SpaceX: Get your ass to Mars

Miscellaneous and Personal Stuff I Must Share
1

SpaceX has announced a 2018 mission to Mars. It will be, apparently, entirely privately funded, though with substantial non-monetary support from NASA, in exchange for flight data from SpaceX under a “no-exchange-of-funds agreement” (that has been in place for some time).

Although they haven’t talked yet about the flight profile in great detail, we can guess that it’s not far from the Red Dragon proposal talked about in this presentation. It uses a highly modified Dragon 2 craft–the same one that they intend to use to ferry astronauts to the space station.

All Mars flights to date have used a parachute. Parachutes do not work well on Mars, due to the thin atmosphere, and the landing vehicle still requires rockets of some kind. Parachutes, unfortunately, do not scale–the most recent (and heaviest) Mars lander was the rover Curiosity, and this is about at the limit of what parachutes can do. Any kind of manned Mars mission, or even the more advanced robotic missions (such as sample return), requires much more.

SpaceX is using a different approach. Dragon will use aerobraking to achieve most of the initial velocity loss. It will use a much more aggressive flight profile than other missions due to the high mass–it will have to dip low into the atmosphere, and then actively alter its center of mass (using ballast on an internal sled) to achieve some degree of lift and control the landing zone. It can dissipate a large amount of kinetic energy this way.

However, it will still be travelling quite fast after this phase–perhaps mach 2-3. The Dragon craft uses retropropulsion for the final phase. This is not trivial, because unlike previous efforts, Dragon requires supersonic retropropulsion. The rockets have a complex interaction with the shock waves from the heat shield, making the process far trickier than one that only cancels a couple hundred m/s of velocity. Fortunately, SpaceX has a great deal of experience in this regime–the only real experience, in fact–from their hypersonic boostback burns supporting their booster landing efforts.

All of this requires a great deal of fuel compared to parachutes, but there appears to be no choice for large payloads, at least when pinpoint landings are required (as they would be for a manned mission). SpaceX hasn’t said exactly how much mass they will end up landing on the surface, though they believe that 2-4 tons of delivered payload is possible, which implies a system mass of 8+ tons. Curiosity was only 0.9 tons, so this is a significant leap.

2018 is a pretty ambitious schedule, especially as the mission requires the (yet to be flown) Falcon Heavy rocket. But the Dragon 2 has been doing pretty well in testing so far, so perhaps it’s not as ambitious as it might seem. And they will be getting no small amount of support from NASA, especially when it comes to deep space communications, so they are not starting from scratch either. Should be fun to watch!

2

I gotta think SpaceX needs a whole-company sick day every time Musk says something like this and they all face-palm themselves into unconsciousness.

2018? No way.

2022? More likely.

3

The people who face palm after management decrees tend to get summarily terminated; for better or worse, SpaceX is not relucatnt to fire people who don’t hew to the mission directives as laid down by Musk. (I think that is a somewhat short-sighted method of management derived from the Silicon Valley approach that doesn’t necessarily work with hardware, but whatever.) However, SpaceX been working toward supersonic retropropulsion and a Mars landing capability for a while with relatively little publicity, and provided the debut of the three core Falcon Heavy is successful and SpaceX is otherwise able to fund the launch while maintaining their other contractual obligations, they’ll likely try to launch this mission because 2018 is the lowest C[SUB]3[/SUB] (energy required to achieve Earth escape on a Mars injection trajectory) opportunity for the next fifteen years or so. Whether they will be successful is another question.

So far, only NASA has successfullly landed probes and landers on Mars, which is literally the most difficult solid body to land on in the solar system. That SpaceX is attempting to use an all-propulsive method of final descent and landing adds to the challenge, but will likely be a necessary technology for landing payloads that are significantly heavier than the Mars Science Laboratory and will contribute to technology for landing on other celestial bodies for which aerodynamic deceleration is not applicable. Even if the SpaceX attempt is launched and ultimately fails, there may be useful information to be gleaned which will inform future efforts.

However, I have some concern about the particular obsession on Mars as an ultimate destination. Mars seems appealing because it has a solid surface with signficant gravity, but the notions of colonizing Mars are misguided in my opinion. We don’t know the long term viability of physiological adaptation to living in the ~1/3 g field of Mars, but in many other ways it is still not well suited to habitation, nor does occupying Mars extend the ability to access the other riches of space, nor develop a sustainable space communication, transportation, and resource utilization infrastructure. Mars is useful as a place to exercise certain capabilities, but I don’t think it makes a very good endgame for exploration and habitation, either in scientific or practical terms.

Stranger

4

Everyone working at SpaceX is there because they believe in the dream of Mars exploration. They certainly aren’t there for the high pay, low hours, and relaxed environment! Perhaps they are delusional on the timeframes, but they certainly believe in the mission. So far, the inspiration seems to be working.

5

I think there’s a strong case to be made for Mars, if not as the endgame, then at least as a strong intermediate point. It is fairly rich in resources; particularly water and carbon. There is reason to believe that it has useful metal deposits.

The gravity is strong enough to make human occupation relatively straightforward (compared to zero-G, in which just about everything is absurdly difficult, from turning a wrench to breathing), while being weak enough that SSTO craft are easy. If Mars proves amenable to occupation, it may be easier to bring resources from the Martian surface than Earth’s.

I suppose I also see asteroid mining and large space habitats as the endgame, but Mars may prove the better industrial base to start from. This is all very long-term, of course.

At any rate, I would like to see advancements in robotic planetary exploration, regardless of the ultimate habitability, and the SpaceX effort will support these goals. Among other things, I would like to see a robotic drilling operation in the relatively short term; one that could dig hundreds of meters. This is going to be a multi-ton probe and can probably only be done with propulsive landings. We have literally only scratched the surface of Mars, and to discover its riches (or to prove there aren’t any), we need to go rather deeper.

6

In the nearby thread about software bugs, I posted an old story from the “Famous Bugs” document, about an early-1970s-era Mars mission that was lost. The lander landed, but due to poorly documented interactions among various software modules, all radio contact was lost.

I wonder if any of the more recent Mars landers have come across that old lander, or if any of Musk’s missions will.

7

Am I getting the following right: The problem with Mars is that the atmosphere is thick enough to produce lots of heat and quickly decay your orbit but not thick enough to really slow you down.

So, if you start getting low enough that the atmosphere becomes significant, without a parachute/wings/retropropulsion, you’d meet the ground at a high speed. Most of that speed would be horizontal speed, correct?

To you and anyone else familiar with these industries: Why is it common and/or useful to do that with software?

Do they not often have a “red team” that strives to see problems in proposed ideas?

8

I like this, if for no other reason than it pulls the logistical part of the mission away from Congress. SpaceX gets media attention, the Falcon Heavy gets to throw a capsule at Mars and NASA gets to examine a secondary approach to landing heavy equipment on Mars. I note they basically killed Low-Density Supersonic Decelerator project.

9

That’s the basic gist of it, but I’d add “to a speed appropriate for landing”.

The Martian atmosphere is extremely valuable for exploration. It allows aerocapture–that is, entering Martian orbit with a minimum of fuel. Airless worlds require a rocket burn to enter orbit (though if you’re lucky, a nearby body may allow a gravity assist). Thin as it is, the Martian atmosphere is enough to slow from injection velocity to orbital velocity.

A lander indeed requires heat shielding, and yet the atmosphere is thin enough that the “easy” methods (parachutes, mainly) either do not work at all or require additional assistance. On Earth, you can land quite large payloads purely by parachute, and gently enough to not damage reasonably robust equipment.

In comparison, the lander for the Mars Pathfinder mission, which landed the Sojourner rover, had a descent speed of 160 mph. And that was one of our smallest landers. It used retrorockets and giant airbags to handle the remaining velocity.

As payload mass increases, this problem just gets worse. As said above, parachutes have a scaling limit; they can only get so big. If you can’t depend on them, then you must depend on the ballistic properties of the lander itself. This is tough, since you may not even have enough time for the lander to reach terminal velocity (which is already going to be much higher for a dense capsule than a light object on a parachute). Designing the lander as a lifting body helps in this regard.

The result is that the short landing burn that other landers used is now done supersonically and for a much longer duration. You need to devote a high fraction of your payload mass to fuel, and can’t get rid of the heat shielding, either. There are no showstoppers here, but it’s a regime that hasn’t been explored in any detail.

All that said, we’d be vastly worse off if Mars had no atmosphere at all. Mars orbital velocity is around 3.8 km/s. Aerobraking, even for dense capsules, can get rid of almost all of that–probably around 3 km/s. The remaining several hundred m/s of velocity is indeed a challenge, but the alternative would have been to get rid of all of it via retropropulsion, as we did on the Moon. It would have made Martian exploration almost impossible (though it would also have made Mars much less interesting as a destination, so maybe it doesn’t matter).

10

That is essentially correct with the caveat that at low speeds you can’t develop enough lift to be useful, period. Although the orbital or injection speed is tangent to the planet’s gravitational field, you’re also accelerating at you fall into the gravity well. The typically proposed solutions are some combination of a lifting body, deployable inflated decelerator (ballutes or an annular bluff shape) and terminal retropropulsion. What SpaceX is proposing is almost all retropropulsion from high supersonic to quasi-static terminal approach. This is a viable option, and even technically achievable, but all of the studies I’ve seen and worked on rate this as requiring the highest vehicle mass to landed payload mass as a vehicle capable of powered landing and return ascent has essentially the same performance as an Earth SSTO. I haven’t seen performance details on the Dragon 2.0/Red Dragon, but just a cursory look at the design seems to indicate that it lacks the necessary performance and/or propellant capacity for a fully powered descent, much less return ascent. (And before we starting talking about in-situ propellant manufacture I’ll point out that Dragon uses storable liquids, e.g. hydrazine and NiTet, which are costly and difficult to manufacture on Earth with readily available precursors, and that methane, the propellant most readily manufactured in situ has nowhere near the performance necessary for a compact single stage ascent vehicle, even in Mars’ 1/3 g field.)

I don’t know if it is “useful” but the periodic culling of the perceived low performers is an inculcated business practice and culture in many Silicon Valley firms. The basic idea is understandable; instead of leaving ‘dead woood’ to waste time and consume resources, you have a scrub to get rid of unproductive employees and bring in fresh blood. In many commercial software development firms with production cycles measured in a few months this may make sense and isn’t terribly disruptive as they will break up teams frequently anyway. In longer term efforts, it is problematic, as even perceived low performers may have critical skills or knowledge that is lost in such clearcutting, and it often serves as an excuse to eliminate people who may be perfectly adequate or better but are disliked by someone in management for whatever reason. SpaceX, like many startups, is something of a cult of personality centered around Musk and Gwynne Shotwell, and anyone who disagrees with their vision is liable to find themselves at the curb in short order, even if it is a valid disagreement based upon technical concerns. For a startup, this kind of rule of iron fist can be helpful in not getting mired in a debate about the appropriate path; often with a startup, inaction is worse than pretty much any other course of action. But at some point in achieving commercial viability you need to have naysayers and alternatives to balance enthusiasm for visionary ideas. In the case of SpaceX, they need to become massively profitable before Musk’s vision of colonizing Mars or whatever is even remotely practical, because nearly everything done to support that effort is a total loss from profit in any foreseeable term. So, they really should be focusing on both providing highly reliable and cost effective launch services AND encouraging the growth of space industry in order to give them more future business. A diversion about landing objects and eventually people on Mars that detracts from their core business is newsworthy but not fiscally rewarding.

It is unfortunate that the LDSD was terminated; despite the failures, much information was learned from the testing, and I personally think that something like their design will be necessary for landing heavy payloads on Mars in the foreseeable future, and may also inform efforts to develop an Earth SSTO. But fundamentally, I don’t think it is sensible to focus all attention on Mars, and especially not a crewed mission. For the cost of a single crewed mission we could pepper the surface of Mars with hundreds of probes and rovers that would cover more ground than any crewed mission without the parasitic cost of returning after the science mission is done.

Meanwhile, we have much to learn about the other planets and moons in the solar system which we may never be able to send crewed missions to, and for which a more generalized space logistics and communication system would have broad applicability and high value. The notion of ‘colonizing’ Mars as some sort of ‘backup Earth’ is about as sensible as putting a sidecar on a racing bike. It will be less costly and ultimately more viable to recreate terrestrial conditions in purpose-built space habitats than we could ever do on Mars, and in doing so also develop the capability to protect the Earth from hazards such asteroids, solar storms, et cetera. Neither SpaceX nor any other commercial entity has either the capital resources or wherewithal to engage in this large scale infrastructure development that is key to future space exploration and exploitation of space resources.

Stranger

11

It may seem that way from a propellant budget sense, but you have to offset propellant saved in aerobraking to the mass of the aerobraking system and the associated systems to protect the vehicle from damage due to aeroheating and erosive effects, which can be substantial. Mars atmosphere is just dense enough that solid particular matter can be held in suspension, with can grossly exacerbate erosion. In the 'Eighties, a biconic profile capsule was preferred for reentry for being able to provide good aerobraking performance and a degree of controlled lift all the way down to almost subsonic speeds. However, knowing what we know now about Mars climate (after the loss of the Mars Climate Orbiter the Mars Reconnaissance Orbiter was modified to achieve most of the mission goals of the lost vehicle), such a vehicle would require a very heavy ablative shield. Although the all-propulsive breaking and landing method would be heavier, it is technicaly less risky and gives a more simple design presuming that it can be launched into and from Earth orbit.

Stranger

12

No denying that. Propulsive descent in a vacuum is certainly easier from a technical perspective. The LM was made from bubblegum and tinfoil and had no problem landing on the Moon; it’s not so easy on Mars.

That said, if you have to land X tons on Mars, you have to consider how you got there. A full propulsive descent may be easier on the lander, but implies a larger launch vehicle–one that may not even exist. That is a technical challenge in itself.

In related news, SpaceX recently up-rated their engines again, and bumped their listed performance numbers. The F9 (in expendable mode) can now lift 22.8 t to LEO, making it technically a heavy-lift launch vehicle.

13

Was this the Viking 1 lander? The lander operated successfully for several years, but ultimately brought down by buggy software that broke the antenna pointing module. Even so, it was an entirely successful mission.

Occasionally, people say (presumably tongue in cheek) that Mars is cursed, due to the number of failed missions to it. But if you look at the Wikipedia page, the American missions were largely successful; it was the Soviet missions that were almost entirely failures. Red planet it ain’t.

14

I would take those claims with more than just a grain of salt. SpaceX is a great marketing company, full of promises on what they can do. Note that the advertise 8.2 metric tons to geosynchronous transfer orbit, but then under their $62M manifest price (which is just a bare manifesting and minimal payload processing accommodations) state that it is the cost for 5.5 mT. You find this sort of bait & switch all over the claims, e.g. they can get you to a GTO orbit, but if you want the GTO orbit that gets you to your required injection point, it’s another $10M. They’re still beating out ULA in the commercial arena, which is to be applauded and encouraged, but their specfic claims of cost or capability should be viewed with a critical eye.

Stranger

15

Yeah, it’s a little disappointing that their site isn’t more clear on what you get for the listed $62M price–it’s clearly not the maximum numbers they list. Musk acknowledged this via Twitter, plus you can see a partial caveat in dark grey text that says “5.5 mT to GTO”, despite being listed as 8.3 mT below. That said, I’m reasonably confident that if you came to them with a blank check, they would indeed be able to put 22 t in LEO.

While we’re talking about rockets–can you explain the insanity of India’s PSLV? It’s a 4.5 stage rocket, with strap-on boosters, solid first and third stages, two different liquid fuel types, and injection of strontium perchlorate for their thrust vector control. It’s nuts!

I can understand all the pieces in isolation. I can’t understand why someone would build a rocket like that. There must be some reason related to how India’s space program works, but I can’t make heads or tails of it.

16

WAG, sanctions and tech transfer restrictions.

17

I figure that there has to be some kind of “local development” angle going, but it’s still weird. Why two different liquid fuels? Why the solid-liquid-solid-liquid arrangement? Why the bizarre liquid injection system for TVC instead of gimbaling or vernier thrusters?

The vehicle is pretty reliable, so I give them credit for that despite the seeming complexity. It’s just really weird.

18

I’m not intimately familiar with the PSLV, but there are reasons to use solid propellant motors both in first stage/strap on applications (high thrust to develop speed quickly) and upper stages (reliability, lack of slosh characteristics, avoiding propellant loading). Strontium perchlorate is frequently used as the working fluid in liquid injectant thrust vector control applications for solid propellant motors, which allows for a simplier fixed nozzle TVC system where high vector response is not required. The notion in using the same propellants in all stages does allow for somewhat simplier logistics, but the cost of propellants is minimal, and the logistics of handling multiple non-cryogenic propellants is not that complicated.

I suspect that part of the reason the PSLV system was designed in that way was political, to share the work between various design bureaus, but it is hard to argue that it works, and I’ve seen crazier vehicle architectures both proposed and actually flown. At the end of the day, the best performance for lowest cost or risk is the goal, so you get things like a winged shuttle lofted with giant solid boosters and a cryogenic sustainer, carrying a payload boosted by a solid propellant kick motor.

Stranger

19

I think so. The story sounds right. Yes, this happened after the mission had completed its basic objectives. But it still goes down in software history as a major blunder.

Here’s an old post where I quoted the story (which discusses how it happened), and here is a link to the actual classic “Famous Bugs” document that has the story. Fun read, about major bugs of olden times.

20

I want us to go to Mars because I want us to find signs of life there - extinct or otherwise.

Landing such a large mass on Mars sounds quite a challenge. Might it be worth sending two rockets and have them transfer fuel in Mars orbit before the descent? With the direct descent as a backup if one rocket fails. Then you could have one Dragon on the surface and one in orbit. Or you could do two descents with different fuel strategies.