This week electrovehicle and space transit magnate Elon Musk unveiled his master plan for the human race to become “a multi-planetary species”, starting with travel to Mars. But while I know it’s an act of remarkable hubris for me to even say this, I believe SpaceX’s plans for their Interplanetary Transport System have some pretty glaring flaws. The good news is they are flaws that are easily corrected by applying some decades-old solutions from the American and Soviet space-faring experience.


After going over the system architecture and basic spacecraft design, I think there are places where the system can be made to use less fuel, reduce manufacturing complexity, increase crew room, deliver more payload, and more.

If you haven’t yet watched it, this lovely little video shows the basics of the SpaceX Interplanetary System:

I want to add that, overall, I’m delighted by the whole Mars/Interplanetary travel plans. I think that even if this does not end up being the way we eventually start to spread off our planet, Musk’s determined and detailed plans for how it could happen are incredibly valuable.

The British Interplanetary Society’s 1938 plans for a lunar mission were so influential on the engineers and astronauts that actually made it happen in the following decades. These deep-dive explorations are crucial in turning vague concepts into achievable reality, and Musk is doing more to advance the goal of human off-planet colonization than pretty much anybody right now. My critique is meant to show I’m taking it seriously.

I think the SpaceX spaceship design and system architecture is elegant and lovely, and I’m sure that’s a big part of its appeal to Musk; it looks futuristic, and fundamentally different than the clunky-looking space hardware we produce now. I suspect that Musk’s desire for something with this look may be directing these plans to the detriment of other concerns.

We’re looking at the same thinking that pushed the Tesla Model X into production with those dramatic and problematic Falcon Doors. Only here, it’s a massive space transportation system, not a rich-guy/gal SUV. My changes will be killing a lot of this visual elegance, but with good reason.

The Changes

The changes I’m proposing involve the System Architecture SpaceX presented and the design of the main crew capsule/lander. I’m assuming all of SpaceX’s underlying technology (launch vehicle systems, vertical landing, etc) and methane-fueled Raptor engines and all that related, underlying tech is just fine.

One of Musk’s slides highlighted the difficulties of getting to Mars because getting anything there is so damn expensive. It’s incredibly expensive to get anything off Earth, which is why weight/mass savings in a spacecraft is so critical.

The Ship

That brings us to the centerpiece of Musk’s plan, the massive crew-carrying spacecraft/lander that the slides just called the “Ship.”

In his talk, Musk described the Ship as being spacious, comfortable, and a place where going to Mars will be “fun.” He talked about watching movies and eating in restaurants on board. We can presume that in addition to the theaters and restaurants, there will be crew quarters, medical facilities, exercise equipment, hygiene/life support equipment, and on and on. All the stuff you’d need to keep up to 100 human colonists alive and not insane for the 80+ day journey to Mars.



Having all this stuff on board for the use and support of up to 100 people means that the ship is huge. Like, really freaking huge. All that stuff is also incredibly heavy, even if it’s made with carbon fiber and has holes Swiss-cheesed into every flat surface. It’s just a lot of stuff.

And everything is an integral part of the Ship, which means it has to get launched from Earth with every single launch of the Ship, it has to enter Mars’ atmosphere (all protected by heat shielding), be braked by rockets (using fuel) as it lands on Mars, and then launched again from the Martian surface back into space, using precious fuel produced in-situ on Mars.

Oh, and then it has to enter Earth’s atmosphere, be braked yet again by rockets, and landed. Then it all starts over.


Since these ships are colonist transport ships and not intended to turn into long-term Mars habitats, those restaurants and toilets and sleeping bunks and galleys will only be used in the orbital and Earth-to-Mars space portion of the trip. For all those launches and landings and re-entries, they’re just dead weight.

This, of course, is insane, and people building spaceships have known it’s insane at least since the late 1960s.

The Hab Module Solution

Let’s look at the most successful spaceship ever built as an example: the Soviet/Russian Soyuz. The Soyuz designers made a point to separate the components that will only be used in space so that they wouldn’t have to build them to withstand being in an atmosphere or have to shield it from re-entry heat or support it with a parachute. The mass payoffs to this approach are significant. This is what the Encyclopedia Astronautica says about it:



The end result of this design approach was remarkable. The Apollo capsule designed by NASA had a mass of 5,000 kg and provided the crew with six cubic meters of living space. A service module, providing propulsion, electricity, radio, and other equipment would add at least 1,800 kg to this mass for the circumlunar mission. The Soyuz spacecraft for the same mission provided the same crew with 9 cubic meters of living space, an airlock, and the service module for the mass of the Apollo capsule alone!

Americans did the same thing when it came to making a lunar lander. By making it a separate vehicle from the Apollo command module, it could be made much lighter, and save mass that can be used for fuel, supplies, whatever.

So, what I propose is that the Ship be reduced in size significantly: at least half its current proposed size. We’ll do this by removing all of the major in-flight crew support systems and amenities and putting those into a separate Hab module that would get launched once and spend the rest of its life in space, never to return to any planet.


The smaller Ships will carry the same amount of people, but with much more limited amenities and more basic life support. Bathrooms, sure, but no zero-G pools or gyms or restaurants or whatever. The current Ship uses six in-space engines and three atmospheric ones; a smaller ship could use four and two, which should still give it a good margin of redundant capacity and backup.

I’m also a little concerned about the stability of a Ship that big landing vertically on Mars, on rough terrain, and having to withstand possible winds and storms. The air pressure is much less on mars, so perhaps this isn’t as big an issue, but why chance it for no good reason?

I have a design idea for the Hab module, too. There’s no reason why we can’t get even more usable space for these potential 100 passengers and colonists in a separate module. I think a Hab module that uses a Bigelow Aerospace-type inflatable module would make sense. There’s even a test one docked to the ISS as we speak.

Hell, let’s use four of them. By arranging the four modules in a sort of overhead-view of a 1970s video game racecar sort of look (where the modules are like big fat wheels) we can get a dual barbell layout that can be spun about the long axis, giving some artificial gravity.

Since Mars has gravity—less than Earth, but still significant—this centrifugal-force synthetic gravity will be a huge help to keep the colonists’ muscles from atrophying. Even the minimum travel time of 80 days is no joke, and generating gravity en route will be a huge help to making the colonists prepared and healthy or their arrival on Mars.

Based on the lovely animation SpaceX made showing their Interplanetary Transportation System, the Ship’s solar panels appear to be only deployed when traveling through interplanetary space; in that case, those too can be moved to the Hab, for even more mass savings when the Ship launches or lands. Plus, thermal radiators could be mounted as well, which I suspect will be needed, even if the animation didn’t show those. Same goes for any number of un-aerodynamic, non-sleek parts like communication antenna arrays and airlocks and other clunky spacecraft stuff.



And, remember, even if the Hab looks bigger than the ship itself, it won’t add more mass than the huge ship, because it’s built only to be in space, which can be orders of magnitude lighter than something that must withstand re-entry, G-loads, an atmosphere, and so on.

The Space Garage

Hab Modules would be built and launched into a parking orbit either around Earth or at the L1 Lagrange Point, which would become a sort of space “garage.” The much smaller Ships would then be launched from Earth—which could be done with a smaller version of SpaceX BFR rocket or with the same one, just using much less fuel—and would then dock with the Hab Units when in orbit.

Supplies for the trip could still be carried on the Ship, and would be transferred to the vast habitable space in the Hab unit. I think perhaps the two inflatable modules closest to where the Ship docks would be crew quarters, 50 per module, and the other two would have all the communal spaces and facilities. The modules would be divided into decks, oriented with ‘up’ being towards the central spine modules.


There are other significant advantages to reducing the massive size of these components and changing the Ship design from something monolithic to something modular. The logistics of building the Ships becomes much more manageable when the size is reduced to something closer to what humans are used to building on a regular basis, and Ships can be built in one facility while Hab Modules are built somewhere else.

Yes, it will take more launches to put the Hab Modules into their orbits, but these are one-time launches, and the fuel and mass savings will be realized for every launch, landing, and re-launch of the ship, which comprise four separate fuel-intensive events for every single Earth-Mars round trip.

A better way to refuel

The way SpaceX proposes handling refueling the Ship prior to the Mars journey also strikes me as bizarre. In the video, we see the huge launcher putting the Ship into orbit, then the launcher re-enters the atmosphere, comes to a controlled landing, gets the fuel tanker module installed on it, and launches again, where the tanker detaches from the booster, goes into orbit, docks with the Ship, fuels it, and lands.



This method makes no sense, because it introduces so many new points of failure and establishes a needless artificial deadline on the fueling procedure. I mean, I know Elon sort of loves giving himself arbitrary deadlines, but there’s absolutely no need here. Think about this process: if anything goes wrong with the booster during re-entry, landing, attaching the fuel tanker craft, and relaunching, that is four significant potential points of failure where the crew in the Ship could be just stuck there in Earth orbit, without fuel.

And there are zero reasons for it to be this way. Tanker ships should be launched before the crew Ship, and should be waiting in a parking orbit, perhaps in the same ‘garage’ location as the Hab Modules. Full fuel tankers should be waiting on-call for the next crew Ship, and dock with the ship as soon as it enters the hangar. The tanker would fuel up the crew Ship, undock, and come to a controlled landing on Earth, ready to be refilled and re-launched. Why would you want to do this with the same launcher under time pressure while the crewed ship is in orbit? A fleet of these things should be waiting, ready to go.

Plus, the new modular nature of the system means that extra tanker modules could be docked to the Ship/Hab spacecraft complex, allowing for an easy way to make the ship capable to undertake missions beyond Mars. Extra supply modules, based on the basic tanker design, could be docked to the Hab as well. There are very good reasons why most large spacecraft we’ve used so far resemble tinker toys.

The revised system architecture

That covers my key redesigns of the Interplanetary Transportation System components; now, I want to revise the System Architecture that Musk showed for a typical colonial-dropoff trip from Earth to Mars and back. I think my changes would result in a significant savings in fuel and a significant amount of extra usable mass that could be used for more supplies, equipment, etc. For reference, here’s Musk’s plan:

Let’s go through the steps. I’ll be adding some more detail than is in the chart:


1. SpaceX BFR Booster launches with huge crew Ship

2. Crew ship enters Earth orbit and waits as…


3. Booster re-enters atmosphere and comes to a controlled vertical landing.

4. Booster gets Fuel Tanker mated to it


5. Booster re-launches and releases Tanker to rendezvous with crew Ship in Earth orbit

6. Booster again re-enters atmosphere, comes to controlled vertical landing for re-fueling and re-use

7. Tanker fuels Ship, then undocks


8. Tanker re-enters atmosphere, lands, gets refilled and prepped for re-use

9. Ship deploys solar panels and fires engines to start journey to Mars


10. Ship reaches Mars, enters Mars orbit

11. Ship enters Mars atmosphere, and lands on the planet (again, it’s huge and carrying all the now-unused crew facilities) by firing its engines as retro-rockets


12. Ship dumps its crew to make Mars tacos or whatever, refuels from Mars-based methane-production facilities (and oxidizer, too), and launches off Mars and into orbit

13. Ship fires engines for return cruise to Earth

14. Ship re-enters Earth atmosphere, comes to controlled landing for re-use.


Okay. Now, here’s the architecture of my proposed modified system:

These steps are a bit different from the chart to get some more detail, but it should all make sense:


0. Prior to initial launch, Hab Modules and Tanker Ships will already have been launched to Garage orbit

1. Smaller version of SpaceX BFR launches smaller version of the crewed Ship


2. Crew Ship enters Earth orbit and adjust orbit to match Garage orbit as…

3. Booster re-enters atmosphere and comes to a controlled vertical landing. It is in no rush to get back into space.

4. Crew Ship rendezvouses with Tanker Ship and docks, refuels


5. Crew Ship undocks with tanker, rendezvous and docks with Hab Module

6. Empty Tanker Ship re-enters atmosphere, comes to controlled landing for re-use


7. Crew Ship and Hab Module Assembly deploys solar panels, antennae, radiators, etc., starts spin, and fires engines to start cruise to Mars

8. Ship and Hab complex reaches Mars, enters Mars orbit


9. Ship (again, much smaller) undocks from Hab, enters Martian atmosphere, lands; while in orbit, the Hab Module will act as a communications and weather satellite

10. Ship dumps its crew to make Mars craft beers or whatever, refuels from Mars-based methane-production facilities (and oxidizer, too), and launches off Mars and into orbit

11. Ship re-docks with Hab Unit


12. Ship/Hab complex fires engines to begin return trip back to Earth

13. Ship/Hab complex enters Earth orbit; at Garage orbit, Ship undocks with Hab, leaving it for the next use in the Garage


14. Ship re-enters Earth atmosphere, comes to controlled landing.

I’m not sure how I managed to get those to have the same number of steps, because I’m pretty sure there’s a few extra steps in my version, but I’m confident that the overall effect will be that these modifications will provide an overall architecture that has significantly less fuel requirements, provides more usable room for humans and cargo, provides additional health benefits by providing artificial gravity en route, simplifies manufacturing and logistics, provides a better margin of redundancy and makes the missions safer, and makes this all cheaper and more possible.


I’m a believer in Musk’s vision that humanity needs to become a multi-planetary society. I’d love to see this happen, and, if I can convince my wife, I hope to retire on Mars and open the first Martian air-cooled Volkswagen restoration shop on the planet.

That is, if they manage to have some air.