The Dragon capsule of SpaceX is seriously considered by NASA for a sample return mission from Mars in 2022, according to Larry Lemke.

Could a Dragon be used also for hopping around on the Moon? Hoover over its surface? Land through skylights of lunar caves? Will the Dragon become a general purpose spaceship with low costs because of serial production and reusability?

Since it wouldn't return to Earth and the Moon has no atmosphere, I suppose its heat shield would be replaced by a lower mass structure or instruments. But otherwise, would any other modifications be necessary for repurposing a used LEO-Dragon for interplanetary missions, except the payload inside it?

According to the frequent poster here HopDavid (if I'm not mistaken), a delta-v of 1.68 km/s is enough to jump from any point of the Moon to any other point, for example from pole to pole. What delta-v could a landed Dragon feasibly achieve, given a certain fraction of its payload being fuel tanks?

  • 1
    $\begingroup$ Thanks! A few notes: In addition to 1.68 for launch, you'd need another 1.68 for a soft landing. There will be a little gravity loss during launch and descent. $\endgroup$
    – HopDavid
    Commented Jul 23, 2014 at 11:27
  • 1
    $\begingroup$ I believe Dragon relies on the booster and upper stage for most of the delta-V to reach LEO. Once in LEO it needs enough delta-V to rendezvous with the ISS and then enough to lower its perigee into the upper atmosphere. It's a guess on my part but I'd say the Dragon has around .3 km/s delta V. $\endgroup$
    – HopDavid
    Commented Jul 23, 2014 at 11:29

2 Answers 2


The Manned Dragon capsule, the Dragon 2, is designed to be able to do a powered escape from a launch stack. That requires at least 1.1 Gees thrust (and ISTR a requirement of 3 Gees). Further, it's designed to perform a powered landing after escaping the launch stack. But it's also required to be able to detach in flight at any altitude, including LEO, and perform a controlled descent to soft landing on a hard surface..

The Dragon 2 capsule has "120,000 pounds of axial thrust"... 60 tons, for an easier measure. (Note that each of the 8 superdraco motors is capable of 8 tons, but due to angles, 3.5 of the 64 tons of total thrust is non-axial (a fancy way of saying sideways), and the rest is throttled back for safety reasons.

At liftoff, the Dragon 1 masses some 26,000 pounds (including 6000 pounds payload capacity); the Dragon 2 probably masses around 30,000 pounds (15 Tons). This means that, on 2 engines, it is probably capable of 1G of thrust.

Two superdraco engines are required as a bare minimum for flight (due to load balancing), and preferably 4, due to the configuration of the thruster nacelles. Each engine has a 25 second duration on internal fuel (according to the wikipedia entry). Assuming that, as with other SpaceX engines, they can be throttled back to about 65%, this puts the minimum thrust to about 19.5 tons, or roughly 0.66 Gee, and possibly up to 38 seconds thrust.

If we assume even deeper throttleback (Elon Musk has stated a goal of 30%), then we could get this down to 10 tons of thrust, and up to about 80 seconds. And there are 4 such pairs.

The notes on the Dragon 1 capsule give an endurance on orbit of 2 weeks. The Dragon 1 is a parachutes-to-water design.

Also note: Earth landing requires 6 of the 8 superdraco engines working. So, if aiming for earth return, either refuelling or limited use are required.

On the Moon

The moon's roughly 0.16 Gee gravity makes the Dragon 2 thrust an excess. In the case of hopping, this means it will do exactly that. Assuming the exhibited 65%, it's going to have to use bursts.

Landing from Earth orbit requires at least 2 pairs. That leaves room for landing and a few hops on the moon. dry-landing on earth is likely to need well more than that.

The capsule itself seems, from the very limited data available, to be suitable for this role.

The capsule, however, relies upon the trunk for the solar panel connections. Those are not a suitable fit for the landing, unless the trunk is fitted with landing gear.

Further, the capsule door is some 6' off the ground when landed. Use as a taxi would require a ladder.

In such a taxi use, the total of about two hundred seconds of thrust and lack of solar panels would necessitate a support infrastructure.

Delta-V estimates

Noting that NASA gave about a 25% above minimum delta-V, as a safety margin. The total delta-V on the Apollo LM was about 15,380 feet per second, just a hair over 4.7 km/s total delta V. Low Lunar Orbit (LLO) to Lunar Surface is about 1.8 km/s delta V.

The best estimates of current Dragon V2 delta V put it around 1.6 km/s delta V. On the other hand, the capsule itself is rigged for external fuel connections, in addition to its internal fuel, so a modified "landing trunk" is a very real possibility. Adding a trunk with another 3-4 km/s worth of fuel, and some landing legs and a ladder, and you have your down-and-back, with the majority of mass beneath the center of thrust (which is a fairly stable condition).

Note that LEO to Earth surface needs about 4.3 km/s of delta-V... most of which Dragon is slated to achieve by drag from Aerobraking. (Versus the Apollo CM, which did it all by drag - first by aerobraking, then by parachutes.)

Bottom Line

It could be pressed into such a use. It would not be optimal. It's not actually designed for this role.


https://en.wikipedia.org/wiki/Delta-v_budget https://en.wikipedia.org/wiki/Apollo_Lunar_Module https://www.reddit.com/r/spacex/comments/33f75a/just_how_much_deltav_will_the_dragon_2_capsule/

  • $\begingroup$ "The capsule, however, relies upon the trunk for the solar panel connections." Is it? So what about the Red Dragon to Mars? $\endgroup$
    – LocalFluff
    Commented Jul 24, 2014 at 6:21
  • $\begingroup$ I'm going solely by what's released as stable designs, and direct speculation from those. $\endgroup$
    – aramis
    Commented Jul 25, 2014 at 1:50
  • $\begingroup$ What is Dragon's delta V budget? If it's .3 km/s as I suspect, it'd be good for only very short lunar hops. $\endgroup$
    – HopDavid
    Commented Jul 25, 2014 at 15:39
  • $\begingroup$ @HopDavid ≅ 4G x 25s ≅ 9.8 x4 x25 ≅ 980 m/s or just about 1 km/s $\endgroup$
    – aramis
    Commented Jul 27, 2014 at 6:47
  • $\begingroup$ NTO/MMH has exhaust velocity of about 3.34 km/s. Plugging in a 1 km/s delta V budget I get the Dragon is about 26% propellent. Actually it'd be a little more since not all the thrust is axial. $\endgroup$
    – HopDavid
    Commented Jul 29, 2014 at 12:15

Dragon V2 has more than enough delta-v to land on Earth. Only problem on Moon it would be too much. SuperDraco engines are throttleable (~7%) and redundant so fewer engines could be used, but still - too much engine. Probably best solution is to run off some scaled versions of SuperDraco on their 3D printer and push through test program. SuperDraco's incompatibility is not a deal-breaker due to SpaceX's CAM flexibility.

  • 2
    $\begingroup$ I believe the reusable Dragons will rely on aerobraking for most of the entry descent and landing delta-v. It will need a burn at the final part of descent for a soft landing. But this burn would be enough to kill terminal velocity. en.wikipedia.org/wiki/Terminal_velocity $\endgroup$
    – HopDavid
    Commented Jul 23, 2014 at 17:50

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.