HLS Option B and the Sustaining Lunar Development Phase (Appendix P)

[yg1968]

I was trying to figure the cost per seat for HLS-Starship. If we assume that each HLS-Starship can hold 4 people and that 32% of Option B is for development and 68% for the mission (*), I get a price of $195M per seat for a NASA (or international partner) astronaut (68% x $1150M / 4 = $195M).

If Crewed Starship is used to get from Earth to Earth orbit, presumably HLS-Starship could then carry 12 astronauts, so you would then get a much lower $65M per seat (68% x $1150M / 12 = $65M$) for HLS-Starship. Presumably, crewed Starship would cost you another $65M per seat, so private astronauts could get to the lunar surface for about $130M per seat.

As a comparison, Russia was offering a trip around the Moon for $150M per seat a few years ago; I had guessed that the cost for a private mission to the lunar surface would be around the same price. I likely wasn't too far off.

(*) The 68% comes from the CCtCap contract where 68% was for the missions and 32% for the development (8 missions x 4 astronauts x $55M per seat / $2600M = 68%). The Boeing numbers are similar (8 missions x 4 astronauts x $90M per seat / $4200M = 68%).

[DanClemmensen]

I was trying to figure the cost per seat for HLS-Starship. If we assume that each HLS-Starship can hold 4 people and that 32% of Option B is for development and 68% for the mission (*), I get a price of $195M per seat for a NASA or international partner astronaut (68% x $1150M / 4 = $195M).

If Crewed Starship is used to get from Earth to Earth orbit presumably HLS-Starship could then carry 12 astronauts, so you would then get a much lower $65M per seat (68% x $1150M / 12 = $65M$) for HLS-Starship. Presumably, crewed Starship would cost you another $65M per seat, so a private astronaut could get to the lunar surface for about $130M per seat.

As a comparison, Russia was offering a trip around the Moon for $150M per seat a few years ago; I had guessed that the cost for a private mission to the lunar price would be around the same price. I wasn't too far off.

(*) The 68% comes from the CCtCap contract where 68% was for the missions and 32% for the development (8 missions x 4 astronauts x $55M / $2600M = 68%). The Boeing numbers are similar (8 missions x 4 astronauts x $90M / $4200M =68%).

I think your result may be valid, but there is a lot to discuss about the details of your analogy with CCtCap. The Option B development (and production) are minor extensions of the larger HLS, so maybe run the calculation on the entire Option A + Option B. But you are looking at the incremental cost of a new mission, so this is closer to the cost of the Crew Dragon extensions and not of the Crew Dragon program.  I'm also not clear on the cost differences between a few very cheap tanker flights (HLS) versus a single more-expensive F9 flight (CCP). Finally, the HLS SS is  expended on the HLS mission, while Falcon US and Crew Dragon trunk are expended on the CCP mission.

[yg1968]

If we take into account Options A & B, the numbers would be a little bit higher but not much more. The total contract value for Options A & B is $4150M for 3 missions ($3000M for Option A plus 1150M for Option B).

If we assume that each HLS-Starship can hold 4 people and that 32% of Option B is for development and 68% for the mission (*), I now get a price of $235M per seat for a NASA (or international partner) astronaut (4150/3 missions x 68%/4 seats= $235M).

If Crewed Starship is used to get from Earth to Earth orbit presumably HLS-Starship could then carry 12 astronauts, so you would then get a much lower $78M per seat (4150/3 missions x 68%/12 seats= $78M) for HLS-Starship. Presumably, crewed Starship would cost you another $78M per seat, so private astronauts could get to the lunar surface for about $156M per seat.

As a comparison, Russia was offering a trip around the Moon for $150M per seat a few years ago; I had guessed that the cost for a private mission to the lunar price would be around the same price. I likely wasn't too far off.

(*) The 68% comes from the CCtCap contract where 68% was for the missions and 32% for the development (8 missions x 4 astronauts x $55M / $2600M = 68%). The Boeing numbers are similar (8 missions x 4 astronauts x $90M / $4200M = 68%).

[JayWee]

_IF_ Starship puts down 100t of cargo onto the Moon surface - that'd be $11.5k/kg price to Moon surface. That's... not bad!
Any idea how approximately that was for Apollo LM (I don't want to do the simple total Apollo cost/kg)?

[Redclaws]

_IF_ Starship puts down 100t of cargo onto the Moon surface - that'd be $11.5k/kg price to Moon surface. That's... not bad!
Any idea how approximately that was for Apollo LM (I don't want to do the simple total Apollo cost/kg)?

Well, then what number do you want to do?  What basis makes sense to you?  It’s not a question with an obvious single right answer, not at all.  How do you credit all of the many components/accomplishments of the program and split it out to kg/$?

[JayWee]

_IF_ Starship puts down 100t of cargo onto the Moon surface - that'd be $11.5k/kg price to Moon surface. That's... not bad!
Any idea how approximately that was for Apollo LM (I don't want to do the simple total Apollo cost/kg)?

Well, then what number do you want to do?  What basis makes sense to you?  It’s not a question with an obvious single right answer, not at all.  How do you credit all of the many components/accomplishments of the program and split it out to kg/$?

Something closest to incremental price of one Apollo mission/payload.

[TheRadicalModerate]

If we take into account Options A & B, the numbers would be a little bit higher but not much more. The total contract value for Options A & B is $4150M for 3 missions ($3000M for Option A plus 1150M for Option B).

If we assume that each HLS-Starship can hold 4 people and that 32% of Option B is for development and 68% for the mission (*), I now get a price of $235M per seat for a NASA (or international partner) astronaut (4150/3 missions x 68%/4 seats= $235M).

If Crewed Starship is used to get from Earth to Earth orbit presumably HLS-Starship could then carry 12 astronauts, so you would then get a much lower $78M per seat (4150/3 missions x 68%/12 seats= $78M) for HLS-Starship. Presumably, crewed Starship would cost you another $78M per seat, so private astronauts could get to the lunar surface for about $156M per seat.

As a comparison, Russia was offering a trip around the Moon for $150M per seat a few years ago; I had guessed that the cost for a private mission to the lunar price would be around the same price. I likely wasn't too far off.

(*) The 68% comes from the CCtCap contract where 68% was for the missions and 32% for the development (8 missions x 4 astronauts x $55M / $2600M = 68%). The Boeing numbers are similar (8 missions x 4 astronauts x $90M / $4200M =68%).

I don't think you can figure this without a cost model for tankers.  SpaceX is likely to be charging more for DDT&E than they did for CCtCap, because they're actually developing three different spacecraft, partially on NASA's dime:  the LSS itself, the depot (presumably with the bulk of the prop transfer hardware on it), and the lift tanker (whose cost is dramatically impacted by how quickly it becomes reusable and how many re-uses they can get out of it).

I would expect that SpaceX has a fairly complicated model for the cost of a tanker, which takes into account their best guess for the probability of extensive reuse after a given number of launches.  I'm using a SWAG below.

We have three different tanker requirements:

1) Uncrewed Option A #1, which only has to land and maybe take off just enough to demonstrate ascent stability.  I use 110t for inert mass (95t LSS dry + 15t crew module), which would require only 3 tankers' worth of prop.  But this is where SpaceX would also launch the depot, which presumably can be reused for at least the first four missions.

2) Crewed Option A #2, which needs to do the complete LEO-NRHO-LS-NRHO flight on one tank of prop.  I get 9 tankers for this.

3) Crewed Option B #1, which should be the same as Option A #2: 9 tankers.

So for Option A ($3000M contract), we have:
- Two LSSes at... a bit more than an F9/D2 mission?  Say $350M apiece?  $700M
- One depot at maybe $80M
- 12 lift tankers, at $70M (assuming a pretty high probability that reusability is slight to nonexistent) = $840M
Total:  $1620M, which would be 54% of the $3000M contract, leaving DDT&E as 46%.

For Option B ($1150M contract):
- One upgraded LSS.  Let's say $300M, as they get better at building them.
- 9 lift tankers, with a higher likelihood of resuability:  $50M each = $450M
Total: $750M,  65% of the contract, with DDT&E being 35% (roughly in line with your guess).

However, if you're really looking at per-seat costs for the long run, you're looking at the SLT contract.  Let's assume that we build a new LSS for SLT #1, incorporating lessons learned from Option B, that it costs roughly the same to build, and that it's 4x reusable.  For each of those four missions, we have an initial mission of the LSS (9 tankers), followed by 3 missions where the LSS is refueled in NRHO (this costs 11 tankers, by my reckoning).

So, SLT #1-#4:
- LSS: $300M, used for all four missions
- 42 tankers (9 + 3*11).  Reusability should be even higher probability:  $35M?  $1470M for all.
Total for four missions:  $1770M.  Assuming 16 crew total:  $111M/seat.

That's just the HLS leg, of course.  If you factor in 4 SLS/Orions at the OIG's marginal cost, you're at $1136M/seat.

Now, if sanity were to break out and SLT really turns out to be staged from LEO with an F9/D2 handling the crew launch/EDL, that takes about 15 tankers, with a post-ascent refueling in NRHO.  For four missions:
- One LSS: $300M
- 60 tankers @ $35M: $2100M
- 4 F9/D2 launches @ $250M: $1000M
Total for 4 Artemis missions:  $3400M.  That's $213M/seat.

That's an all-SpaceX mission.  If you want to service the other SLT vendors in NRHO with an LSS that goes LEO-NRHO-LEO, that costs 5 tankers, more or less.

Let's call it Commercial Cislunar Transit, for 8 missions:
- One LSS (which should be easily able to be 8x reusable, since it doesn't land on the Moon)
- 40 tankers (8 missions @ 5/mission) @ $35M: $1400M
- 8 F9/D2 launches @ $250M: $2000M
Total for 8 missions: $3400M, which is $106M/seat, plus whatever the App P SLT cost is.

[yg1968]

I was trying to calculate how much a private lunar surface mission would cost. So I assumed 12 people on crewed Starship and on HLS-Starship. I doubt that there will be any mission using F9/Dragon and HLS-Starship. I don't think that either NASA or SpaceX are really interested in those for different reasons: NASA because they have to use SLS because of Congress and SpaceX because they believe that Starship is the future, not Dragon. 

[TheRadicalModerate]

I was trying to calculate how much a private lunar surface mission would cost. So I assumed 12 people on crewed Starship and on HLS-Starship. I doubt that there will be any mission using F9/Dragon and HLS-Starship. I don't think that either NASA or SpaceX are really interested in those for different reasons: NASA because they have to use SLS because of Congress and SpaceX because they believe that Starship is the future, not Dragon.

The problem with estimating this is that it's far enough in the future that you really can't predict what the retail price of a tanker will be--which of course won't keep me from trying to guess.

If you refuel both in VLEO and in an LEO+2000 HEEO, that's 13-14 lift tankers.  When Starship is crew-certified for launch and EDL, I'd guess that tankers will go for no more than $15M apiece.  (If you refuel in LLO instead of HEEO, it's 17 tankers.)

I'm intensely skeptical that a Starship that lands in regolith dust will be reusable for more than 5 flights before it's easier just to scrap it.  If you can build a crew-rated Starship for $300M, that would be $270M a mission.  For a crew of 12, that's $22.5M/seat.

A lot depends on how quickly full crew-rating for Starship occurs.  Frankly, I don't think it will ever happen for a NASA mission, and my money's on no earlier than 2030 for something that SpaceX thinks is safe enough for informed consent.  I think

IMO, the moment that SpaceX does a D2-assisted lunar surface mission, or even a D2-assisted LEO-NRHO-LEO transit mission, SLS/Orion is over.  The cost disparity between such a mission and an SLS/Orion mission is simply too huge to be justified, even using Congress-logic.  I'd guess that you'd get at least 2-3 D2-assisted missions that way, and maybe more, if NASA is particularly conservative.

[yg1968]

My calculations likely overestimate the price but I would rather over estimate the price than under estimate it. A commercial lunar surface mission with F9/Dragon 9 and HLS-Starship would cost you about a $1B ($250M for F9/Dragon + $1150M x 68% for HLS-Starship = $1032M)(*). If you divide that by four, you get about $258M per seat. I find that too expensive for a private lunar surface mission, so that's why I don't think that SpaceX will ever offer that as an option for private missions.

The more likely scenario is a combination of crewed Starship and HLS-Starship with 12 people which gives you a price per seat between $130M and $156M according to my fairly conservative calculations above. That price is more reasonable since it is similar to what Russia was charging for a trip around the Moon a while ago (i.e., $150M). You would also expect a trip to the surface of the Moon would cost you twice as much as a trip to LEO.

(*) For the 68% ratio, see the assumptions in my post above.

[jketch]

My calculations likely overestimate the price but I would rather over estimate the price than under estimate it. A commercial lunar surface mission with F9/Dragon 9 and HLS-Starship would cost you about a $1B ($250M for F9/Dragon + $1150M x 68% for HLS-Starship = $1032M)(*). If you divide that by four, you get about $258M per seat. I find that too expensive for a private lunar surface mission, so that's why I don't think that SpaceX will ever offer that as an option for private missions.

I don't think that's too expensive for a commercial mission. Numerous people have paid nearly half that for a trip to orbit. I'm not sure how much Isaacman and Maezawa for their various missions, but I bet the total for each significantly exceeds $258M. A trip to the lunar surface is far more exclusive (and frankly, cooler) than what either of them are getting.

[NaN]

My calculations likely overestimate the price but I would rather over estimate the price than under estimate it. A commercial lunar surface mission with F9/Dragon 9 and HLS-Starship would cost you about a $1B ($250M for F9/Dragon + $1150M x 68% for HLS-Starship = $1032M)(*). If you divide that by four, you get about $258M per seat. I find that too expensive for a private lunar surface mission, so that's why I don't think that SpaceX will ever offer that as an option for private missions.

Another possibility is adding back the 3 Crew Dragon seats that were deleted during development, presumably a minor effort. This would allow a crew of 7, without needing to launch or reenter on Starship. While Dragon is small for a crew of 7, they would spend the bulk of the time on Starship. You still have to bring the HLS back to LEO, but it drops the per-seat price considerably from the 4-crew option and is a viable intermediate option.

[DanClemmensen]

... I doubt that there will be any mission using F9/Dragon and HLS-Starship. I don't think that either NASA or SpaceX are really interested in those for different reasons: NASA because they have to use SLS because of Congress and SpaceX because they believe that Starship is the future, not Dragon.

I thought that also, but then SpaceX announced that they will build a fifth Crew Dragon capsule AND they will extend the life of each capsule from five missions to ten or even fifteen. Prior to the extension, they had approximately one remaining unmanifested mission (eight flown, eleven on the manifest=19, but 4*5 lifetimes=20). After the extension, they will have between 31 and 56 unmanifested missions. Why?

[DanClemmensen]

My calculations likely overestimate the price but I would rather over estimate the price than under estimate it. A commercial lunar surface mission with F9/Dragon 9 and HLS-Starship would cost you about a $1B ($250M for F9/Dragon + $1150M x 68% for HLS-Starship = $1032M)(*). If you divide that by four, you get about $258M per seat. I find that too expensive for a private lunar surface mission, so that's why I don't think that SpaceX will ever offer that as an option for private missions.

Another possibility is adding back the 3 Crew Dragon seats that were deleted during development, presumably a minor effort. This would allow a crew of 7, without needing to launch or reenter on Starship. While Dragon is small for a crew of 7, they would spend the bulk of the time on Starship. You still have to bring the HLS back to LEO, but it drops the per-seat price considerably from the 4-crew option and is a viable intermediate option.

Apparently, it is a big deal to reinstate those seats and it may not be possible.:
     https://spaceflightnow.com/2019/12/07/after-redesigns-the-finish-line-is-in-sight-for-spacexs-crew-dragon/
Also, any redesign would require re-certification for crew, but the whole reason to use Crew Dragon instead of Crew Starship is that it is already certified.

[Coastal Ron]

Apparently, it is a big deal to reinstate those seats and it may not be possible.:
     https://spaceflightnow.com/2019/12/07/after-redesigns-the-finish-line-is-in-sight-for-spacexs-crew-dragon/
Also, any redesign would require re-certification for crew, but the whole reason to use Crew Dragon instead of Crew Starship is that it is already certified.

I don't know if it would be easy to add those three additional seats, but certifying that configuration would likely not be as hard as the initial crew certification. And there are numbers between 4 & 7 too that could be considered...

[yg1968]

... I doubt that there will be any mission using F9/Dragon and HLS-Starship. I don't think that either NASA or SpaceX are really interested in those for different reasons: NASA because they have to use SLS because of Congress and SpaceX because they believe that Starship is the future, not Dragon.

I thought that also, but then SpaceX announced that they will build a fifth Crew Dragon capsule AND they will extend the life of each capsule from five missions to ten or even fifteen. Prior to the extension, they had approximately one remaining unmanifested mission (eight flown, eleven on the manifest=19, but 4*5 lifetimes=20). After the extension, they will have between 31 and 56 unmanifested missions. Why?

For the Commercial LEO Destinations program would be my guess.

[VSECOTSPE]

I thought that also, but then SpaceX announced that they will build a fifth Crew Dragon capsule AND they will extend the life of each capsule from five missions to ten or even fifteen. Prior to the extension, they had approximately one remaining unmanifested mission (eight flown, eleven on the manifest=19, but 4*5 lifetimes=20). After the extension, they will have between 31 and 56 unmanifested missions. Why?

Margin.  They may lose a capsule.

[DanClemmensen]

... I doubt that there will be any mission using F9/Dragon and HLS-Starship. I don't think that either NASA or SpaceX are really interested in those for different reasons: NASA because they have to use SLS because of Congress and SpaceX because they believe that Starship is the future, not Dragon.

I thought that also, but then SpaceX announced that they will build a fifth Crew Dragon capsule AND they will extend the life of each capsule from five missions to ten or even fifteen. Prior to the extension, they had approximately one remaining unmanifested mission (eight flown, eleven on the manifest=19, but 4*5 lifetimes=20). After the extension, they will have between 31 and 56 unmanifested missions. Why?

For the Commercial LEO Destinations program would be my guess.

But SpaceX will try to crew-qualify the Starship as soon as they possibly can, so these expensive Crew Dragon missions serve only to provide early access in the event that there is a demand before crewed SS is available.
(edited to correct the quoting)

[oldAtlas_Eguy]

If a Crewed Starship is launching by 2030. Then the prices for a LSS to the lunar surface would be $450M or less Total. And then also with 12 persons transported for a 3 month stay on the surface in a ISS amount of roominess in the LSS for those 12. Would be ~$40M per person. It would involve a LSS leaving LEO for Lunar orbit. A tanker leaving LEO for Lunar orbit to fuel the LSS once it returns from the surface for it's return to LEO. A Cargo to LEO to replenish the supplies in the LSS prior to the mission. And finally a Crew Starship to and from LEO to Earth surface to get some number of crew to the LSS and then wait for their return to LEO. For much shorter periods such as 1 month of 25 persons for price of ~$20M per person. And then for a truly short duration of 2 weeks on the surface of 50 for $10M per person.

There are way too many options and too much speculation other than to identify floor and cieling for the cost per person to the surface for all commercial trip to the Lunar surface. My set was what I see as a near term floor that is possible in 10 years or less. But there are always some difficulties encountered along the way that can cause what is believed and what turns out to be realistic. F9 was believed that it could get prices to around $30 to 20M per flight as well. The reality is the best so far attained is a cost of $30M and a Price floor of $45 to 50M. The real note here is that 60% of cost savings occur in the ability to use a single vehicle 4 times. That is because the hardware effect on cost starts to become irrelevant to the operations costs of the vehicle once use gets to 10 or more. Such that it does not take much reuse to get a 50% price reduction and the ech subsequent factor 2 price reduction gets harder and harder to achieve.

Thus prices for HLS will drop rapidly at first then start slowing it's downward travel.

Which also predicts that SpaceX will likely make a lot of profit from that Option B contract mod. And subsequent mission prices may decrease applying additional pressure on any HLS competitor to keep up.

The BIG BUT is the Starship needs to work (fly to orbit) as well as on-orbit refueling.

[TheRadicalModerate]

Apparently, it is a big deal to reinstate those seats and it may not be possible.:
     https://spaceflightnow.com/2019/12/07/after-redesigns-the-finish-line-is-in-sight-for-spacexs-crew-dragon/
Also, any redesign would require re-certification for crew, but the whole reason to use Crew Dragon instead of Crew Starship is that it is already certified.

That sounds like something that would be hard to revert to for a NASA-certified design, but my understanding is that the change from 7 to 4 was about NASA getting cold feet about parachute and landing forces:

A requirement change from NASA also contributed to delays, Shotwell said.

After SpaceX had already designed the interior layout of the Crew Dragon spacecraft, NASA decided to change the specification for the angle of the ship’s seats due to concerns about the g-forces crew members might experience during splashdown.

The change meant SpaceX had to do away with the company’s original seven-seat design for the Crew Dragon.

“With this change and the angle of the seats, we could not get seven anymore,” Shotwell said. “So now we only have four seats. That was kind of a big change for us.”

If SpaceX considered the forces acceptable but NASA didn't, then reducing or eliminating the rotation of the seats¹ might allow the other 3 seats to be added back in for private missions.  This is obviously a big change, and might be more trouble than it's worth.  But if it enabled crews of 7 to the LSS, that might drum up quite a bit of early private lunar business for SpaceX, allowing the cost to be split amongst more billionaires.  And a private mission, either cislunar or to the surface, would of course be an excellent way of shaming NASA/Congress into using the 4-person D2 to reduce the price of at least some Artemis missions.

Again, this only makes sense if Starship isn't crew-certifiable for launch/landing for a long time, or that it's not certifiable by NASA for its missions.  Prudence would seem to dictate that SpaceX have a Plan B if they can't allay NASA's fears--or their own, for that matter.

___________
¹Off-topic, which is why this is in a footnote:  My understanding (which is a little murky) is that there are two positions for the seats:

1) Perpendicular to the axis of thrust of liftoff, a potential abort, or on-orbit operations.  This is rotated up so that the crew are very close to the control consoles.

2) Perpendicular to the load vector when the parachutes pop and the D2 splashes down, which isn't through the axis of symmetry.  I've been assuming that this seat position is the same as the "boarding" position, where the tops of the seats are rotated down (i.e., aft).

One thing I don't understand is that this seems to require rotating the crew farther away from the controls during reentry.  That seems like a terrible design for a flight mode with high accelerations. 

Maybe I have this backwards, and the flight position isn't perpendicular to the launch acceleration, but is to the landing accelerations?  That would make adding the 3 seats an even bigger mod to the D2, because they'd have to move the consoles if the seats were more perpendicular to the axis of symmetry.