Mars Sample Return Options

[John Whitehead]

A decade ago in this forum I offered my perspectives on the Mars ascent vehicle (MAV) challenge (first post #56 on 2007Sep19, last post #205 on 2008Jul14). Now seems like a good time for an update considering that Mars Sample Return has been in the news lately. Recent science results have strengthened the case for doing MSR. In April 2018, NASA and ESA signed an agreement to collaborate on MSR, and there is a hope that the 2026 launch opportunity can be used to go and get the samples to be collected and cached by the Mars 2020 rover.
https://mepag.jpl.nasa.gov/announcements/20...%20(Signed).pdf

For MSR mission planning, a huge unknown is how small a MAV "can be," or conversely how large does a MAV "have to be" in order to reliably reach Mars orbit with an appropriate payload mass. The answer strongly influences mission scale and cost. The assumptions used in mission planning are necessarily arbitrary, because MAV size depends very steeply on the propellant mass fraction of tiny rocket stages, the likes of which have never been built. Over the past ten years, studies and research have continued, but the big picture has not changed. During this time, I no longer have done rocket development professionally, but I have not given up on efforts to raise awareness of what is missing.

Here is my assessment of the situation. There are mission planners whose comfort zone includes the notion that propulsion is straightforward, as has been the case for essentially all planetary science to date. There is a lot of refined expertise for building spacecraft propulsion systems, using technology very much like satellites have, typically by buying engines, tanks, and valves, then designing and assembling unique new systems in a modular fashion by connecting these flight-proven components with metal tubing. An excellent example of such modular propulsion is the hydrazine system used for Mars Science Lab (MSL) entry descent and landing (EDL) in 2012. Unfortunately, all such components are way too heavy to build a MAV. Separately, there is a propulsion research community that focuses on propellant chemistry, propellant material properties, propellant mixing with ignition and combustion efficiency, and the shape of rocket nozzles. All this is important for obtaining a high exhaust velocity. University teaching in rocketry is mostly limited to the topics listed here.

So, what is missing? There is no community of engineers who focus on building components far less heavy than on satellites. The people toward the top of the org charts seem generally unaware of what is missing. During the past ten years, money allocated for MAV has been spent on propellant research, including one scheme that had fuel and oxidizer mixed together in the same tank, and another line of research for a new solid fuel to be burned with a liquid oxidizer (hybrid propulsion). The main advantage of the hybrid propellants has been explained as long-term cold storage on Mars without a lot of insulation and heating energy, but can it be made lightweight? Research papers say very little about how heavy the parts "have to be," or how lightweight they "can be."

The social system phenomenon of ignoring the importance of lightweight parts is widespread, not unique to MAV. In the early 1990's, the DC-X was advertised as "SSTO technology," while the parts summed up to a mass that was ten times too heavy for it to reach orbit.
https://en.wikipedia.org/wiki/McDonnell_Douglas_DC-X
One NASA friend of mine at that time referred to the vertical takeoff and landing tests as "all the performance of a hot air balloon." Yes of course the DC-X inspired many people, a good thing, but a SSTO vehicle was never built.

For the MAV, perhaps one major positive step during the past ten years is that a 2-stage solid rocket is no longer being promoted at NASA as the most likely workable solution, as occurred from about 1999 to 2012. My trajectory analysis published in 2005 (Journal of Spacecraft and Rockets) showed that tiny solid rockets burn very quickly, accelerating too rapidly at low altitudes, hence extra atmospheric drag, even in the thin atmosphere of Mars. A JPL paper published in 2016 echoed my comments about solid rockets. They included the fact that excess thrust means that the steering components would also be relatively heavier than for optimum thrust, the latter pointed out in my 1997 AIAA paper.

For a comparison regarding schedule, most of a decade was required to develop the MSL EDL propulsion system, using well-known flight-proven technology (including updated Viking engines). Considering the modular methods for conventional space propulsion (bend tubing to connect the parts as necessary), the physical configuration of the components was not intimately tied to the overall system design. For a MAV, brand new components have to be custom-designed for a brand new propulsion system, with the parts and the system much more tightly linked together as one engineering problem. So, does it make sense that now there is a notion that a MAV will be sent from Earth to Mars as early as 2026?
https://spacenews.com/esa-awards-mars-sampl...ans-take-shape/

It is widely appreciated that it is cool to be an astronaut, and cool to work on Mars rovers. There should be advertising to the effect that building a MAV is going to be one of the coolest engineering jobs over the next ten years, so why is there no such message?

There are two different kinds of difficulty faced by engineers, one being project complexity, and another being physical limits. Complexity can be managed, and it can be overcome by money, time, and teamwork. It is harder for an organization to manage its way past physical limits. Two kinds of physical limits determine what is possible with rockets as we know them. One is the energy in chemical bonds relative to the mass of molecules, which determines rocket exhaust velocity. The other is material strength-to-weight ratios, which determines whether a rocket stage can be over 80 percent propellant (MAV) or less than 70 percent propellant (good enough for lunar landing). Large launch vehicles crack the mass nut by running tanks at very low pressures, and pumping propellants to thrust chambers at much higher pressures. NASA publications during the past ten years have mentioned some work on different kinds of pumps for MAV, but little is offered in the way of explanation for why this is not being pursued more. My guess is that not enough people are focusing primarily on the question of how heavy the parts are going to be.

The TRL scale is used widely by NASA to describe the "Technology Readiness Level" for new technologies in development. It seems crazy to me that the TRL scale does not make any formal mention of being lightweight enough to fly, versus being too heavy to fly. My guess is that the people who wrote the TRL scale tended to think of engineering functionality as a separate thing from how heavy the implementation is. Their assumption may be appropriate for a new electrical circuit design, in which mass is all about packaging rather than the electrical functionality.

There is the notion that a successful Mars 2020 mission will greatly increase the priority and funding for developing a MAV, because Mars 2020 will collect samples and seal them up in specialized containers that superficially could be described as titanium test tubes. Conversely, however, Mars 2020 ideally would itself include a MAV. Unfortunately for MAV development, the can has been kicked down the road for decades. Still lacking MAV technology after all these years, leaving the geologists' selected samples lying on Mars is all that can be done now.

Below are a few particular events from the past decade or so.

In the Spring of 2008, NASA HQ hosted a 3-day gathering in Houston to discuss all the technologies needed to implement MSR. Each separate technology was represented by a group of people, each group taking a turn sitting around the front table to discuss technical details of rovers, Mars arrival, Mars orbit rendezvous, biohazard containment, etc. When it was time for the MAV discussion, there was only one NASA manager, a business development person from each of two aerospace companies, and myself. There was very little interest in having a technical discussion.

At the AIAA Space 2008 conference in San Diego, I presented a paper titled, "Defining the Mars Ascent Challenge for Sample Return." That was probably the first time I fully realized, and shared, the observation that the social system as a whole did not appreciate that the MAV would be more than just another planetary spacecraft propulsion system. Also that a MAV will not appear from someone else's "technology pipeline" like new computers do.

In 2009 I noticed that the Decadal Survey for planetary science (National Academy of Science and National Academy of Engineering) stated that Mars ascent propulsion would "evolve" from lunar ascent technology. I submitted my correction to the people who were preparing the subsequent decadal survey. Now the current version (up through 2023) says that the MAV is a huge challenge.

In 2012, NASA had an open invitation for anyone to submit abstracts for a meeting in Houston, for any ideas at all related to the future of Mars exploration.
My own submission is at the following URL (LPI = Lunar and Planetary Institute).
https://www.lpi.usra.edu/meetings/marsconce...12/pdf/4290.pdf

In April 2018, the 2nd International Mars Sample Return Conference was held in Berlin.
My submission to that meeting is at the following URL.
https://www.hou.usra.edu/meetings/marssampl...18/pdf/6035.pdf

At the Berlin conference, there was a lot of discussion of rovers, and sample handling back on Earth after the mission. The most detailed treatment of MAV was my own submission. The chief engineer of JPL gave a talk that had two slides about MAV, basically revealing that hopes are being pinned on propellant and combustion research. This is nothing new. In all my years of doing propulsion R&D for various government organizations, most people seemed to believe that "making smoke and fire" was 90 percent of the way to a successful rocket ship.

As of July 2018, the latest hybrid propulsion MAV research paper from JPL (AIAA) includes a statement that "there is a risk that additional mass may be necessary."

Regarding the MSL EDL propulsion, three excellent papers can be found on the JPL website (click the links for "show full item record").
https://trs.jpl.nasa.gov/handle/2014/44893 (Design and Development)
https://trs.jpl.nasa.gov/handle/2014/44137 (Fabrication, Assembly, and Test)
https://trs.jpl.nasa.gov/handle/2014/44159 (Lessons learned)
Even using "proven technology" it could not have happened without these seasoned specialists facing and overcoming unknowns along the way. It seems noteworthy that all these detailed papers don't include information about how heavy the components were, as though the peer review community of spacecraft propulsion experts is not particularly interested in mass.

Thanks for reading,

John W.

[Explorer1]

Would successful ISRU change be easier on mass requirements? Rather than lugging a full tank of propellant all the way to Mars, land an empty tank and fill it at the launch site?
Of course, the landing site has to have the raw materials, and there has to be some sort of refinery/power source to build it. Would this strategy outweigh (pardon the pun) the traditional methods? Mr Musk's long term Mars plans involve doing just this with a (much larger) vehicle...

[John Whitehead]

ISRU could be a benefit if the propellant to be produced can weigh more than all the equipment sent to Mars to produce and store and transfer the propellant. This is not likely to happen on the small scale of a science mission. Generally the ISRU community has focused on chemistry and propellants and combustion, with little to say about the realistic mass of a real propellant production plant.

Even if an empty MAV landed next to a filling station on Mars, the challenge of designing the MAV itself would not change, i.e. the MAV needs to be made with very lightweight engines and tanks relative to thrust produced and propellant carried. An empty MAV sent to Mars could be somewhat larger than a full MAV, which would make it somewhat easier to achieve a high propellant fraction.

John W.

[hendric]

I am not a rocket scientist, but I have played a lot of Kerbal Space Program.

With Mars escape velocity 1/2 that of Earth's, I am guessing at a back-of-the-envelope a MAV would need 1/4 the mass (E=.5MV^2) of an Earth-orbiting vehicle. I took a look at the Orbital Science Pegasus, since it launches at altitude and would probably have a closer flight profile (max Q etc) than a ground launch from sea level. Their Pegasus (https://en.wikipedia.org/wiki/Pegasus_(rocket)) weighs 18,500 kilograms and delivers a payload of 443 kg to LEO. So about 18,000 kg of "rocket". So 18,000/4 = 4,500 kg for 443 kg payload. I think we could probably usefully return 20kg of samples + container, so that would be ~95% mass reduction, so ~225 kg. They launch at 40,000 ft, which is higher pressure than Mars (equal to about 115,000). So there would definitely be less drag. Also, with less atmosphere/maxQ I assume your ascent profile/gravity turn can be more aggressive. With the lower atmospheric drag, maybe a nontraditional shape could be used. Does a MAV have to be a long/thin "rocket" shape? Maybe it can be short and fat, perhaps even round, to increase the mass/surface area equation so that less non-propellant skin material or tank material is needed?

How much would military missile design (esp. for fighter jets or SAMs) apply here?

[mcaplinger]

There is the notion that a successful Mars 2020 mission will greatly increase the priority and funding for developing a MAV...

I share your skepticism, but as the saying goes, "it's a lousy war but it's the only one we've got."

I've watched MSR flail around my whole professional career in aerospace. I've always felt that there were lots of viable ways to do it, and we just needed to have the will to pick one and follow it through, staying within a clearly defined and achievable set of budget constraints. But there is certainly a minimum viable budget.

People seem to be expecting some magic solution that will all of the sudden make MSR more affordable. I doubt that such a solution exists.

I have a Mars meteorite on my desk that Mike Malin gave me on the 25th anniversary of my working for him. I've felt for a while that barring some out-of-the-box event like Elon Musk's Mars plans actually coming to fruition, that's as close as I'll get to a Mars sample.

EDIT: BTW, I think you sell solids short. I've been active in amateur rocketry with small solids and hybrids for several years now, and I'm pretty impressed with what can be done with solids (less optimistic about hybrids, which is why I find this latest interest in them a bit perplexing.) As for electronics mass, there's a lot that's possible, but it would require a different mindset about reliability. The pendulum was swinging in that direction, but after the Mars98 failures (from which the wrong lessons were learned, IMHO) it swung most of the way back again.

[mcaplinger]

Does a MAV have to be a long/thin "rocket" shape? Maybe it can be short and fat, perhaps even round...

You can find any shape you want in various concept art for MAVs over the decades. I recall at least one short fat one with spherical exposed tanks from the 90s.

[siravan]

Regarding ISRU, has there ever been a concrete plan to send a demo Sabatier reactor as an addition to a science mission to field test one (just to make some Methane, not actually launching a rocket)?

[mcaplinger]

Regarding ISRU, has there ever been a concrete plan to send a demo Sabatier reactor as an addition to a science mission...

Not methane, but I presume you're aware of MOXIE? https://mars.nasa.gov/mars2020/mission/instruments/moxie/

IMHO, ISRU is one of those supposedly enabling technologies that have served to distract MSR from simpler, more workable architectures for decades.

[djellison]

IMHO, ISRU is one of those supposedly enabling technologies that have served to distract MSR from simpler, more workable architectures for decades.

Seconded. ISRU is enabling for future human exploration architectures. It's an unnecessary source of complication and cost for sample return.

One thing I remember hearing at some point - something similar to an AMRAAM missile would make a capable MAV first stage. A small mono-prop upper stage for orbit insertion would finish the job quite well.

[John Whitehead]

something similar to an AMRAAM missile would make a capable MAV first stage. A small mono-prop upper stage for orbit insertion...

It would be nice to know the source of this claim. In Post number 90, I pointed out that a MAV launched on Earth would have a range of roughly 500 km. The AIM-120C-5 AMRAAM can go only about 100 km with a 18-kg payload. https://en.wikipedia.org/wiki/AIM-120_AMRAAM
So an 18-kg upper stage (including the samples) would have to do most of the work to reach Mars orbital velocity. Presently the notional MAV payload is 18 kg, plus another 18 kg for non-propulsion items including guidance (JPL 2017). So the 152-kg AMRAAM would have to be scaled up a lot, to something like a ton. After all the decades of MSR planning and MAV research, the big unknown is still, "how large and heavy is necessary?" In order to deliver a MAV to Mars within the capability of the current landing technology (MSL and Mars 2020), the MAV needs significantly better rocket technology than AMRAAM.

[John Whitehead]

With Mars escape velocity 1/2 that of Earth's, I am guessing at a back-of-the-envelope a MAV would need 1/4 the mass (E=.5MV^2) of an Earth-orbiting vehicle.

Richard, here is my perspective on your rough calculations, which seem to have the following three assumptions:
1. Propellant energy is all converted into payload kinetic energy.
2. Non-payload mass is only propellant.
3. Rocket stages can be scaled down while keeping the same relative capability (same delta V for same percent payload, same propellant mass fraction, same thrust-to-weight ratio).

Regarding Assumption 1, I recall reading decades ago that rockets are "momentum machines," not "energy machines" (or "power machines") which can be explained as follows. Considering one extreme, if there is a tiny amount of propellant, we have a simple conservation of momentum equation. For 1/4 of the propellant, we get 1/4 of the velocity, not half. At the other extreme, as a rocket approaches 100 percent propellant, changing the relative payload mass by a factor of 4 does not change the velocity obtained. This makes sense because most of the work of the rocket is to accelerate the propellant that remains on board throughout the flight. And that's the main reason that rockets are not "energy machines" (focusing propellant energy onto the payload), namely that much of the energy is needed to accelerate the propellant that remains on board and is later expelled. What is amusing, however, is that somewhere in between the extremes, we actually do get half the velocity for 1/4 as much propellant relative to payload (but only for the reason that the number 1/2 exists within the range of extremes from 1/4 to 1). So maybe that explains why it might be easy to have the impression that the energy equation applies.

Assumption 2 essentially says that tanks and engines might have no mass, which we can only wish for. Other than propellants and combustion, building lightweight components is THE engineering problem. There is a joke that a physicist doing a rocket calculation observes that the propellant mass dominates, so just neglect the rocket hardware mass (actually I've encountered two real-world PhD physicists who said just that, LOL).

Assumption 3 unfortunately does not work either, which is why it is so hard to pin down how heavy or how lightweight an actual MAV might be. Regarding your example, a miniature Pegasus cannot be built according to the criteria I listed above for Assumption 3.

[hendric]

Thanks for the information John. I know that #3 prevents a direct scaling (ie, a 1000psi tank has to be a certain thickness, no matter the volume), but for a back of the envelope I think it's ok. #2 is hard to estimate without a dry mass of the rocket. #1 is pure derp on my part, sorry.

On the missile comparison, I think a better comparison is a Phoenix long-range AA missile. 190km range @ 470kg weight. https://en.wikipedia.org/wiki/AIM-54_Phoenix

Also, don't mistake a missile "warhead" for a payload - warhead mass is just the amount of high explosives. I assume there is significant overhead for radar seeking guidance.

There are/were several companies interested in using balloon launched rockets to reach LEO. Maybe there is some cross-pollination possible with MAV from that area.

[mcaplinger]

There are/were several companies interested in using balloon launched rockets to reach LEO.

Find one that isn't vaporware. This is a problem that sounds easy to a casual observer and ends up being not very easy at all. Unfortunately, MAV design has been plagued by casual observers and technologists selling their own particular solution to the exclusion of good system tradeoffs for a long time.

[hendric]

Surprised NASA hasn't tried to sponsor an X-Prize like competition to get creative juices flowing. Having some actual hardware would answer a lot of questions.

[mcaplinger]

Surprised NASA hasn't tried to sponsor an X-Prize like competition to get creative juices flowing. Having some actual hardware would answer a lot of questions.

I'm sorry, but IMHO this misses the point. The problem is not a lack of "creative juice" -- you are buying into the myth that somehow this would be easy with the right out-of-the-box thinking. The reality is that it's a hard technical problem that requires enough money to solve and the will to follow through.

Give me an example of an X-prize that led to something useful. The Ansari X-prize led to a few flights of a vehicle that now hangs in the NASM but has proven a real challenge to commercialize, and the GLXP led to a whole lot of promises and no flights at the time the prize was cancelled.

I'd suggest that you take this discussion over to NSF, but there you will simply be told that Elon will return all the samples you want in a few years.