JEAN LEGALL: Good afternoon, ladies and gentlemen. I'm Jean LeGall. I'm President of the French Space Agency and the President elect of the International Astronautical Federation, and it is my pleasure to welcome you here at the 67th International Astronautical Congress. Elon Musk is founder, C.E.O., and lead designer of SpaceX. Elon founded SpaceX in 2002 with the goal of revolutionizing space technology and ultimately enabling humans to become a multiplanetary species, and that's the plan he's going to lay out for us today. SpaceX has had a number of firsts including as the first private company to deliver cargo to and from the International Space Station and the first entity to land a nautical thrust booster back on land and on ships out at sea. Please join me in welcoming Elon Musk. [APPLAUSE]. ELON MUSK: Thank you. Thank you very much for having me. I look forward to talking about the SpaceX Mars architecture. And what I really want to achieve here is to make Mars seem possible, make it seem as though it's something that we can do in our lifetimes and that you can go.
And is there really a way that anyone can go if they wanted to? I think that's really the important thing. So, I mean, first of all, why go anywhere? Right? I think there are really two fundamental paths. History is going to bifurcate along two directions. One path is we stay on Earth forever and then there will be some eventual extinction event. I don't have an immediate doomsday prophecy, but eventually history suggests there will be some doomsday event.
The alternative is to become a space-faring civilization and a multiplanet species, which I hope you agree that is the right way to go. Yes? [APPLAUSE]. That's what we want. [APPLAUSE]. Yeah. So how do we figure out how to take you to Mars and create a self-sustaining city, a city that is not really an outpost but can become a planet in its own right and thus we can become a truly multiplanet species? You know, sometimes people wonder, Well, what about other places in the solar system? Why Mars? Well, just to sort of put things into perspective, this is — this is what — this is an actual scale of what the solar system looks like. So we're currently in the third little rock from the left. That's Earth. Yeah, exactly. And our goal is to go to the fourth rock on the left. That's Mars. But you can get a sense for the real scale of the solar system, how big the sun is and Jupiter, Neptune, Saturn, Uranus.
And then the little guys on the right are Pluto and friends. This sort of helps to see, it's not quite to scale, but it gives you a better sense for where things are. So our options for going to — for becoming a multiplanet species within our solar system are limited. We have, in terms of nearby options, we've got Venus. But Venus is a high-pressure — super high-pressure hot acid bath. So that would be a tricky one. Venus is not at all like the goddess. This is not in no way similar to the actual goddess.
So it is really difficult to make things work on Venus. Mercury is also way too close to the sun. We could go potentially on to one of the moons of Jupiter or Saturn, but those are quite far out, much further from the sun, a lot harder to get to. Really, it leaves us with one option if we want to become a multiplanet civilization, and that's Mars. We could conceivably go to our moon, and I have nothing against going to the moon, but I think it's challenging to become multiplanetary on the moon because it's much smaller than a planet. It doesn't have any atmosphere. It's not as resource rich as Mars. It has a 28-day day, whereas the Mars day is 24 1/2 hours. And in general, Mars is far better suited to ultimately scale up to a self-sustaining civilization.
Just to give some comparison between the two planets, there are actually — they're remarkably close in a lot of ways. In fact, we now believe that early Mars was a lot like Earth. And, in fact, if we could warm Mars up, we would, once again, have a thick — a thick atmosphere and liquid oceans. So, where things are right now, Mars is about half again as far from the sun as Earth. So it has decent sunlight. It's a little cold, but we can warm it up. It has a very helpful atmosphere which, in the case of Mars being primarily CO2 with some nitrogen and argon and a few other trace elements, means that we can grow plants on Mars just by compressing the atmosphere. And it has nitrogen, too, which is also very important for growing plants. It will be quite fun to be on Mars, because you will have gravity which is about 37% that of Earth, so you will be able to lift heavy things and bound around and have a lot of fun.
And the day is remarkably close to that of Earth. So we just need to change that bottom row, because currently we have 7 billion people on Earth and zero on Mars. So there's been a lot of great work by NASA and other organizations in early exploration of Mars and understanding what Mars is like, where could we land, what's the composition of the atmosphere, where is there water, or ice, we should say. And we need to go from these early exploration missions to actually building a city. The issue that we have today is that if you look at a Venn diagram, there's no intersection of sets of people who want to go and can afford to go. In fact, right now you cannot go to Mars for infinite money. Using traditional methods, you know, if taking sort of a holistic style approach, an optimistic cost number would be about $10 billion per person.
For example, the Apollo program, the cost estimates are somewhere between $100 billion to $200 billion in current-year dollars, and we sent 12 people to the surface of the moon, which was an incredible thing and I think probably one of the greatest achievements of humanity. But that's — that's a steep price to pay for a ticket. That's why these circles only just barely touch. So you can't create a self-sustaining civilization if the ticket price is $10 billion a person. What we need is a closer — is to move those circles together. And if we can get the cost of moving to Mars to be roughly equivalent to a median house price in the U.S., which is around $200,000, then I think the probability of establishing a self-sustaining civilization is very high.
I think it would almost certainly occur — not everyone would want to go. In fact, I think a relatively small number of people from Earth would want to go. But enough would want to go and who could afford the trip that it would happen. And people could get sponsorship. And I think it gets to the point where almost anyone, if they saved up and this was their goal, they could ultimately save up enough money to buy a ticket and move to Mars. And Mars would have labor shortage for a long time so jobs would not be in short supply. But it is a bit tricky because we have to figure out how to improve the cost of trips to Mars by 5 million percent. So this is — this is not easy. I mean, it's — and it sounds like virtually impossible but I think there are ways to do it. This translates to an improvement of approximately 4 1/2 orders of magnitude. These are the key elements that are needed in order to achieve the 4 1/2 order of magnitude improvement. Most of the improvement would come from full reusability, somewhere between 2 and 2 1/2 orders of magnitude.
And then the other two orders of magnitude would come from refilling in orbit, propellant production on Mars and choosing the right propellant. So I'm going to go into detail on all of those. Full reusability is really the super hard one. It's very difficult to achieve reusability for even an orbital system, and that challenge becomes even substantially greater for a system that has to go to another planet.
But as an example of the difference between reusability and expendability in aircraft — and you can actually use any form of transport. You could say a car, bicycle, horse. If they were single-use, almost no one would use them. It would be too expensive. But with frequent flights, you can take something like an aircraft that costs $90 million and if it was single use, you would have to pay half a million dollars per flight.
But you can actually buy a ticket on Southwest right now from L.A. to Vegas for $43, including taxes. So that's — I mean, that's a massive improvement. Right there it's showing a four order of magnitude improvement. Now this is harder — the reusability doesn't apply quite as much to Mars because the number of times they can reuse the spaceship is — the spaceship part of the system is less often because the Earth-Mars rendezvous only occurs every — every 26 months.
So you get to use the spaceship part roughly every two years. Now, you get to use the booster and the tanker as frequently as you'd like. And so it makes — that's why it makes a lot of sense to load the spaceship into orbit with essentially tanks dry and have it have really quite big tanks that you then use the booster and tanker to refill while it's in orbit and maximize the payload of the spaceships that when it goes to Mars, you really have a very large payload capability. So as I said, refilling in orbit is one of the essential elements of this. Without refilling in orbit, you would have a half order of magnitude impact roughly on the cost.
By "half order of magnitude," I think the audience mostly knows, but what that means is each order of magnitude is a factor of ten. So not refilling in orbit would mean a 500%, roughly, increase in the cost per ticket. It also allows us to build a smaller vehicle and lower the development cost, although this vehicle is quite big. But it would be much harder to build something that's five to ten times the size. And it also reduces the sensitivity of performance characteristics of the booster rocket and tanker. So if there's a shortfall in the performance of any of the elements, you can actually make up for it by having one or two extra refilling trips to the spaceship. So this is — it's very important for reducing the susceptibility of the system to a performance shortfall. And then producing propellants on Mars is actually also very obviously important. Again, if we didn't do this, it would have at least a half order of magnitude increase in the — in the cost of a trip.
So 500% increase in the cost of the trip. And it would be pretty absurd to try to build a city on Mars if your spaceships just kept staying on Mars and not going back to Earth. you would have this, like, massive graveyard of ships. You would have to, like, do something with them. So it really wouldn't make sense to — to leave your spaceships on Mars. You really want to build a propellant plant on Mars and send the ships back. So and Mars happens to work out well for that because it has a CO2 atmosphere, it's got water rights in the soil, and with H2O and CO2 you can produce CH4, methane, and oxygen, O2. So picking the right propellant is also important. Think of this as maybe there's three main choices.
And they have their merits, but kerosene or rocket propellant-grade kerosene which is also what jets use. Rockets use a very expensive form of highly refined form of jet fuel essentially which is a form of kerosene. It helps keep the vehicle size small, but because it's a very specialized form of jet fuel, it's quite expensive. Your reusability potential is lower. Very difficult to make this on Mars, because there's no oil. So really quite difficult to make the propellant on Mars. And then propellant transfer is pretty good but not great. Hydrogen, although it has a high specific impulse, is very expensive, incredibly difficult to keep from boiling off because liquid hydrogen is very close to absolute zero as a liquid.
So the insulation required is tremendous, and the cost of — the energy cost on Mars of producing and storing hydrogen is very high. So when we looked at the overall system optimization, it was clear to us that methane actually was the clear winner. So it would require maybe anywhere from 50 to 60% of the energy on Mars to refill propellants using the propellant depot. And just the technical challenges are a lot easier. So we think — we think methane is actually better on just really almost across the board. And we started off initially thinking that hydrogen would make sense, but ultimately came to the conclusion that the best way to optimize the cost per unit mass to Mars and back is to use an all-methane system, or technically deep-cryo Methalox. So those are the four elements that need to be achieved. So whatever system is designed, whether by SpaceX or anyone, we think these are the four features that need to be addressed in order for the system to really achieve a low cost per — a cost per ton to be of service on Mars.
This is a simulation about the overall system. (Music). (Video). [APPLAUSE]. So what you saw there is really quite close to what we will actually build. It will look almost exactly what you saw — like what you saw. So this is not an artist's impression. The simulation was actually made from the SpaceX engineering CAD models. So this is not — you know, it's not just, well, this is what it might look like. This is what we plan to try to make it look like. In the video, you got a sense for what this system mock architecture looks like. The rocket booster and the spaceship take off, loads the spaceship into orbit. The rocket booster then comes back. It comes back quite quickly, within about 20 minutes. And so it can actually launch the tanker version of the spacecraft, which is essentially the same as the — as the spaceship but filling up the unpressurized and pressurized cargo areas with propellant tanks.
So they look almost identical. This also helps slow the development cost, which obviously will not be small. And then the propellant tanker goes up. It will go — actually, it will go up multiple times, anywhere from three to five times, to fill the tanks of the spaceship in orbit. And then once the spaceship is — the tanks are full, the cargo has been transferred, and we reach the Mars rendezvous timing, which as I mentioned is roughly every 26 months, that's when the ship would depart.
Now, over time there would be many spaceships. You would ultimately have, I think, upwards of a thousand or more spaceships waiting in orbit. And so the Mars colonial fleet would depart en masse. Kind of like Battlestar Galactica, if you have seen that thing. Good show. So it's a bit like that. But it actually makes sense to load the spaceships into orbit because you have got two years to do so and then make frequent use of the booster and the tanker to get really heavy reuse out of those. And then with the spaceship you get less reuse because you have to prepare for how long is it going to last? Well, maybe 30 years.
So that might be 12 to maybe 15 flights with the spaceship at most. So you really want to maximize the cargo of the spaceship and use the booster and the tanker a lot. So the ship goes to Mars, gets replenished, and then returns to Earth. So going into some of the details of the vehicle design and performance — and I'm going to gloss over — I'll only talk a little bit about the technical details in the actual presentation, and then I'll leave the detailed technical questions to the Q and A that follows. This is to give you a sense of size. It's quite big. (Laughter). [APPLAUSE]. The funny thing is in the long-term, the ships will be even bigger than this. This will be relatively small compared to the Mars interplanetary ships of the future. But it kind of needs to be about this size because in order to fit a hundred people or thereabouts in the pressurized section plus carry the luggage and all of the unpressurized cargo to build propellant plants and build everything from iron foundries to pizza joints to you name it, we need to carry a lot of cargo.
So it really needs to be roughly on this sort of magnitude, because if we say like the — that same amount of threshold for a self-sustaining city on Mars for civilization would be a million people. If you only go every two years, if you have a hundred people per ship, that's 10,000 trips. So I think at least a hundred people per trip is the right order of magnitude, and I think we may actually end up expanding the crew section and ultimately taking more like 200 or more people per flight in order to reduce the cost per person. But it's — you know 10,000 flights is a lot of flights. So you really want ultimately on the order of a thousand ships. It will take a while to build up to a thousand ships. And so I think if you say, When would we reach that million-person threshold? From the point at which the first ship goes to Mars, it's probably somewhere between 20 to 50 total Mars rendezvous.
So it's probably somewhere between maybe 40 to 100 years to achieve a fully self-sustaining civilization on Mars. So that's sort of the cross-section of the ship. In some way, it's not that complicated, really. It's made primarily of an advanced carbon fiber. The carbon fiber part is tricky when dealing with deep cryogens and trying to achieve both liquid and gas impermeability and not have gaps occur due to cracking or pressurization that would make the carbon fiber leaky. So this is a fairly significant technical challenge, to make deep and cryogenic tanks out of carbon fiber. And it's only recently that we think the carbon fiber technology has gotten to the point where we can actually do this without having to create a liner, some sort of metal liner, quad liner on the inside of the tanks, which would add mass and complexity.
It's particularly tricky for the hot gaseous oxygen pressurization. So this is designed to be autogenously pressurized, which means that the fuel and the oxygen, we gasify them through heat exchanges in the engine and use that to pressurize the tanks. So we will gasify the methane and use that to pressurize the fuel tank. Gasify the oxygen. Use that to pressurize the oxygen tank. This is a much simpler system than what we have with Falcon 9, where we use helium for pressurization and we use nitrogen for gas thrusters. In this case, we would autogenously pressurize and then use gaseous methane and oxygen for the control thrusters. So really, you only need two ingredients for this, as opposed to four in the case of Falcon 9 and actually five if you consider the ignition liquid.
It's sort of a complicated liquid to ignite the engines. That isn't very usable. In this case we would use spark ignition. So this gives you a sense of vehicles by performance, sort of current and historic. I don't know if you can actually read that. But in expandable mode, the vehicle, of course, we are proposing would do about 550 tons and about 300 tons in reusable mode. That compares to satisfy max capability of 135 tons. But I think this really gives a better sense of things. The white bars show the performance of the vehicle; in other words, the payload-to-orbit of the vehicle. So you can see essentially what it represents is what's the size efficiency of the vehicle.
And most rockets, including ours — ours as they're currently flying — the performance bar is only a small percentage of the actual size of the rocket. But with the interplanetary system which we will initially use for Mars, we've been able to — or we believe massively improve the design performance. So it's the first time a rocket's sort of performance bar will actually exceed the physical size of the rocket. This gives you a more direct sort of comparison. This is — the thrust that is quite enormous, talking about liftoff thrusts of 13,000 tons. So it's quite tectonic when it takes off. But it is — it is a fit on Pad 39A, which NASA has been kind enough to allow us to use, where — because they somewhat oversized the pad in doing Saturn 5 and, as a result, we can actually do a much larger vehicle on that same launch pad.
And in the future, we expect to add additional launch locations, probably adding one on the south coast of Texas. But this gives you a sense of the relative capability, if you can read those. But these vehicles have very different purposes. This is really intended to carry huge numbers of people, ultimately millions of tons of cargo to Mars. So you really need something quite large in order to do that. So talk about some of the key elements of the interplanetary spaceship and rocket booster. We decided to start off the development with what we think are probably the two most difficult elements of the design. One is the Raptor engine. And this is going to be the highest chamber pressure engine of any kind ever built and probably the highest thrust-to-weight.
It's a full-flow staged combustion engine which maximizes the theoretical momentum that you can get out of a given source fuel and oxidizer. We subcool the oxygen and methane to densify it. So compared to when — propellants normally use close to their boiling point in most rockets. In our case, we actually build the propellants close to their freezing point. That can result in a density improvement of up to around 10 to 12%, which makes an enormous difference in the actual results of the rocket. It also makes the — it gets rid of any cavitation risk for the turbo pumps and it makes it easier to feed a high-pressure turbo pump if you have very cold propellant.
Really one of the keys here, though, is the vacuum version of Raptor having a 382-second ISP. This is really quite critical too to the whole Mars mission. And we can get to that number or at least within a few seconds of that number, ultimately maybe exceeding it slightly. So the rocket booster in many ways is really a scaled-up version of the Falcon 9 booster. You will see a lot of similarities, such as the grid fins. Obviously clustering a lot of engines at the base. And the big difference really being that the primary structure is an advanced form of carbon fiber as opposed to limited lithium and that we use autogenous pressurization and get rid of the helium and the nitrogen. So this uses 42 Raptor engines. It's a lot of engines, but we use an I.N. on the Falcon 9. And with Falcon Heavy, which should launch early next year, there's 27 engines on the base. So we've got pretty good experience with having a large number of engines.
It also gives us redundancies. So that if some of the engines fail, you can still continue the mission and be fine. But the main job of the booster is to accelerate the spaceship to around 8 1/2 thousand kilometers an hour. For those that are less familiar with orbital dynamics, really it's all about velocity and not about height. So really that's the job of the booster. The booster is like the javelin thrower. You've got to toss that javelin, which is the spaceship. In the case of other planets, though, which have a gravity well which is not as deep, so Mars, the moons of Jupiter, conceivably maybe even one day Venus — the — well, Venus will be a little trickier.
But for most of the solar system, you only need the spaceship. So you don't need the booster if you have a lower gravity well. No booster is needed on the moon or Mars or any of the moons of Jupiter or Pluto. You just need the spaceship. The booster is just there for heavy gravity wells. And then we've also been able to optimize the propellant needed for boost-back and landing to get it down to about 7% of the liftoff prop propellant load. We think with some optimization maybe we can get it down to about 6%. And we also are now getting quite comfortable with the accuracy of the landing. If you have been watching the Falcon 9 landings, you will see that they are getting increasingly closer too to the bull's-eye. And we think, particularly with the addition of additional — with the addition of some thrusters and maneuvering thrusters, we can actually put the booster right back on the launch stand. And then those fins at the base are essentially centering features to take out any minor position mismatch at the launch site.
So that's what it looks like at the base. So we think we only need to gimbal or steer the center cluster of engines. There's seven engines in the center cluster. Those would be the ones that move for steering the rocket, and the other ones would be fixed in position, which gives us the best concentration of — we can max out the number of engines because we don't have to leave any room for gimbaling or moving the engines. And, like, this is all designed so that you could actually lose multiple engines even at liftoff or anywhere in flight and continue the mission safely. So for the spaceship itself, in the top, we have the pressurized compartment. And I'll show you a fly-through of that in a moment.
And then beneath that is the — is where we would have the unpressurized cargo, which would be really flat packed in a very dense format. And then below that is the liquid oxygen tank. The liquid oxygen tank is probably the hardest piece of this whole vehicle because it's got to handle propellant at the coldest level and the tanks themselves actually form the air frame. So the air frame structure and the tank structure are combined, as it is in all modern rockets. And in aircraft, for example, the wing is really a fuel tank in wing shape. So it has to take the thrust loads of ascents, the loads of reentry, and then it has to be impermeable to gaseous oxygen, which is tricky, and non-reactive to gaseous oxygen. So that's the hardest piece of the spaceship itself, which is actually why we started on that element as well. And I'll show you some pictures of that later. And then below the oxygen tank is the fuel tank, and then the engines are mounted directly to the thrust cone on the base.
And then there are six of the vacuum — the high efficiency vacuum engines around the perimeter, and those don't gimbal. And then there are three of the sea-level versions of the engine which do gimbal and provide the steering. Although we can do some amount of steering if you're in space with differential thrust on the outside engines. The net effect is a cargo to Mars of up to 450 tons, depending upon how many refills you do with the tanker. And the goal is at least 100 passengers per ship. Although I think we will see that number grow to 200 or more. This chart is a little difficult to interpret at first, but we decided to put it there for people who wanted to watch the video afterwards and sort of take a closer look and analyze some of the numbers. The column on the left is probably what's most relevant. And that gives you the trip time. So depending upon which Earth-Mars rendezvous you are aiming for, the trip time at 6 kilometers per second departure blast speed can be as low as 80 days. And then over time, I think we could probably improve that.
Ultimately, I suspect that you would see Mars transit times of as little as 30 days in the more distant future. It's fairly manageable, considering the trips that people used to do in the old days would routinely take sailing voyages that would be six months or more. So on arrival, the heat shield technology is extremely important. We have been refining the heat shield technology using our Dragon spacecraft.
We are now on version 3 of PICA, which is the phenolic-impregnated carbon ablator. And it's getting more and more robust with each new version, with less ablation, more resistance, less need for refurbishment. The heat shield is basically a giant brake pad. How it's like how good can you make that brake pad against the extreme conditions and the cost of refurbishment and make it so you could have many flights with no refurbishment at all. This is a fly-through of the crew compartment. I just want to give you a sense of what it would feel like to actually be in the spaceship.
I mean, in order to make it appealing and increase that portion of the Venn diagram of people who actually want to go, it's got to be really fun and exciting, and it can't feel cramped or boring. But the crew compartment or the occupant compartment is set up so you can do zero-G things, you can float around. It would be like movies, ElectroPuls, cabins, a restaurant. It will be, like, really fun to go. You are going to have a great time. (Laughter). So the propellant plant on Mars, again, this is one of those slides that I won't go into in detail here, but people can take that offline. The key point being that the ingredients are there on Mars to create a propellant plant with relative ease, because the atmosphere is primarily CO2 and there's water ice almost everywhere. You've got the CO2 plus H2O to make methane CH4 and oxygen O2 using the Sabatier reaction. The trickiest thing really is the energy source, which think we can do with a large field of solar panels.
So then to give you a sense of the cost, really the key is making this affordable to almost anyone who wants to go. And we think, based on this architecture, this architecture, assuming optimization over time, like the very first flights would be fairly expensive. But the architecture allows for a cost per ticket of less than $200,000, maybe as less — maybe as little as $100,000 over time, depending upon how much mass a person takes.
So we're right now estimating about $140,000 per ton to the trips to Mars. So if a person plus their luggage is less than that, take into account food consumption and life support, then we think that the cost of moving to Mars ultimately could drop below $100,000. So funding, talking about funding sources. So we have steel underpants; launch satellites; send cargo to space station; Kickstarter, of course; followed by profit. So obviously it's going to be a challenge to fund this whole endeavor. We do expect to generate pretty decent net cash flow from launching lots of satellites and serving the space station for NASA, transferring cargo to and from space station, and then I know there's a lot of people in the private sector who are interested in helping fund a base on Mars and then perhaps there will be interest on the government sector side to also do that. Ultimately, this is going to be a huge public-private partnership. And I think that's — that's how the United States was established, and many other countries around the world, is a public-private partnership.
So I think that's probably what occurs. And right now we're just trying to make as much progress as we can with the resources that we have available and just sort of keep moving both forward. And, hopefully, I think as we — as we show that this is possible, that this dream is real, not just a dream, it is something that can be made real, I think the support will snowball over time. And I should say also the main reason I'm personally accumulating assets is in order to fund this. So I really don't have any other motivation for personally accumulating assets except to be able to make the biggest contribution I can to making life multiplanetary. [APPLAUSE]. Time lines. Not the best at this sort of thing. But just to show you where we started off. In 2002, SpaceX basically consisted of carpet and a mariachi band. That was it. That's all of SpaceX in 2002. As you can see, I'm a dancing machine. And, yeah, I believe in kicking off celebratory events with mariachi bands.
I really like mariachi bands. But that was what we started off with in 2002. And really, I mean, I thought we had maybe a 10% chance of doing anything, of even getting a rocket to orbit, let alone getting beyond that and taking Mars seriously. But I came to the conclusion if there wasn't some new entrant into the space arena with a strong ideological motivation, then it didn't seem like we were on a trajectory to ever be a space-faring civilization and be out there among the stars.
Because, you know, in '69 we were able to go to the moon and the space shuttle could get to low-Earth orbit, and then after the space shuttle got retired. But that trend line is down to zero. So I think what a lot of people don't appreciate is that technology does not automatically improve. It only improves if a lot of really strong engineering talent is applied to the problem that it improves. And there are many examples in history where civilizations have reached a certain technology level and then have fallen well below that and then recovered only millennia later. So we go from 2002 where we're basically — we're clueless. And then with Falcon 1, the smallest useful little rocket that we could think of which would deliver a half a ton to orbit, and then four years later we developed the — we built the first vehicle. So we dropped the main engine, the upper stage engine, the air frames, the fairing and the launch system and had our first attempt at launch in 2006, which failed. So that lasted about 60 seconds, unfortunately. But it's 2006, four years after starting, is also when we actually got our first NASA contract.
And I just want to say I'm incredibly grateful to NASA for supporting SpaceX, you know, despite the fact that our rocket crashed. Of course, I'm NASA's biggest fan. So, you know, thank you very much to the people that had the faith to do that. Thank you. [APPLAUSE]. So then 2006, followed by a lot of grief. And then, finally, the fourth launch of Falcon 1 worked in 2008. And we were really down to our last pennies.
In fact, I only thought I had enough money for three launches and the first three bloody failed. And we were able to scrape together enough to just barely make it and do a fourth launch. And thank goodness that fourth launch succeeded in 2008. That was a lot of pain. And then also at the end of 2008 is when NASA awarded us the first major operational contract, which was for resupplying cargo to the space station and bringing cargo back. Then a couple years later we did the first launch of Falcon 9, version 1.
And that had about a 10-ton-to-orbit capability. So it was about 20 times the capability of Falcon 1, and also was assigned to carry our Dragon spacecraft. Then 2010 is our first mission to the space station. So we were able to finish development of Dragon and dock with the space station in 2010. so — Sorry, 2010 is expendable Dragon — expendable Dragon. 2012 is when we delivered and returned cargo from the space station.
2013 is when we first started doing boat take-off and landing tests. And then 2014 is when we were able to have the first orbital booster do a soft landing in the ocean. The landing was soft. The (inaudible) exploded. But the landing — for seven seconds, it was good. And we also improved the capability of the vehicle from 10 tons to about 13 tons to LEO. And then 2015, last year, in December, that was definitely one of the best moments of my life when the rocket booster came back and landed at Cape Canaveral. That was really … [APPLAUSE]. Yeah. So that really showed that we could bring an orbit-class booster back from a very high velocity all the way to the launch site and land it safely and with almost no refurbishment required for reflight. And if things go well, we're hoping to refly one of the landed boosters in a few months. So, yeah — and then 2016, we also demonstrate landing on a ship. The landing on the ship is important for the very high-velocity geosynchronous missions. And that's important for reusability of Falcon 9 because about roughly a quarter of our missions are sort of servicing the space station.
But then there's a few other low-Earth-orbit missions. But most of our missions, probably 60% of our missions, are commercial geo missions. So we've got to do these high-velocity missions that really need to land on a ship out to sea. They don't have enough propellants on board to boost back to the launch site. So looking into the future, next steps, we were kind of intentionally a bit fuzzy about this time line. But we were going to try to make as much progress as we can. Obviously, it's with a very constrained budget. But we are going to try to make as much progress as we can on the elements of interplanetary transport booster and spaceship, and hopefully we'll be able to complete the first development spaceship in maybe about four years and start doing suborbital flights with that.
In fact, it has enough capability that you could maybe even go to orbit if you limit the amount of cargo with the spaceship. Well, you have to really — you have to really strip it down. But in tanker form, it could definitely get to orbit. It can't get back, but it can get to orbit. Actually, I was thinking like maybe there is some market for really fast transport of stuff around the world, provided we can land somewhere where noise is not a super big deal, because rockets are very noisy. But we could transport cargo to anywhere on Earth in 45 minutes at the longest. So most places on Earth would be maybe 20, 25 minutes. So maybe if we had a floating platform off the coast of, you know, say — off the coast of New York, say 20, 30 miles out, could you go from New York to Tokyo in, I don't know, 25 minutes; across the Atlantic in ten minutes.
Really most of your time would be getting to the ship, and then it would be real quick after that. So there's some intriguing possibilities there. Although, we're not counting on that. And then development of the booster — we actually think the booster part is relatively straightforward because it's — it amounts to a scaling up of the Falcon 9 booster. So there's — we don't see a lot of sort of show-stoppers there. Yeah. But then trying to put it all together and make this actually work to Mars, if things go super well, it might be kind of in the ten-year time frame.
But I don't want to say that's when it will occur. It's, like, this huge amount of risk. It's going to cost a lot. Good chance we don't succeed, but we're going to do our best and try to make as much progress as possible. And we're going to try to send something to Mars on every Mars rendezvous from here on out. So Dragon 2, which is a propulsive lander, we plan to send to Mars in a couple years, and then do probably another Dragon mission in 2020. In fact, we want to establish a steady cadence, that there's always a flight leaving, like there's a train leaving the station.
With every Mars rendezvous we will be sending a Dragon — at least a Dragon to Mars and ultimately the big spaceship. So if there's a lot of interest in putting payloads on Dragon, you know you can count on a ship that's going to transport something on the order of at least two or three tons of useful payloads to the surface of Mars. [APPLAUSE]. That's part of the reason we designed Dragon 2, to be a propulsive lander. As a propulsive lander, you can go anywhere in the solar system. So you can go to the moon. You can go to — well, anywhere, really.
Whereas, if something relies on parachutes or wings, then you can pretty much only — well, if it's wings, you can pretty much only land on Earth because you need a runway, and most places don't have a runway. And then anyplace that doesn't have a dense atmosphere, you can't use parachutes. But propulsive works anywhere. So the Dragon should be capable of landing on any solid or liquid surface in the solar system. I was real excited to see that the team managed to do the — all our Raptor engine firing in advance of this conference. I just want to say thanks to the Raptor team for really working seven days a week to try to get this done in advance of the presentation, because I really wanted to show that we've made some hardware progress in this direction.
And the Raptor is a really tricky engine. It's a lot trickier than Merlin because it's a full-flow stage combustion, much higher pressure. I'm kind of amazed it didn't blow up on the first firing. Fortunately, it was good. It's kind of interesting to see the mock diamonds forming. [APPLAUSE]. And part of the reason for making the engine sort of small, Raptor, although it has three times the thrust of a Merlin is only about the same size as a Merlin engine because it has three times the operating pressure. That means we can use a lot of the production techniques that we've honed with Merlin. We are currently producing Merlin engines at almost 300 per year. So we understand how to make rocket engines in volume.
Even though the Mars vehicle uses 32 on the base and 9 on the upper stage, so we are at 51 engines to make — that's well within our production capabilities for Merlin. And this is a similarly sized engine to Merlin except for the expansion ratio. So we feel really comfortable about being able to make this engine in volume at a price that doesn't break our budget. We also wanted to make progress on the primary structure. So, as I mentioned, this is really a very difficult thing to make, to make something out of carbon fiber. Even though carbon fiber has incredible strength-to-weight, when you want to then put super cold liquid oxygen or liquid methane — particularly liquid oxygen — in a tank, it's subject to cracking and leaking, and it's very difficult to make. Just the sheer scale of it is also challenging, because you've got to lay out the carbon fiber in exactly the right way on a huge mold, and you've got to cure that mold at temperature.
And then — it's just hard to make large carbon fiber structures that could do all of those things and carry incredible loads. So that's the other thing we want to focus on is the Raptor and then building the first development tank for the Mars spaceship. So this is really the hardest part of the spaceship. The other pieces we have a pretty good handle on. But this was the trickiest one. We wanted to tackle it first. You get a size for how big the tank is, which is really quite big. Also big congratulations to the team that worked on that. They were also working seven days a week to try to get this done in advance of the IAC. We managed to build the first tank, and the initial tests with the cryogenic propellants actually look quite positive. We have not seen any leaks or major issues. This is what the tank looks like on the inside.
So you can get a real sense for just how big this tank is. It's actually completely smooth on the inside, but the way that the carbon fiber plies lay out and reflect the light makes it look faceted. So then what about beyond Mars? So as we thought about the system — and the reason we call it a system, because generally I don't like calling things systems because everything is a system, including your dog — is that — is that it's actually more than a vehicle.
There's obviously the rocket booster, the spaceship, the tanker, and the propellant plant, the in situ propellant production. If you have all of those four elements, you can actually go anywhere in the solar system by planet hopping or moon hopping. So by establishing a propellant depot on — in the asteroid belt or on one of the moons of Jupiter, you can go to — you can make flights from Mars to Jupiter no problem. In fact, even from — even without a propellant depot at Mars, you can do a fly-by of Jupiter without a propellant depot.
So — but by establishing a propellant depot, let's say, you know, Enceladus or Europa or — there's a few options, and then doing another one on Titan, Saturn's moon, and then perhaps another one further out on Pluto or elsewhere in the solar system, this system really gives you freedom to go anywhere you want in the greater solar system. So you can actually travel out to the Kuiper belt or the Earth cloud. I wouldn't recommend this for interstellar journeys, but this — just this basic system, provided we have filling stations along the way, is — means full access to the entire greater solar system. [APPLAUSE]..