NASA's Artemis Program Updates and Discussion Thread 6

[TheRadicalModerate]

We have discussed ad nauseam how Starship (Depot/Tanker/HLS) plus D2 can completely supersede SLS/Orion. However, If Starship fails and the Appendix P hardware succeeds, is there a mission plan that does not need SLS/Orion? How does crew reach LEO, and how does crew get from LEO to the Moon and back?

I can see a couple options:

-the most direct option would be the launch a crew Dragon to dock with the Blue Moon HLS transporter/lander in LEO. The crew transfers over to HLS, then Dragon undocks and remains in LEO under remote control. HLS proceeds with its mission, performing a LTI burn, a lunar landing, lunar takeoff, and TEI burn. Blue Moon redocks with Dragon in LEO, the crew transfers back, and finally returns to Earth.

The primary unknown factor is Blue Moon's ability to reach LEO from the lunar surface. As spec'd, it's only required to reach Gateway in NRLO. Some investigation would also be required about its ability to sustain crew on the LEO to Lunar surface trip and back, although the lunar surface and translunar space are similar enough that I don't think this will be a major issue.

BM2 on its own can't even come close to executing this mission.  If it did, it'd be the size of Starship.  However, there are lots of other options:

1) Keep the split between the vehicle that brings the crew to and from lunar orbit, and the one that goes down to the surface.  You could then use a "ferry" crewed BM2, carried by the CT, to go LEO-NRHO (or LEO-LLO), where it would transfer the crew to the HLS BM2.  Upon HLS's return from the surface, the ferry would carry the crew back to LEO, where it would transfer to the D2 or Starliner.

2) Make a "no propulsion" version of the BM2 crew module and permanently dock it onto a CT.  The CT would then haul the BM2 from LEO to lunar orbit and back.  The D2 would bring up both the crew and the consumables to reprovision it.

3) If all you really want to get rid of is SLS, then launch Orion on a new Glenn, dock it with the CT, and have the CT take it to lunar orbit.  The Orion could then return directly to EDL when the mission was over.

Surface stays are limited by Dragon's free flying design limit of 10 days.

That free-flight limit is for when the D2 is crewed the entire time.  We suspect (but don't know) that the limit can be made considerably longer if the D2 is uncrewed.

-Alternatively the first crew Dragon could return after docking to HLS, and a second Dragon could be launched upon their return to LEO. This would increase risk to the crew, as a launch failure on the 2nd Dragon would be unsurvivable, but would allow full usage of HLS's 30 day design limit with weeklong stays on the Moon.

As I said above, I don't think you need to do this for extending the uncrewed free-flight time, but there is The RAAN Problem to worry about:  If the lunar segment aborts early, the returning vehicle and the D2 will have very different orbital nodes, which can require unreasonably large amounts of delta-v to make the orbital planes line up.  This may be a reason for a second D2 to be on warm standby.

The other alternative we've discussed is to take the D2 along to lunar orbit and back.  It's not used during the translunar and lunar segments; it just sits there, idle but docked.  If it's docked to a BM2 or a BM2 crew module, the BM2 system supports the crew, and the D2 is just there so the RPOD back in LEO is a non-issue.

This does require some modification to the D2:  radiation-hardening the avionics, and ensuring that the docking system can be pushed by the CT/BM2 stack.  However, unlike Starship, the CT will be very low thrust, so pushing the D2 around eyeballs-out can be considerably easier.

-Assuming a commercial crew station with multiple docking ports is available, like Haven-2, Dragon and Blue Moon could meet at the station for the crew transfer. The Dragon could then be left docked until HLS returns.

This is still vulnerable to the RAAN problem.  Again, I doubt the D2 free-flight limit is only 10 days if the D2 is uncrewed.

[jadebenn]

... than pay ...Bechtel to build ... ML2.

This is already too late; NASA has already paid most of the money for the construction of ML-2.

https://www.usaspending.gov/award/CONT_AWD_80KSC019C0013_8000_-NONE-_-NONE-

You can also see from recent photography that ML-2 is structurally complete. It has been for a few months now. They've moved on to outfitting.

EDIT: A recent view:
https://twitter.com/i/status/2012506591644229665

It's also looking likely the OIG report significantly overestimated the final price tag (based on the spending slope thus far), but I'll not count those chickens before they hatch...

[TheRadicalModerate]

... than pay ...Bechtel to build ... ML2.

This is already too late; NASA has already paid most of the money for the construction of ML-2.

https://www.usaspending.gov/award/CONT_AWD_80KSC019C0013_8000_-NONE-_-NONE-

Hey, it's cost-plus.  There's plenty of time to expand the award in the out-years.

You can also see from recent photography that ML-2 is structurally complete. It has been for a few months now. They've moved on to outfitting.

So it's more than halfway done!  (Partially kidding, but not completely.)

[RocketFan1959]

Tickets for the launch are now available on the KSC site. The ticketing app has been flaky this morning, but with enough patience (and page refreshes), you can buy a ticket to see the launch at either the Saturn V Center ($250) or the Visitor's Center ($99).

https://www.kennedyspacecenter.com/event/nasa-space-launch-system-sls-artemis-ii/

[jadebenn]

So it's more than halfway done!  (Partially kidding, but not completely.)

Actually completely fair! There's still a lot of work to do.

What I was trying to get at is that most of the ML-2 equipment is already manufactured. If you killed the contract tomorrow, I wouldn't be surprised if the closeout fees ate most or all of the remaining value. Might genuinely be cheaper to finish it.

Now, that's admittedly a different conversation than "should we have Block 1B at all," and EUS work seems a lot more immature so I can see value in that perspective (even if I strongly disagree). It just seems like ML-2 itself isn't really an avoidable cost anymore.

Could be wrong, though. Feel free to point and laugh at me if Bechtel does something stupid again. 

[catdlr]

Tickets for the launch are now available on the KSC site. The ticketing app has been flaky this morning, but with enough patience (and page refreshes), you can buy a ticket to see the launch at either the Saturn V Center ($250) or the Visitor's Center ($99).

https://www.kennedyspacecenter.com/event/nasa-space-launch-system-sls-artemis-ii/

And it's already sold out.

[catdlr]

https://twitter.com/NASA_Johnson/status/2014754388762366306

NASA's Johnson Space Center

@NASA_Johnson
·
Where mission success meets the ocean. 🌕

On this week’s #HWHAP, hear how @NASA and its partners are preparing for @NASAArtemis
 II recovery operations in the Pacific Ocean.

Pod Cast:  https://go.nasa.gov/4ag4wDV

[321]

We have discussed ad nauseam how Starship (Depot/Tanker/HLS) plus D2 can completely supersede SLS/Orion. However, If Starship fails and the Appendix P hardware succeeds, is there a mission plan that does not need SLS/Orion? How does crew reach LEO, and how does crew get from LEO to the Moon and back?

I can see a couple options:

-the most direct option would be the launch a crew Dragon to dock with the Blue Moon HLS transporter/lander in LEO. The crew transfers over to HLS, then Dragon undocks and remains in LEO under remote control. HLS proceeds with its mission, performing a LTI burn, a lunar landing, lunar takeoff, and TEI burn. Blue Moon redocks with Dragon in LEO, the crew transfers back, and finally returns to Earth.

The primary unknown factor is Blue Moon's ability to reach LEO from the lunar surface. As spec'd, it's only required to reach Gateway in NRLO. Some investigation would also be required about its ability to sustain crew on the LEO to Lunar surface trip and back, although the lunar surface and translunar space are similar enough that I don't think this will be a major issue.

BM2 on its own can't even come close to executing this mission.  If it did, it'd be the size of Starship.  However, there are lots of other options:

1) Keep the split between the vehicle that brings the crew to and from lunar orbit, and the one that goes down to the surface.  You could then use a "ferry" crewed BM2, carried by the CT, to go LEO-NRHO (or LEO-LLO), where it would transfer the crew to the HLS BM2.  Upon HLS's return from the surface, the ferry would carry the crew back to LEO, where it would transfer to the D2 or Starliner.

2) Make a "no propulsion" version of the BM2 crew module and permanently dock it onto a CT.  The CT would then haul the BM2 from LEO to lunar orbit and back.  The D2 would bring up both the crew and the consumables to reprovision it.

3) If all you really want to get rid of is SLS, then launch Orion on a new Glenn, dock it with the CT, and have the CT take it to lunar orbit.  The Orion could then return directly to EDL when the mission was over.

Surface stays are limited by Dragon's free flying design limit of 10 days.

That free-flight limit is for when the D2 is crewed the entire time.  We suspect (but don't know) that the limit can be made considerably longer if the D2 is uncrewed.

-Alternatively the first crew Dragon could return after docking to HLS, and a second Dragon could be launched upon their return to LEO. This would increase risk to the crew, as a launch failure on the 2nd Dragon would be unsurvivable, but would allow full usage of HLS's 30 day design limit with weeklong stays on the Moon.

As I said above, I don't think you need to do this for extending the uncrewed free-flight time, but there is The RAAN Problem to worry about:  If the lunar segment aborts early, the returning vehicle and the D2 will have very different orbital nodes, which can require unreasonably large amounts of delta-v to make the orbital planes line up.  This may be a reason for a second D2 to be on warm standby.

The other alternative we've discussed is to take the D2 along to lunar orbit and back.  It's not used during the translunar and lunar segments; it just sits there, idle but docked.  If it's docked to a BM2 or a BM2 crew module, the BM2 system supports the crew, and the D2 is just there so the RPOD back in LEO is a non-issue.

This does require some modification to the D2:  radiation-hardening the avionics, and ensuring that the docking system can be pushed by the CT/BM2 stack.  However, unlike Starship, the CT will be very low thrust, so pushing the D2 around eyeballs-out can be considerably easier.

-Assuming a commercial crew station with multiple docking ports is available, like Haven-2, Dragon and Blue Moon could meet at the station for the crew transfer. The Dragon could then be left docked until HLS returns.

This is still vulnerable to the RAAN problem.  Again, I doubt the D2 free-flight limit is only 10 days if the D2 is uncrewed.

If sized right for propulsion trip LEO>LLO(LL)>LEO a reusable BM2/CT stack with single refueling port for both is all that is needed to develop to make moon missions a routine. Plus of cause LEO refueling from NG S2.
D2 would carry crew to and from LEO and could be equipped with propulsion pack to be able for some change of orbital plane plane to meet early or late return of BM2/CT stack.

[TheRadicalModerate]

D2 would carry crew to and from LEO and could be equipped with propulsion pack to be able for some change of orbital plane plane to meet early or late return of BM2/CT stack.

If you have the D2 in a 350km circular orbit with a 23º inclination, a 19.13º difference in RAAN costs 1000m/s of delta-v to re-align.¹  19.13º is how far the Moon moves on average in 1.45 days.

So you're talking about a pretty big propulsion pack.


__________
¹planeAngle = arccos(cos(inc1)*cos(inc2)+sin(inc1)*sin(inc2)*cos(RAAN1-RAAN2))

Δv = 2*v*sin(planeAngle/2)  (this version only works for circular orbits of the same radius)



BTW:

If sized right for propulsion trip LEO>LLO(LL)>LEO a reusable BM2/CT stack with single refueling port for both is all that is needed to develop to make moon missions a routine.

I'd be interested to see a design for this.  The flamey end of the CT and the flamey end of the BM have to point in opposite directions, so the logical place for them to dock is on the tops of their LH2 tanks.  I don't know where a GS2 would dock to refuel them both at the same time if they were docked like this.

My guess is that they have to be undocked and refuel separately.

[321]

D2 would carry crew to and from LEO and could be equipped with propulsion pack to be able for some change of orbital plane plane to meet early or late return of BM2/CT stack.

If you have the D2 in a 350km circular orbit with a 23º inclination, a 19.13º difference in RAAN costs 1000m/s of delta-v to re-align.¹  19.13º is how far the Moon moves on average in 1.45 days.

So you're talking about a pretty big propulsion pack.

I did not mean it must be able to catch it in any case, only in most. A kind of reasonable compromise.

[TheRadicalModerate]

D2 would carry crew to and from LEO and could be equipped with propulsion pack to be able for some change of orbital plane plane to meet early or late return of BM2/CT stack.

If you have the D2 in a 350km circular orbit with a 23º inclination, a 19.13º difference in RAAN costs 1000m/s of delta-v to re-align.¹  19.13º is how far the Moon moves on average in 1.45 days.

So you're talking about a pretty big propulsion pack.

I did not mean it must be able to catch it in any case, only in most. A kind of reasonable compromise.

Aborts from lunar orbit aren't instant no matter what, so it's perfectly reasonable to pick an elapsed time after the event that triggers the abort, and design around that.  If that number is a couple of weeks, no problem.  But if it's a day or two + trans-earth transit time, then you have to figure out whether the D2 (or the HLS) can provide that amount of delta-v.  If not, then there's no reasonable compromise.

I don't know what the limit is on extra mass on a crewed D2 mission.  2.0t?  That's considerably heavier than it flies with crew to the ISS.  So if the usual wet mass for a D2 is about 12.5t and the Dracos can generate 300s of Isp, then you can get about 440m/s of extra delta-v.  That's about 15 hours off the nominal return time.  That leaves a whole bunch of unreasonableness left in the abort capabilities.

We've had this argument at great length over on the thread for discussing this kind of mission based on SpaceX hardware, without getting broad agreement on what's acceptable.  (I don't think we understand the orbital mechanics completely.)  I suspect forcing a handful of days' delay is probably acceptable.  But we're nowhere near making that the only black zone in the abort.

[DanClemmensen]

D2 would carry crew to and from LEO and could be equipped with propulsion pack to be able for some change of orbital plane plane to meet early or late return of BM2/CT stack.

If you have the D2 in a 350km circular orbit with a 23º inclination, a 19.13º difference in RAAN costs 1000m/s of delta-v to re-align.¹  19.13º is how far the Moon moves on average in 1.45 days.

So you're talking about a pretty big propulsion pack.

I did not mean it must be able to catch it in any case, only in most. A kind of reasonable compromise.

Aborts from lunar orbit aren't instant no matter what, so it's perfectly reasonable to pick an elapsed time after the event that triggers the abort, and design around that.  If that number is a couple of weeks, no problem.  But if it's a day or two + trans-earth transit time, then you have to figure out whether the D2 (or the HLS) can provide that amount of delta-v.  If not, then there's no reasonable compromise.

I don't know what the limit is on extra mass on a crewed D2 mission.  2.0t?  That's considerably heavier than it flies with crew to the ISS.  So if the usual wet mass for a D2 is about 12.5t and the Dracos can generate 300s of Isp, then you can get about 440m/s of extra delta-v.  That's about 15 hours off the nominal return time.  That leaves a whole bunch of unreasonableness left in the abort capabilities.

We've had this argument at great length over on the thread for discussing this kind of mission based on SpaceX hardware, without getting broad agreement on what's acceptable.  (I don't think we understand the orbital mechanics completely.)  I suspect forcing a handful of days' delay is probably acceptable.  But we're nowhere near making that the only black zone in the abort.

Even the lowest-cost HLS-based mission is still very high cost compared to anything except SLS/Orion. Putting a second D2 on 48-hour standby is not expensive compared to the rest of the mission. It gets a bit more expensive if you actually need to launch it, but that only occurs in the abort case.

[Greg Hullender]

. . . but there is The RAAN Problem to worry about:  If the lunar segment aborts early, the returning vehicle and the D2 will have very different orbital nodes, which can require unreasonably large amounts of delta-v to make the orbital planes line up.  This may be a reason for a second D2 to be on warm standby.

Is there any evidence that anyone actually cares about this? It takes a maximum of 9 days for a launch window to open up, but the flight time is already so long that this makes almost no difference, medically, given how long it takes to return from the moon in the first place. If you could get someone to an Earth hospital within 24 hours, that would make a difference, but when you're already talking several days, adding as much as a month wouldn't make much difference. All this effort to avoid the RAAN problem is effort wasted.

This is pretty easy information to find online. Why does this topic keeps coming up?

[TheRadicalModerate]

Even the lowest-cost HLS-based mission is still very high cost compared to anything except SLS/Orion. Putting a second D2 on 48-hour standby is not expensive compared to the rest of the mission. It gets a bit more expensive if you actually need to launch it, but that only occurs in the abort case.

I agree, with a few provisos:

1) There aren't very many D2's, and at least one of them is inaccessible while docked at ISS or a successor CLD.  In the CLD era, more than one of them may be inaccessible.  Balancing lunar logistics with LEO logistics is a non-trivial task.

2) A launch failure of the backup D2, or even a multi-day pad glitch, reduces the probability of a good outcome after the abort.

3) "A bit more expensive" is still something to avoid if you can.

Is there any evidence that anyone actually cares about this? It takes a maximum of 9 days for a launch window to open up, but the flight time is already so long that this makes almost no difference, medically, given how long it takes to return from the moon in the first place. If you could get someone to an Earth hospital within 24 hours, that would make a difference, but when you're already talking several days, adding as much as a month wouldn't make much difference. All this effort to avoid the RAAN problem is effort wasted.

This is pretty easy information to find online. Why does this topic keeps coming up?

It's a major difference between a mission with a vehicle capable of direct EDL (Orion or D2-on-the-HLS-nose) and one that requires an LEO-based RPOD after a propulsive insert back into LEO.  Minimizing the time from abort to getting the crew safely on the ground is going to be a major parameter in overall pLOC.  That's not wasted effort.

If the difference between the two architectures is a couple of days, that's one thing.  I'm not convinced of that.¹  If it's a couple of weeks, that's a completely different situation.

Medical emergencies are only one class of contingencies that need to be planned for.  Another higher-runner will be loss-of-pressure accidents, where the crew has to breathe and hydrate on suit-based consumables.  Orion crew survival suits last for 6 days.

That said, Dan is correct that if there's a problem, the simplest solution (not the least expensive, but simplest) is a second standby D2.  The other solution is D2-on-the-nose.  Whether that's less expensive depends on the DDT&E needed to make a D2 secure on the nose of the transporting vehicle (Starship or possibly a Blue CT or BM2) and its avionics radiation-hardened.

__________
¹Three potential problems:

1) Time and delta-v to get out of lunar orbit.  From NRHO apoapsis, this should be close to a fixed value, but it depends on leaving from apoapsis, which only comes around every 6.5 days.  From a polar LLO, time isn't much of an issue, but you may have a significant plane-change to get into a decent trans-earth orbit.

2) Relative RAANs of the trans-earth orbit and the LEO parking orbit.  I think (but am not yet sure) that this is a two-week cycle, not four.  We need to know how many of the days and that cycle are black zones, where the delta-v to do the rendezvous is unacceptably high.

3) Inclination, which we haven't talked about much.  Until recently, I'd been assuming that the inclination of the trans-earth orbit was easy to adjust at lunar departure.  Now I'm not sure.

[DanClemmensen]

Even the lowest-cost HLS-based mission is still very high cost compared to anything except SLS/Orion. Putting a second D2 on 48-hour standby is not expensive compared to the rest of the mission. It gets a bit more expensive if you actually need to launch it, but that only occurs in the abort case.

I agree, with a few provisos:

1) There aren't very many D2's, and at least one of them is inaccessible while docked at ISS or a successor CLD.  In the CLD era, more than one of them may be inaccessible.  Balancing lunar logistics with LEO logistics is a non-trivial task.

2) A launch failure of the backup D2, or even a multi-day pad glitch, reduces the probability of a good outcome after the abort.

3) "A bit more expensive" is still something to avoid if you can.

Agreed on all points. I see this as a short-term solution for the first one or two missions, and I feel that for those, simpler is better and zero DDT&E is better.

By the time Artemis V, I expect a crewed EDL Starship will be available, and missions can use this Ship to take crew from Earth surface to LLO and from LLO to Earth's surface. The only reason for D2 shortage would be if the Artemis cadence increases to more than one a year, and I see this as a good problem to have.

[TheRadicalModerate]

Agreed on all points. I see this as a short-term solution for the first one or two missions, and I feel that for those, simpler is better and zero DDT&E is better.

By the time Artemis V, I expect a crewed EDL Starship will be available, and missions can use this Ship to take crew from Earth surface to LLO and from LLO to Earth's surface. The only reason for D2 shortage would be if the Artemis cadence increases to more than one a year, and I see this as a good problem to have.

I think there's about a 30% chance that Starship, as currently architected, will never be crew-certifiable for launch and EDL, at least not to NASA standards.  I also think the amount of time for the empirical evidence and the probabilistic risk assessment model to converge to a credible pLOC¹ will be longer than you think.

That's not really a reason to abandon the warm standby as your primary abort methodology, but it is a reason to worry about the long-term logistics of such a methodology.

___________
¹Note that "credible pLOC" doesn't appear to be something that NASA is worried about with SLS/Orion.  But I suspect that there will be a double standard when it comes to things that might displace SLS/Orion.

[DanClemmensen]

Agreed on all points. I see this as a short-term solution for the first one or two missions, and I feel that for those, simpler is better and zero DDT&E is better.

By the time Artemis V, I expect a crewed EDL Starship will be available, and missions can use this Ship to take crew from Earth surface to LLO and from LLO to Earth's surface. The only reason for D2 shortage would be if the Artemis cadence increases to more than one a year, and I see this as a good problem to have.

I think there's about a 30% chance that Starship, as currently architected, will never be crew-certifiable for launch and EDL, at least not to NASA standards.  I also think the amount of time for the empirical evidence and the probabilistic risk assessment model to converge to a credible pLOC¹ will be longer than you think.

That's not really a reason to abandon the warm standby as your primary abort methodology, but it is a reason to worry about the long-term logistics of such a methodology.

___________
¹Note that "credible pLOC" doesn't appear to be something that NASA is worried about with SLS/Orion.  But I suspect that there will be a double standard when it comes to things that might displace SLS/Orion.

SpaceX intends to fly Hundreds of Starship EDL missions, mostly Pez and Tankers, before Artemis V. If NASA rejects crewed EDL Starship in favor of SLS/Orion after that, we have no hope of a viable long-term Artemis, or any other rational decisions from NASA.

[Vultur]

I think there's about a 30% chance that Starship, as currently architected, will never be crew-certifiable for launch and EDL, at least not to NASA standards

I can see a chance that politics alone prevents it. I don't think I'd go as high as 30% though.

I also think the amount of time for the empirical evidence and the probabilistic risk assessment model to converge to a credible pLOC¹ will be longer than you think.

I think that depends heavily on how fast Starship cadence scales. I absolutely do not believe the more extreme claims (once/hour within 4 years?) but I do think it's very plausible that Starship will scale faster than anything before (because SpaceX is the most experienced organization in the world at this, and because Starship & its manufacturing and GSE are being designed and built for mass scale in a way even F9 wasn't early on).

I think that assuming they don't hit a fundamental showstopper (which I now think is pretty unlikely) things will be painfully slow until they're "suddenly" fast. But once they're fast evidence base will be built very quickly.

If Artemis V is likely to be something like 2031 (?) I think there's a good chance that Starship has a pretty extensive record by then.

[TheRadicalModerate]

SpaceX intends to fly Hundreds of Starship EDL missions, mostly Pez and Tankers, before Artemis V.

We've gone over this elsewhere, but here's my prediction:

1) I don't think we're going to see more than 50 consecutive successful flights for quite a while.

2) Meanwhile, early operational failures will make NASA less confident that the PRA failure tree has accounted for all the failures of imagination. 

Ironically, finding some failures of imagination will probably give NASA more confidence in the PRA in long run.  I think if we see 2-3 operational failures separated by 30-40 successes, followed by 50 successes, then NASA will start to believe in the model.

It's not practical to get to a sufficient confidence interval, with a sufficient confidence level, on empirical data alone.  It's hundreds of consecutive successful missions.

If NASA rejects crewed EDL Starship in favor of SLS/Orion after that, we have no hope of a viable long-term Artemis, or any other rational decisions from NASA.

If they reject crewed EDL Starship in favor of D2+HLS and whatever Blue puts together, long-term Artemis will be viable just fine.  As you said, even with the expense of tying up a second D2, the system is still about 25% the cost of SLS/Orion.

[DanClemmensen]

If NASA rejects crewed EDL Starship in favor of SLS/Orion after that, we have no hope of a viable long-term Artemis, or any other rational decisions from NASA.

If they reject crewed EDL Starship in favor of D2+HLS and whatever Blue puts together, long-term Artemis will be viable just fine.  As you said, even with the expense of tying up a second D2, the system is still about 25% the cost of SLS/Orion.

That is probably a rational decision, a lot safer than SLS/Orion+HLS. The irrational decision is a different case, where they declare EDL Starship to be unsafe because it has only launched 40 times while SLS/Orion is safe because it has launched four times.