Using Crew Dragon for Lunar Gateway

[DreamyPickle]

How about using 1 or 2 of the as of yet developed lunars starship landing engines?

Those are methane/oxygen pressure-fed thrusters, it does not make much sense to use them on a vehicle that doesn't also use Raptor. There is a NSF thread will lots of speculation that their fuel system will be tightly integrated with Raptor: https://forum.nasaspaceflight.com/index.php?topic=46574.0

Apollo and Orion use hypergolics on their service module for good reasons and SpaceX is not working on an equivalent.

[spacenut]

Apollo used hydrolox for their Service Module, which also produced their electricity.  Small hypergolic thrusters were used on the Command Module during re-entry. 

[TomH]

Apollo used hydrolox for their Service Module, which also produced their electricity.  Small hypergolic thrusters were used on the Command Module during re-entry.

Apollo service module used hypergolic propellants. It used liquid hydrogen and oxygen only for the fuel cells to provide electricity with the by-product being water that was used for consumption by the astronauts. However, as I used to tell my students, don't take my word for it; go look it up.

[kkattula]

Dragon 2, by itself, doesn't have the delta v (after TLI) to get to to NRHO, let alone back.  Unmodified D2 delta v is approximately 300 m/s. However, from early reports it does have a heat-shield capable of lunar return, although they'd want to test it before crewed lunar flights. It also has the necessary duration, when docked to a supporting vehicle like ISS or Gateway.

So, assuming fairly standard D2 is launched with crew to LEO on a Falcon 9, how do you get it to Gateway, and later bring it back? Assuming D2 can handle the Gateway docking and proximity operations, there are 3 remaining propulsion tasks:

1)  Trans Lunar Injection (TLI)  approx. 3200 m/s
2)  Powered Lunar Flyby and NRHO Insertion  approx. 450 m/s
3)  NRHO Departure and Powered Lunar Flyby  approx. 450 m/s

2) & 3) Together, requires a long duration vehicle that can provide docking and power for D2, as well as propulsion. IMO, this is not too different from the Dragon XL that SpaceX are already developing for Gateway Logistics. Replacing the pressurized cargo module with larger propellant tanks holding 7 tons in total, and assuming a dry mass of 6 tons, it would have over 900 m/s delta v with a 13 ton D2 docked.  Using small Draco thrusters for all maneuvers, there wouldn't be any issues with over stressing the docking adaptor or high G 'eyes out' burns. It would also need some sort of adaptor on the other end to dock with the TLI stage. Such a Dragon XXL could be launched to LEO by an F9, with plenty of loiter time.

1) Requires a high energy stage, but cannot be too high thrust, to again avoid over stressing the docking adaptor or high G 'eyes out' burns:

 - F9 2nd Stage would be too much thrust and wouldn't have enough delta v, even if launched with no payload by FH (5 tons dry plus 27 tons propellant in LEO).

 - A fully fueled Centaur in LEO would be ok on the thrust, but still not nearly enough delta v.

 - A fully fueled DCSS 5m would be enough, maybe launched on a Falcon Heavy? But launch would have to be synchronized within a few hours of D2 launch. Also would need additional station keeping abilities to allow Dragon XXL to dock.

What's really needed is something that can loiter in LEO for weeks or months, like a very, very large version of Dragon XXL, call it Dragon XXXXL that would perform all 3 operations.  However, it would have to be around 80 tons wet, so too heavy for even F9H to lift.


How about 2 x Dragon XXXL?  Each one about 10 tons empty with 40 tons of hypergolic propellant and a high ER, vacuum, 310+ Isp engine (similar to AJ10-190).  Each would be launched by F9H, one (call it X2) with a temporary inter-stage with docking adaptor on the tail, say 2 tons. They would all meet in LEO, X2's tail docking to X1's nose, then D2 docking to X2's nose. The burns would be:

a) X1 performs about 40% of the TLI, 1300 m/s, and is discarded empty, as is the inter-stage.
b) X2 performs the remaining 1900 m/s of the TLI, using about 29 tons of propellant.
c) X2 performs the Powered Lunar Flyby and NRHO Insertion burns, using about 5 tons of propellant
d) X2 performs the NRHO Departure and Powered Lunar Flyby burns, using about 4 tons of propellant

All this would need optimizing. e.g. they would probably want more engines on X1 to reduce gravity losses, there's about 200 m/s margin for RCS split between X1 & X2, FH reusable vs expended payload, etc. The propellant would fit into a couple of 3m diameter, near spherical tanks, so these aren't huge vehicles.


Disclaimer:  SpaceX are of course focused on Starship, so they would not be interested in anything like this now.

[Negan]

Unless there's an issue with burn life, the Dracos should do fine for both NRHO insertion and TEI, both of which require a lunar-flyby burn.  The real issue is propellant.  You probably need at least 950m/s of delta-v.  At Isp=300s, that's a mass ratio of 1.38, which means that you'd need another 2.7t of MMH and NTO.  That's roughly an extra 2.3m³ of volume.

I wonder if Dragon could use the transposition, docking, and extraction maneuver to hookup to a small module after TLI to give it the extra delta-v.

Edit: Of course this is a work around only needed if the propulsion module can't stay in the trunk.

[TheRadicalModerate]

Absolute minimum I could see in terms of work:

1a) Get the MVac to throttle down to 370kN, about 38% of max thrust.  Current min throttle is rated as 64%.
OR:
1b) Beef up the docking port and D2 structure to handle a 500kN load.  IDSS currently specifies 300kN.
OR:
1c) Some combination of the two above.

2) Work out how to do an F9/D2 launch and an FHE launch from the same pad within 1-2 days of each other.  Note that the D2 has to launch first, because the FHE S2 has very limited mission life.

3) Figure out how to do FHE launch, D2-S2 RPOD, and TLI, all within about 5 hours of one another, to stay within the S2 mission life.

4) Add extra 2.7t MMH/NTO tanks in the trunk, and interconnect it to the Draco systems, to get up to 950m/s of delta-v for NRHO insertion and TEI.

5) Double consumables, from one week to 15-17 days.  This stuff would also have to go in the trunk and is... one tonne?

6) Ensure that free-flying with no crew for up to two weeks works OK and doesn't further degrade consumables.

7) Extra comm and nav gear.

8) Work out some barely-acceptable solar storm sheltering procedure.

[catiare]

Absolute minimum I could see in terms of work:

1a) Get the MVac to throttle down to 370kN, about 38% of max thrust.  Current min throttle is rated as 64%.
OR:
1b) Beef up the docking port and D2 structure to handle a 500kN load.  IDSS currently specifies 300kN.
OR:
1c) Some combination of the two above.

2) Work out how to do an F9/D2 launch and an FHE launch from the same pad within 1-2 days of each other.  Note that the D2 has to launch first, because the FHE S2 has very limited mission life.

3) Figure out how to do FHE launch, D2-S2 RPOD, and TLI, all within about 5 hours of one another, to stay within the S2 mission life.

4) Add extra 2.7t MMH/NTO tanks in the trunk, and interconnect it to the Draco systems, to get up to 950m/s of delta-v for NRHO insertion and TEI.

5) Double consumables, from one week to 15-17 days.  This stuff would also have to go in the trunk and is... one tonne?

6) Ensure that free-flying with no crew for up to two weeks works OK and doesn't further degrade consumables.

7) Extra comm and nav gear.

Work out some barely-acceptable solar storm sheltering procedure.

Sounds like a kludgy solution but I think that if we had to do this as fast as we can this would be the plan with the least path of resistance to make it happen. Too bad FH and F9+D2 use the same pad. If they use different pads then items 2 and 3 would not be an issue.

[TheRadicalModerate]

Sounds like a kludgy solution but I think that if we had to do this as fast as we can this would be the plan with the least path of resistance to make it happen. Too bad FH and F9+D2 use the same pad. If they use different pads then items 2 and 3 would not be an issue.

#3 would still be an issue.  Once your "transfer stage" S2 is in orbit, the clock is running. 

Note also that #2 is only an issue to the extent that it's poor form to leave a crew sitting in orbit, consuming consumables and eating into mission duration time, for very long.  1-2 days isn't a problem.  3-4 days is pretty close to the edge.  I'd guess that anything over 5 days is unacceptable.

[Negan]

Sounds like a kludgy solution but I think that if we had to do this as fast as we can this would be the plan with the least path of resistance to make it happen. Too bad FH and F9+D2 use the same pad. If they use different pads then items 2 and 3 would not be an issue.

#3 would still be an issue.  Once your "transfer stage" S2 is in orbit, the clock is running. 

Note also that #2 is only an issue to the extent that it's poor form to leave a crew sitting in orbit, consuming consumables and eating into mission duration time, for very long.  1-2 days isn't a problem.  3-4 days is pretty close to the edge.  I'd guess that anything over 5 days is unacceptable.

Gemini 11 docked 1 hour 34 minutes after launch so there's that. I would add a bit more consumables to #5 to help with #2.

Edit: I would also add a Dragon XL to the FH since a lot of the 2nd stage deliverable TLI mass is being wasted.

[TheRadicalModerate]

Gemini 11 docked 1 hour 34 minutes after launch so there's that. I would add a bit more consumables to #5 to help with #2.

Yup, and the Russians can get to ISS on a 3-hour "fast track" rendezvous. I'm not sure whether the D2's long RPODs (19-27 hrs) are out of an abundance of caution or whether there's some technical limitation that takes that long.  Anybody know what the deal is here?

[watermod]

FIne...
For my fantasy time:   Launch a MoonShip with everything you dream you need on it. If it is just to the gateway it can be a stripped down MoonShip with no landing features so a tug.
Launch repeated fueling missions via StarShip tankers until you have what ever fuel you need plus...
When it's fully ready launch your Dragon the normal way from LC-39A and dock with the MoonShip.
Ride the MoonShip to the station or the moon.
Do whatever and then the Dragon still docked (or re-docked) can be taken by the MoonShip back to Earth orbit.
Un-dock and land like a normal Dragon.
If desired send fueling StarShips up to fuel the MoonShip for the next trip.
A stripped down MoonShip could be living area, big tanks, only vacuum engines, a dock and not much more. Quite similar to  the current SN-10 with vacuum engines, a docking port and some living space.  No fins. No aerodynamic stuff.  No landing stuff. Just a docking port.

[sdsds]

[...]
Do whatever and then the Dragon still docked (or re-docked) can be taken by the MoonShip back to Earth orbit.
Un-dock and land like a normal Dragon.
[...]

You present a great vision. Note the MoonShip needn't bring the Dragon all the way back to LEO; it merely needs to put it onto a reentry trajectory you believe its heatshield can tolerate.

Note also the capsule used in this architecture needn't be Dragon. It could work with Orion too.  It's almost as if Orion was designed to work in conjunction with a highly capable lander. Oh, wait....  ;-)

[Jorge]

Gemini 11 docked 1 hour 34 minutes after launch so there's that. I would add a bit more consumables to #5 to help with #2.

Yup, and the Russians can get to ISS on a 3-hour "fast track" rendezvous. I'm not sure whether the D2's long RPODs (19-27 hrs) are out of an abundance of caution or whether there's some technical limitation that takes that long.  Anybody know what the deal is here?

Both commercial crew providers are capable of fast rendezvous. The key to enabling fast rendezvous is on the ISS side, not the visiting vehicle side. The phase angle between ISS and the VV at orbital insertion must fall within a narrow range for fast rendezvous. In order to ensure that this phase window coincides with the planar window at the launch site, the timing and sizing of ISS reboosts must be carefully controlled. Russia currently controls the ISS reboost plan for the benefit of Soyuz. NASA's commercial crew and ISS visiting vehicle requirements are written to ensure that VV providers are not dependent on ISS maneuvers for successful rendezvous (i.e. VVs must have the consumables/free-flight duration to support "slow" rendezvous), so early flights are built around a 24-hour rendezvous timeline to demonstrate this. Since an ISS reboost profile that provides for fast rendezvous from KSC/CCAFS may not provide fast rendezvous from Baikonur, enabling fast rendezvous for future NASA commercial crew missions will require negotiation with the Russians.

[envy887]

Gemini 11 docked 1 hour 34 minutes after launch so there's that. I would add a bit more consumables to #5 to help with #2.

Yup, and the Russians can get to ISS on a 3-hour "fast track" rendezvous. I'm not sure whether the D2's long RPODs (19-27 hrs) are out of an abundance of caution or whether there's some technical limitation that takes that long.  Anybody know what the deal is here?

Both commercial crew providers are capable of fast rendezvous. The key to enabling fast rendezvous is on the ISS side, not the visiting vehicle side. The phase angle between ISS and the VV at orbital insertion must fall within a narrow range for fast rendezvous. In order to ensure that this phase window coincides with the planar window at the launch site, the timing and sizing of ISS reboosts must be carefully controlled. Russia currently controls the ISS reboost plan for the benefit of Soyuz. NASA's commercial crew and ISS visiting vehicle requirements are written to ensure that VV providers are not dependent on ISS maneuvers for successful rendezvous (i.e. VVs must have the consumables/free-flight duration to support "slow" rendezvous), so early flights are built around a 24-hour rendezvous timeline to demonstrate this. Since an ISS reboost profile that provides for fast rendezvous from KSC/CCAFS may not provide fast rendezvous from Baikonur, enabling fast rendezvous for future NASA commercial crew missions will require negotiation with the Russians.

Doesn't Baikonur's latitude also give the Russians more leeway in getting plane angles with the correct phasing, since it's much closer to the ISS inclination?

[TheRadicalModerate]

Both commercial crew providers are capable of fast rendezvous. The key to enabling fast rendezvous is on the ISS side, not the visiting vehicle side. The phase angle between ISS and the VV at orbital insertion must fall within a narrow range for fast rendezvous. In order to ensure that this phase window coincides with the planar window at the launch site, the timing and sizing of ISS reboosts must be carefully controlled. Russia currently controls the ISS reboost plan for the benefit of Soyuz. NASA's commercial crew and ISS visiting vehicle requirements are written to ensure that VV providers are not dependent on ISS maneuvers for successful rendezvous (i.e. VVs must have the consumables/free-flight duration to support "slow" rendezvous), so early flights are built around a 24-hour rendezvous timeline to demonstrate this. Since an ISS reboost profile that provides for fast rendezvous from KSC/CCAFS may not provide fast rendezvous from Baikonur, enabling fast rendezvous for future NASA commercial crew missions will require negotiation with the Russians.

So if SpaceX were to try to get a D2 to NRHO, the sequence would go something like this:

1) Launch F9/D2 with crew (T+0/00:00:00).

2) Launch FHE ASAP after crew launch.  Assuming 14 days of consumables on the D2, The FHE would have to be rolled out, checked out, and launched in 2 days, so let's say T+2/00:00:00.

3) FHE orbits #1 and #2 (up to T+2/3:00:00-ish):  D2 does RPOD with a docking attachment mounted on the FHE S2.

4) FHE orbit #3 (up to T+2/4:30:00): Pre-TLI checkout.

5) FHE orbit #4 (up to T+6:00:00): TLI from the S2, with the D2 accelerated eyeballs-out.

The FH test flight did a successful interplanetary injection 6 hours post-launch, so this could work.  It still requires substantially increasing the propellant and consumables on the D2, as well as figuring out a way to deal with the S2 acceleration forces on the D2 and its docking system.

Random thought:  I wonder if the S2 docking system could include some struts that took some load off of the docking ring by distributing it to somewhere else on the D2.  I have no clue where that would be, but at the very least it would be fairly easy to build struts that closed over the chines of the trunk fins.

Something like this, although an order of magnitude less kludgy?

(D2 renders adapted from Giuseppe De Chiara (Archipeppe68) - Own work, CC BY-SA 4.0, https://commons.wikimedia.org/w/index.php?curid=91111433.)

[sdsds]

it would be fairly easy to build struts that closed over the chines of the trunk fins.

Wow. Taken to an extreme you would have a docking ring for the trunk and wouldn't need to carry much of any load through the capsule. Makes one wonder what that load path would look like, with the capsule in tension being pulled by the trunk.

Then again there's another capsule that was designed from the beginning for eyeballs-out acceleration....

[TheRadicalModerate]

it would be fairly easy to build struts that closed over the chines of the trunk fins.

Wow. Taken to an extreme you would have a docking ring for the trunk and wouldn't need to carry much of any load through the capsule. Makes one wonder what that load path would look like, with the capsule in tension being pulled by the trunk.

Then again there's another capsule that was designed from the beginning for eyeballs-out acceleration....

MPCV was designed for eyeballs-out, but with very low acceleration.

Another thought:  I've been trying to find ways to use the docking port as-is.  It occurs to me that SpaceX might want to try out a Starship prototype on D2.  Since Starship has to dock tail-to-tail for refueling, it probably has to use multiple capture systems, all working together to damp out motion errors.  You could do the same thing on a Dragon trunk, with a matched passive system on the S2...

[sdsds]

another capsule that was designed from the beginning for eyeballs-out acceleration....

MPCV was designed for eyeballs-out, but with very low acceleration.

In the ESAS/CxP 1.5 launch architecture, wouldn't both TLI and LOI have been eyeballs out? In particular TLI would have been powered by the J-2X of the Ares V EDS, with Orion docked on the nose of the lander.

Moot now. But surely MVac could be throttled back so it provided "only" the acceleration that would have been provided by J-2X.

[TheRadicalModerate]

In the ESAS/CxP 1.5 launch architecture, wouldn't both TLI and LOI have been eyeballs out? In particular TLI would have been powered by the J-2X of the Ares V EDS, with Orion docked on the nose of the lander.

Moot now. But surely MVac could be throttled back so it provided "only" the acceleration that would have been provided by J-2X.

Ah, I found the problem.  The lunar architecture for Constellation assumed that the Orion CEV (23t) would launch separately and rendezvous with both the EDS (26.4t at burnout) and the Altair lander (45.9t), then EDS would take the whole stack to TLI.

The J-2X has a thrust of about 1306kN but had one throttle point at 84%, so burnout thrust would be 1097kN.

So max eyeballs-out acceleration on the CEV (note that I erroneously said MPCV above) would have been 11.5m/s², which would have generated 264kN of force on the docking ring.  That's within spec for the IDSS/NDS.

[sdsds]

I was reminded of this discussion when reviewing the planned launch of the Intuitive Machines lunar lander. SpaceX has apparently agreed to launch the lander with enough oxygen and methane to power it from GTO through TLI and all the way to the lunar surface. Surely a cluster of the type of engine on that lander, combined with appropriate propellant tankage, could power a departure stage that could be the target of a quick rendezvous with a separately launched crew Dragon. SpaceX might be particularly comfortable with that approach ... once they had a successful launch of the lunar lander under their belt.