Dry Mass

75




Arrival Propellant

518




Return Module Dry Mass

77




Departure Propellant

343




Earth Entry Vehicle

40




Margin on Soyuz-Fregat Capability




324

Table 1: Mass Budget for Mars Mission


  1. TECHNOLOGY DEVELOPMENT / PRECURSOR MISSION

The MSR mission involves a number of critical operations that have not, or will not have been done before, specifically:




      • Relaunch of a sample from a planetary body

      • Rendezvous and sample transfer in orbit

      • Recovery with a passive re-entry vehicle at hyperbolic velocities.

A preliminary “demonstration” mission is therefore advisable. An interesting possibility is to adapt the systems and equipment for a lunar sample return mission (possibly in the lunar polar region), demonstrating the three critical aspects close at hand where near-real-time communications are possible. A precursor mission of this sort would clearly have a scientific value in its own right.


The Lander would need to be modified to descend using rocket braking, replacing the entry and descent systems with a hydrazine monopropellant propulsion system (as a cheap and reliable development). The Orbiter would now require only a small V capability (~800 m/s each for entry and departure), which could be accommodated with the return capability of the Mars Orbiter. In this instance it is not necessary to have a separating arrival stage. The resulting mass budget for the lunar mission is shown below in Table 2.





Mass (kg)

Mass at launch




1122

Orbiter at launch




449

Orbiter dry mass

154




Orbiter propellant

295




Lander at Separation




680

Descent stage dry mass

130




Descent propellant

479




Ascent vehicle dry mass

41




Ascent propellant

30




Table 2: Mass Budget for Lunar Mission



  1. CONCLUSION

Use of a micro-spacecraft approach to a Mars Sample Return mission suggests the feasibility of returning 200 gm of surface and sub-surface fines with a spacecraft assembly within projected Soyuz-Fregat launch performance. The projected mission would use direct entry of the Lander into the Martian atmosphere, following the approach adopted by Beagle 2, followed by Mars orbit rendezvous, and direct entry into the Earth’s atmosphere by the returning vehicle.


The design studied has a system level margin of 26% on the maximum projected capability of Soyuz-Fregat, but has some mass sensitivities due to the high V requirement of the returning vehicle. However, the V requirements are subject to the stay time at Mars and other mission design features.
Key operational technologies not yet demonstrated (or not demonstrated in the Mars Express/Beagle 2 mission) include re-launch of a sample from a planetary surface, rendezvous and sample transfer in orbit, and passive re-entry return to Earth at hyperbolic velocities. It is suggested that the Mars Sample Return hardware could be modified with limited effort to achieve a lunar sample return mission that would both demonstrate the requisite capabilities and also fulfil a valuable scientific objective.


База данных защищена авторским правом ©shkola.of.by 2016
звярнуцца да адміністрацыі

    Галоўная старонка