≡ Menu

Interstellar (Precursor) Mission & Vehicle Design

by Marc Millis

Tau Zero’s first graduate student project has been completed. Berkeley Davis, a 2nd Lt. at the United States Air Force Institute of Technology, Dayton Ohio, completed his Masters thesis on a deep space probe to perform Claudio Maccone’s gravitational lens mission (FOCAL). For those unfamiliar with FOCAL, it is a mission to utilize the gravitational lens effect that begins at approximately 550 AU from the Sun, one that in the view of Maccone will offer huge magnifications for the study of targets like the Cosmic Microwave Background. For more, see the Centauri Dreams archives.

Maccone, Deep Space Flight and Communications: Exploiting The Sun as a Gravitational Lens (Springer, 2009).

Davis, Berkeley. R. (2012) Gravitational Lens: The Space Probe Design (Thesis), AFIT/GA/ENY/12-M06, Air Force Institute of Technology.

To provide a realistic baseline on what is possible, the student was asked to constrain his design to commercially off-the-shelf technology. The mission involves taking a 12 m diameter radio telescope beyond 550 Astronomical Units (AU), continuing outward thereafter, to examine the gravitational lensing of our own Sun. A secondary mission, which occurs before that point, is to measure the magnetic fields, particles, and dust while traveling through our Solar System and the transition through the edge of our Solar System (the termination shock, the heliosheath, heliopause) and into true interstellar space.

Image: Beyond 550 AU, we can start to take advantage of the Sun’s gravitational lens, which may allow astrophysical observations of a quality beyond anything we can do today. Credit: Adrian Mann.

In short, it is found that this mission could be performed with EXISTING technology for roughly $3B-$5B (FY 2011 dollars), and that it would take roughly 34 years to reach the edge of our Solar System, and roughly 110 years to reach its primary mission point of 550 AU, and continuing thereafter for almost 80 years of data taking until the spacecraft reaches about 1000 AU, where it will have likely exceeded its 2-century life-time projection.

Using those goals and constraints, the student designed a two-stage vehicle that is delivered into orbit by a “Delta IV-H/Star48/Star37” launcher. The 1st stage, which has a 22 kW solar array, uses four “NEXT” ion thrusters to propel the vehicle from Earth orbit to Jupiter while thrusting almost constantly in a spiral trajectory for 17 years.

At Jupiter that boost stage is jettisoned, and the main stage completes a Jupiter gravity assist. The main stage also includes four “NEXT” ion thrusters powered by 20 Radioisotope Thermal Generator (RTGs) which have roughly 4.4 kW at this point in the mission. The spacecraft thrusts continuously for roughly another 17 years until the propellant runs out at roughly 90 AU. At this point its velocity is 6.7 AU/yr which is almost twice the speed of Voyager (3.6 AU/yr).

For the next 20 or so years it coasts through the border between our Solar System and true interstellar space, taking data for the secondary mission. Then after about 55 more years, it reaches 550 AU, the closest point where gravitational lensing would ideally begin. By this point its velocity has slowed to 6.2 AU/yr. It takes another 12 years to reach 625 AU, which is the closest predicted for observable signals at the focal point. The spacecraft will continue coasting outward for the next 60 years and will be able to continue taking data (observations of our Sun’s gravitational lensing) until the spacecraft passes 1000 AU roughly 180 years after launch. Provisional estimates of the number of cycles of the attitude control system, processor, etc, suggest that this vehicle might function for 2-centuries.

The thesis also made the following recommendations:

– Since spacecraft power is the most limiting technological factor at this time, that should be the focus of next-step research for interstellar missions.

– To complete such missions with RTGs, the production of Plutonium-238 would have to fully resume.

– Mission durations are longer than practical for on-Earth life tests, so novel testing techniques will need to be created to ensure the spacecraft will still be functioning by the time it reaches its interstellar mission location.

– Note: Mission uses about 10% of total annual Xenon production – Xenon which will not return to Earth.

This study was merely a first-cut at examining these possibilities, and these findings should not be considered the last work on this specific topic.

1. Stelio Montebugnoli10. Giovanni Vulpetti
2. Ed Belbruno11. Maria Sarasso
3. Jean Heidmann12. Rinaldo Bertone
4. Jorg Strobl13. Franco Palutan
5. Gregory Matloff14. Vittorio Banfi
6. Ettore Antona15. Mario Pasta
7. Constance Bangs16. Federico Bedarida
8. Renato Pannunzio17. Luciano Santoro
9. Sigfrido Leschiutta18. Claudio Maccone


Boost stage

4 Ion Thrusters = 225 kg (615 W – 7.2 kW ea)
Xe tank = 309 kg
Xe load = 2996 kg
Solar Power 900 kg, 22 kW
Thrusts from LEO to Jupiter
Jettisoned at Jupiter

Main Stage

4 Ion Thrusters = 225 kg (615 W – 7.2 kW ea)
Xe tank = 217 kg
Xe load = 1888 kg
RTG Power: 20 General Purpose Heat Source Radioisotope Thermal Generators
Each 58 kg, 246W initially, with half-life of 90 yrs
12 Attitude control thrusters, 0.8 kg ea
Attitude Control Propellant Tanks = 12 kg
Attitude Control Propellant = 151 kg
Science Payload, 51kg, 40W
• 12 m radio telescope (doubles as communication High Gain Antenna)
• Magnetometers
• Particle detectors
• Dust detection

Thrusts from Jupiter to 200 AU (runs out of propellant)
Coasts from 200 AU outward
100 kBit/sec @1000AU


Comments on this entry are closed.

  • Daniel Highmountain March 30, 2012, 11:46

    Hm.. we will fly by this mission after 50-80 years, with a much better system, which in turn will be passed by yet another mission after 100 years.. So.. I dont know..

  • Adam Crowl March 30, 2012, 12:34

    Nice design.

  • Paul Gilster March 30, 2012, 13:01

    Daniel, the idea isn’t to fly this particular mission — it’s a design study. The idea is to figure out the parameters of a mission like this if it were to fly with the hardware we have today — it’s an entirely theoretical study that points to today’s state of the art and helps us see what needs to be improved as technology matures in various areas. We all hope to see a FOCAL mission launched that takes less than centuries to get to its destination, but this study shows how steep the challenges are.

  • Tom W March 30, 2012, 14:25

    Interesting study. Would be interested to know how the Falcon Heavy affects travel times. I realize it has not flown yet but should have within the next year. You could double the size of your first stage from 4 to 8 Ion thrusters as it can put double the payload of the Delta 4 H into LEO. You wouldn’t need more plutonium as the first stage is fully Solar powered. Seems like the easiest next step to lower journey times.

  • Eric Rooney March 30, 2012, 14:27

    Cool study. I think some computer simulations are in order. And as he mentioned, all the testing would have to be done using “novel testing techniques”.

  • FrankH March 30, 2012, 15:05

    It’s a cool design study, but at the same time, somewhat depressing; with our current technology it would take 110+ years to reach an operating distance of “only” 550AU.

    Orion is still the only currently feasible propulsion system that could complete the mission in a relatively short period of time.

  • Marc G Millis March 30, 2012, 15:59

    The point of this study was to determine a REALISTIC performance baseline (expected lower bound) on what could be done with EXISTING technology, stuff that can be bought and assembled without further development. We’re trying to inject reality checks here. Although this study’s numbers merit scrutiny, it does set the example to beat.

    For those promoting other solutions, it would be nice to have your system-level studies identify the weakest links (perhaps ‘real’ magnetic nozzles, effective heat rejection, or energy supply accumulation). The entire vehicle is only as ready as its weakest link. If we know, reliably, what those weakest points are, then we know where to have research engineers concentrate their attention.

    Ad astra incrementis,


  • Alex Tolley March 30, 2012, 16:11

    I also wonder if a better solution isn’t available “off the shelf”. The route uses a single gravity assist (Jupiter), could it do better with multiple gravity assists, possibly in the inner solar system?

    If nothing else, the mission is far too long duration to be feasible, none of the scientists would even be alive when it reaches its first milestone distance of 550 AU. Who would work on the project, what agency would even fund such a long term mission? What this does emphasize is the need for new, high performance propulsion systems just to reach these modest interstellar distances. OTOH, maybe this indicates that the gravitational lensing experiment is not as workable as other telescopes for the same performance, such as telescope swarms that could combine their signals.

  • Angelo Genovese March 30, 2012, 16:19

    Very interesting study. In particular I agree with the first recommendation about the power source, RTGs have too high specific mass (235 kg/kW) for reasonable extra-solar missions. Presently, the only power source that can make the job is an advanced nuclear reactor with a specific mass of 25 kg/kW or less. This powerful source can be coupled with advanced high power ion thrusters like the NASA´s HiPEP thruster with a specific impulse close to 10,000 s, which is more than double the NEXT specific impulse (4000 s); this will greatly reduce the propellant needed. This way we get a significant reduction of the mission time.

  • Rob Henry March 30, 2012, 17:06

    I am curious as to what need there is to synthesise an isotope (Pu-228) with a half life of just one second. If the probe manufactures it, what does it use it for?

  • Damien RS March 30, 2012, 19:29

    “It’s a cool design study, but at the same time, somewhat depressing; with our current technology it would take 110+ years to reach an operating distance of “only” 550AU.”

    With our current technology and a particular budget, $3-5 billion here. What could an extra $5 billion in power/propulsion buy? I don’t know, but it’s possible the answer wuold be more cheering.

  • mike March 30, 2012, 20:05

    is the effective activity of the mission beyond civilization’s window to receive data? some will say that technology will improve and surpass a mission launched today. what if civilization collapses and no mission is sent ? what if a mission is sent and civilization collapses?

    the mission would become a derelict.

    look at the price of gasoline and other fossil fuel inputs. i dont want to mention deep water horizon but there is a rig in the north sea leaking natural gas and predicted to explode if a relief well isnt started right now.

    what would world war 3 do to mission time tables? famine? pandemic? financial collapse? ecological disruption? CME? all these things seem about to happen. any two would pretty much put an end to space exploration.

    any group that wants a mission of centuries should better study the society
    it lives in to determine if that society will last long enough to get a return of data on the investment made.

  • lepton March 31, 2012, 2:02

    1. What single object is that important to justify a mission like this?
    2. Can we afford multiple launches?
    3. Is this even the best way to spend the money and effort?

    I personally would rather see a HUGE segmented free floating ST.

  • bill March 31, 2012, 3:10

    @Rob Henry

    “I am curious as to what need there is to synthesise an isotope (Pu-228) with a half life of just one second. If the probe manufactures it, what does it use it for?” should be Pu-238

  • Thomas Römer March 31, 2012, 6:08

    @Rob Henry: Guess that’s supposed to be Pu-238 with a half-life of 87.7 yrs and a good power density. Wonder if they calculated the thing for Am-241 (4x half-life, 1/4 power density compared to Pu-238) or other possible RTG power sources.

  • Paul Gilster March 31, 2012, 8:34

    The Pu-238 typo is now fixed in the text.

  • Andrew Higgins March 31, 2012, 9:39

    @lepton: “1. What single object is that important to justify a mission like this? […] I personally would rather see a HUGE segmented free floating ST.”

    The increase in light-gathering achieved by positioning a telescope at the focus of the sun’s gravitational lens is something like 10^8.

    You would need to increase the diameter of the primary of a space telescope by a factor of 10,000 to obtain a similar increase in light gathering ability. Of course, the huge ST can look any direction, while FOCAL only gains by looking back along the line passing through the sun.

  • Eniac March 31, 2012, 14:34

    Very nice study. Here is what I would suggest for potential improvements (if not already discussed in the full text):

    1) Consider a trajectory going closer to the sun first, where more solar power and the Oberth effect make ion propulsion much more effective.
    2) Raise Isp to optimize flight time with given constraints
    3) Use reactors instead of RTGs (as others have said)
    4) Use propellant other than Xenon
    5) Save some propellant to adjust trajectory at ~ 600 AU to be exactly radial to allow stationary observations. Best direction might be opposite to galactic center.

  • A. A. Jackson March 31, 2012, 15:45

    Reading this reminded me of a little footnote about Einstein.
    Just finished Lawrence Krauss’a new book “A Universe From Nothing.” (Cosmology is really mind blowing these days!)
    Krauss relates something I knew and something that has been recently found.
    In 1936 an amateur astronomer, Rudi Mandl, wrote Einstein about using a star’s gravitational field as a lens (We , of course, know about the famous bending of light measurement.) Einstein made a little calculation and sent a short paper to Science, “Lens-Like Action of a Star by the Deviation of Light in the Gravitational Field.” (It’s on the web.)
    Turns out Einstein knew the answer to the question in 1912, before he even presented The General Theory of Relativity. Did not publish the result until 1936 and might never have , except for giving a guy a friendly answer.
    (Others, in between, had also made the calculation.)
    That would make this year the 100th anniversary of gravitational lensing.

  • Peter Popov March 31, 2012, 16:26

    Well, this study clearly demonstrates that
    1. Ion thrusters are dreadfully heavy
    2. RTG have terribly low power density
    3. The resulting flight time is way too long.

    This scheme looks way to complicated for a modest goal: 2 tons of century old electro-mechanical equipment, running on a kilowatt of power. If one spends a century building telescopes on the Moon, the results would probably be much more impressive.

    I wonder if this study did check out (and disprove) other obvious options:
    * Minimize equipment mass
    * Use multiple gravity assists
    * Use a chemical stage in Jupiter’s gravity well
    * Try the above close to the Sun

    Has anyone seen a thorough study on using chemical propulsion just above Jupiter’s atmosphere to get a high escape velocity?

    If the goal is just to test how gravitational lensing works, one could probably build an electromagnetic accelerator around the moon, in say 20-30 years, and then just shoot out small integrated sensor packages at high velocities, in arbitrary directions :-) You could justify a moon base along the way …

  • Joy March 31, 2012, 22:22

    I know that it didn’t get far, but circa 1987 JPL was studying a “Thousand AU” (TAU) mission concept, It was supposed to use an ion drive powered by a high energy/mass 1 MW fission reactor (AFAIK never built) and have a hyperbolic solar departure velocity of 20 AU/YR. So it would have reached the mission goal in “only” 50 years. Even that ambitious programme had limited appeal to funding agencies.

  • Nathan April 1, 2012, 0:21

    I would love to see a optical telescope at the focal point to look at a extrasolar planet.

    Paul, do you know someone who has worked out the best resolution possible one can image a extrasolar planet? I.e. taking into consideration pointing stability of telescope and so on.

  • Tom W April 1, 2012, 6:26

    Marc, is there going to be a further study to propose near-term solutions to improve journey time? It would seem the next logical step after setting the baseline with this study.

  • Michael Spencer April 1, 2012, 7:26

    We’ve seen many papers and discussion about the gravitational lens at 550AU, discussions that bring up a few questions: what is the resolution of such a device? and, what is the field of view? As the device moves further away from the sun, what happens to the images? How long would the thing be at a useful station?

    Would such a device show us the blurred images that we see from Hubble and other sources with multiple images (and a common center), requiring explanation of professionals, or would we see ultra-crisp images accessible to the layman?

    I wonder, too, about data transmission speeds and bandwidths from sources so far away.

  • Paul Gilster April 1, 2012, 13:49

    For those who’ve been asking about the resolutions possible with a FOCAL mission, I need to check with Claudio Maccone for the answer, and bear in mind that there is controversy about just how useful an image we could obtain — there are those who doubt the lens mission would be workable in the first place. Maccone, of course, champions the idea and thinks we would be able to get extremely high quality data. The communications issue that Michael asks about above is always a problem for deep space probes but probably solvable (by the time any FOCAL mission might actually be launched, which means decades from now) by evolving laser methods. But yes, comms are tough and get to be more and more of a problem as we push further out. Lesh’s work on communications via laser from an Alpha Centauri probe is to the point:


  • Ole Burde April 1, 2012, 15:25

    In the framework of a masterthessis this is an impressive work , and it is a good way of making a real contribution to the exploration of space.
    Noboddy should expect a thessis to be the final blueprint for anything , but rather to be a shortcut to establishing the first stage of a working hypotesis in a certain area , SOMETHING to start improoving on .
    The very long duration of the project is not necesarily a vaste of time either , because the challenge of long duration missions will have to be investigated , not as an irritating sideeffect , but as one of the major bottlenecks to the whole idea of interstellar flight . Any kind of a try to design and build equipment that can last for centuries is urgently needed to get started , because however “smart” the testing procedures may become , it will never be the same as the real endurance test of Time itself , with all the infinite possiblities of unforeseen sideffects of unforeseen sideffects of…

  • Eniac April 1, 2012, 16:27

    Paul, do you know someone who has worked out the best resolution possible one can image a extrasolar planet? I.e. taking into consideration pointing stability of telescope and so on.

    Here are some more posts from the archives with relevant comments on the subject (yes, some are mine) :-)


  • anooo April 2, 2012, 0:45

    This study just proves that there will be no actual/real interstellar mission (unlike voyager which turned into interstellar mission after finished examining neptune) until better and more efficient ion thruster discovered.

  • ljk April 2, 2012, 8:42

    Alex Tolley said on March 30, 2012 at 16:11:

    “If nothing else, the mission is far too long duration to be feasible, none of the scientists would even be alive when it reaches its first milestone distance of 550 AU. Who would work on the project, what agency would even fund such a long term mission?”

    I think there is merit in working on a project that will last beyond ones years and benefit future generations. This is the same reason one has children, to have a part of ones self live beyond us.

    By the way, it is about 272,000 Astronomical Units to the Alpha Centauri system, just to put in perspective how much further we have to go in several senses to become true interstellar explorers.

  • ljk April 2, 2012, 8:56

    mike said on March 30, 2012 at 20:05:

    “is the effective activity of the mission beyond civilization’s window to receive data? some will say that technology will improve and surpass a mission launched today. what if civilization collapses and no mission is sent ? what if a mission is sent and civilization collapses?

    the mission would become a derelict.”

    LJK replies:

    By that thinking we should not bother to do or build anything, as civilization could collapse at any time for multiple reasons. Heck, a couple of planetoids big enough to take out a city just passed by Earth at celestially close distances this weekend. A few parameter changes and someone would be having a very bad day.

    I recall reading the written memoir of one physicist who lived during the Cold War who would look at construction projects and think why bother since the Soviets would probably be bombing it into radioactive dust at any time. I consider it no small feat that World War 3 did not happen back when both sides had tens of thousands of nuclear weapons each and many were just itching to use them.

    Mike then says:

    “any group that wants a mission of centuries should better study the society
    it lives in to determine if that society will last long enough to get a return of data on the investment made.”

    LJK replies:

    Good idea. Perhaps if humanity ever becomes a big more enlightened we might focus on long interstellar journeys as reasons to keep society intact to allow future generations see what such missions might find out there.

  • Brasidas April 2, 2012, 9:19

    It seems it would be somewhat disappointing to spend all that time and money, go all that way, and only be able to look along the line passing back through the sun. Are there options for putting the craft in a path that moves within the focal shell, but allows viewing more of the universe as the craft moves? For example, would it be possible to maneuver the craft into an orbit of the sun at the focal distance? If its too far, would it be possible to accelerate the craft along a path that is tangential to the focal sheel, but maximizes the distance it remains within the focal shell? Could a little extra fuel allow the craft to change directions once it reaches the focal distance, so that it is moving tangent to the focal circle within the focal shell, and possibly to change directions each time it reahces the edge of the focal shell to keep it within the shell? Or are we now talking about additional centuries of mission time, making this impractical?

    Also, would the focus be through a single line passing back through the sun, or would it be more of a hollow cylinder around the sun, taking advantage of the gravitational lensing on all sides of the sun?

  • Paul Gilster April 2, 2012, 9:37

    Brasidas, at our current stage of technology, just getting the FOCAL mission out past 550 AU is quite a stretch (and actually, we have to go further to allow for coronal effects). Early FOCAL missions would not be capable of some of the trajectories you’re talking about, so we’d more likely be thinking of using multiple missions for specific targets once we’ve proven that the technique works and have found propulsion strategies that fit.

    The key will be to choose the target well, as FOCAL assumes the ability to image what is directly on the other side of the Sun from the spacecraft. If the goal is to study the Cosmic Microwave Background, the situation is eased, but if we’re trying to study, say, a nearby star system, then the need for precise targeting is emphasized. A useful feature is that the focus extends to infinity, so the craft continues to make observations well past the initial focal distance. Maccone’s book goes into the background of all this.

  • Mike Lockmoore April 2, 2012, 13:00

    Peter Popov: Regarding the “electromagnetic accelerator”, I’ve imagined a similar initiative, though in my version an accelerator ring would be in free space, perhaps the L2 point. Nano-probes could be accelerated to perhaps several percent of c. They would not carry propellent, but would carry ultra-miniaturized power, comms, and science payloads. If a “stream” of these are fired in a well-defined pattern, they could form a moving communications network (bucket-brigade style) and could cover many points of the focal shell that would allow us back home to build up interesting images. Maybe serve as a breadcrumb trail for later human missions.

    If each probe masses 1g, then we could have million of them launched to collect the data for a nice 1000×1000 image (and undoubtedly additional data as well) and the launcher would only need to spend the energy to accelerate a metric ton of material. And the launch rate of the probes could be managed to keep within a reasonable power budget for the launch accelerator ring.

    And if the situation ever warrants, the launcher could be re-purposed as an Earth-Defense pea-shooter to deflect a wayward asteroid or comet or even take out an interstellar invasion fleet . ;-)

  • jkittle April 3, 2012, 9:06

    nice study and as a baseline it is superb. I think one bottom line is that exploratoin of the ENTIRE solar system is now within reach using this technology, a point drivien home byu the new horizon mission. This is not a trivial fact. Use of thorium or enrichedd uranium reactors to send probes around the solar system and seems very plausible. We are still decidingon targets but an extended mission out to look at the Trans neptune objects , sort of like the DAWN mission might be pretty useful. it turns out these iceballs are nto all the same contrast Pluto with Hamea with Eris, each of which have very different properties.
    One question is , if we Tweak this technology to use lighter ( thus more efficient ) ions than Xenon, and use nuclear reactors instead of RTG,s can Ion drives support Manned exploration out to Mars, the asteroid belt and perhaps to the Jupiter Trojans? With a vehicle that has a large balloon habitat and an area with centrifical ” artificial ” gravity, can mission be extended to 10 years?

  • Marc G Millis April 3, 2012, 10:23

    In reply to some comments:
    – This study was CONSTRAINED to EXISTING technology to find out what the results would be. It is NOT a mission proposal. It is a sanity check on our current ability.
    – To all those who suggest revisiting this with different technology or different trajectories, PLEASE DO SO! I have no plans for a sequel since this was just to determine contemporary situations. I have the data points needed, the lower bound (so to speak).
    – For those who do want to follow this with changes in assumptions, PLEASE use realistic values (not salesmanship) when conducting your assessments and then publish those findings for the rest of us to ponder.

  • Marc G Millis April 3, 2012, 11:27

    OOPS – Found another typo:
    In two places, fix this error:
    4 Ion Thrusters = 225 kg (615-7225 kW)

    Should be this (change the power data):
    4 Ion Thrusters = 225 kg (615W-7.2 kW ea)

  • Paul Gilster April 3, 2012, 13:31

    Typo now fixed in the text.

  • Eniac April 4, 2012, 7:08

    As for targeting, A FOCAL type telescope will never be freely pointable, because it would take huge velocities to move around the sun by any substantial angle. This is not as much of a disadvantage as it would seem at first, though, because the very high resolution will provide plenty of data from a single spot, perhaps more than can be practically shipped home.

    As Paul says, the microwave background has the advantage of being interesting everywhere, reducing the requirement for pointing accuracy. The same would go for the galactic center, I imagine, or any nearby galaxy. Extrasolar planets would make a great target, but here we have to not only point accurately, but also follow the planet’s orbit and the star’s proper motion. This is possible, but requires propellant and is likely too challenging for a first mission.

    The diffraction limit would in principle let us see details on nearby extrasolar planets down to just a few meters, if not less. However, because the aperture is a ring, not a full circle, and the focus is a line, not a point, there will be much less light than there would be with a regular telescope of equivalent size. I have not seen an analysis of whether the reflected light available from a planet and focused by the sun would have sufficiently high photon counts to observe the planetary surface in any detail. Has anyone?

  • ljk April 4, 2012, 9:51

    If FOCAL can be equipped with a radio telescope, it could also pick up an amazing amount of cosmic signals greatly amplified, if I recall reading about this correctly. Radio SETI at its near ultimate.