For a manned mission to Mars that isn't a one-way trip, we'll need an ascent stage that can reach orbit from Mars' surface. How big would this need to be to transport e.g. 3 astronauts to orbit?
We've seen examples for Earth (Soyuz comes to mind as a sort-of minimal solution at 600 tons loaded) and the Moon (LM ascent stage was roughly the size of a car). A Mars ascent stage would fall somewhere between these extremes, but where?
Would it be feasible to launch the descent + ascent stages from Earth on a single launcher, or would it need to be assembled in orbit?

  • $\begingroup$ Just to be clear, you're talking about a launched from Earth rocket, fully fueled on Earth? $\endgroup$
    – PearsonArtPhoto
    Commented Nov 18, 2013 at 17:19
  • $\begingroup$ Yes on both counts. $\endgroup$
    – Hobbes
    Commented Nov 18, 2013 at 17:23
  • $\begingroup$ ether way a mars assent vehicle designed to carry 3 to 4 crew members would be big. probably the biggest vehicle needed for the mars mission. thats where I think one or possibly two launches of the upcoming SLS rocket will come in. if they can get the Block 2 cargo flying by that time they probably would be maxing out the 140 tone lift capacity of the SLS just launching the mars assent vehicle. its possible but it would be very expensive and challenging. its one of the reasons NASA predicts a human mission to mars won't happen untill the 2030's or 2050's. $\endgroup$
    – user12104
    Commented Oct 26, 2015 at 15:55

4 Answers 4


Okay, so let's start from a ship somehow constructed and fueled on Mars, and work our way back from that. This convenient delta v chart from Wikipedia shows us a good starting point:

enter image description here

Okay, so the delta v to Low Mars Orbit is about 4.1 km/s. The delta V to get to Earth from there is 2.9 km/s (Earth C3 to Low Mars Orbit). Bottom line is, you need to lift 3 people and supplies and provide it with a delta v of about 6.0 km/s. I'm not an expert in the supplies category, so let's just pick a number, say, let's carry a Dragon capsule fully loaded. Wikipedia gives about 7500 kgs, which is probably a fair starting point. That's not quite as much as required for an Earth Launch, but would be similar to a Falcon 9 launch from Mars to Earth. Let's call it half of one. In fact, The Case for Mars states that around 96 tons of fuel is required to make the trip for 4. Few (None existing) rockets can take that to LEO, let alone to the Martian surface. Still, there is an alternative that doesn't involve orbital assembly.

The key focus is 2 points. First of all, fuel the rocket on the surface of Mars. If you take hydrogen from Earth, you can easily combine it with carbon dioxide and have methane and oxygen, which is a very good rocket fuel. And that requires only 16 tons of fuel, not too bad to manage. Secondly, send the return rocket to the surface of Mars first, let it fuel itself for the return trip home, and you simply have to land close enough to the ship to get there on the surface, which shouldn't be too difficult, as you will have a large range with on board capabilities to get where you need to go.

  • $\begingroup$ It's not quite reciprocal. It's going to take something more to get off Mars than it would take to land on Mars. Given how nasty the rocket equation is this can't be ignored. $\endgroup$ Commented Nov 19, 2013 at 19:52
  • $\begingroup$ Truth is, it just takes a parachute for the most part to land on Mars, and far more to take off, but the delta v will be only slightly greater than 4.1 to get in to Mars orbit. $\endgroup$
    – PearsonArtPhoto
    Commented Nov 19, 2013 at 19:57
  • $\begingroup$ Yeah, but I assumed you were taking the delta-v figure off that image. It always takes a bit more to go up than down and that number is for down. $\endgroup$ Commented Nov 20, 2013 at 20:22
  • $\begingroup$ Is this due to atmospheric drag? $\endgroup$
    – Hobbes
    Commented Nov 21, 2013 at 10:37

I plugged the data provided by @PearsonArtPhoto into the rocket equation, with Isp=304 seconds (Merlin 1C in vacuum), and I get a launch mass of 262 tons to get a Dragon into Mars orbit. A Falcon 9v1.0 weighs 333 tons.
To get from Mars surface directly to Earth C3, I get a launch mass of 3713 tons, or 1.3 times the weight of a Saturn V. So you really need a Mars Orbit rendezvous on the return trip.

  • $\begingroup$ Nice analysis, but did you use the weight of an entire Dragon rocket, or just the capsule? Not sure why your launch mass is so high. It seems like a Saturn 5 would be enough. With a dry mass of 200 metric tonnes(estimated Saturn 5 dry mass), and wet mass of 2970 metric tonnes, and ISP of 304, I get 8043 km/s dV. This is more than enough for the return trip to Earth C3 according to the dV map. I am using this calculator: strout.net/info/science/delta-v $\endgroup$
    – AaronLS
    Commented Mar 12, 2017 at 5:51
  • $\begingroup$ it's been a while, but I think I used the loaded weight of a Dragon capsule+trunk. $\endgroup$
    – Hobbes
    Commented Mar 12, 2017 at 11:25

The Mars ascent stage would need to mass approximately 22,000 kg, assuming an ascent vehicle very similar to the complete Apollo lunar lander (LM), a two-stage vehicle using hypergolic propellant rocket engines.

I base this on the roughly 4 km/sec delta v astronautix.com describes for the LM, and multiply the mass of the LM by 1.5 to account for the increase of crew to three.


Would it be feasible to launch the descent + ascent stages from Earth on a single launcher, or would it need to be assembled in orbit?

As Hobbes and PearsonArtPhoto calculated, it is not feasible to start from Earth as there are no rockets ready, not even from United Launch Alliance or SpaceX, BlueOrigin or Bigelow Space Exploration.

The question is, why not to assemble in space?

Furthermore taking Hydrogen from Earth to Mars distorts the fact that 90% of known visible matter in the universe is hydrogen (H2), the lightest element we know. Why take it from Earth where hydrogen resources are really limited. It's interesting seeing demand for Hydrogen cars on our planet - extracting the H2 out of water H2O via Electrolysis - and at the same time fear that the next war will be about water resources ...

No doubt Hydrogen is a good choice for energy storage in space (where there is plenty of H2).

Back to space assembly, there is a factory in orbit originally planned for 2014, now unfortunately postponed to 2015: Bigelow Inflatable Space Module.

Bigelow BA330 project originally should be plugged together to yield a huge factory habitat. I now remember it's been called Complex Alpha for 16 and Complex Bravo for 32 workers. manufacturer website, also mentions year 2014 for launch.

  • $\begingroup$ I'm not sure which question you're answering, but it isn't the one at the top. See the comments under it, if that's unclear, but the question is about ascent stages from the Martian surface. Your first point is also rather moot, if you then mention orbital assembly. ULA's Delta IV Heavy is a perfectly capable launcher then, of course assuming multiple launches to lift all the required supplies to Earth orbit first. And there are other launch systems that are operational and come close to Delta IV-H in lifting capacity, mind you, not American, such as Proton-M or Ariane 5. $\endgroup$
    – TildalWave
    Commented Nov 21, 2013 at 13:56
  • $\begingroup$ At the risk of causing a lynch mob to gather, why don't we just wait until we establish a factory on the Moon that can build entire (or at least 90+%) spacecraft for Mars? $\endgroup$ Commented Jun 20, 2014 at 20:12
  • $\begingroup$ There is some point to it, the moon could be used for amazing epicness. We don't want to abuse it, of course - and it's still a huge distance so we must accelerate, maybe shift efforts towards a great overall world instead of egoism - should be a boost, because war is costly (as seen by Chinese policy in the medieval ages to prevent war, e.g. preferring buying peace instead, which at occasions may be cheaper ... though not liking money anyway, perhaps all money and gold / underlying value should be shot to the moon ... then surely mankind will not take long to settle on Mars haha). $\endgroup$
    – Radagast
    Commented Jan 13, 2016 at 10:17
  • $\begingroup$ Mars surface is closer in terms of delta-v requirements than lunar surface. $\endgroup$
    – ORcoder
    Commented Feb 27, 2018 at 22:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.