Could Blue Origin offer its own rocket to the Moon, Page 2: low cost crewed lunar landers.

 Copyright 2025 Robert Clark

 In the last blog post, "Could Blue Origin offer it’s own rocket to the Moon?", I suggested that with technically feasible upgrades of the New Glenn booster engine, New Glenn might be transformed into a Saturn V-class, 100 tons to LEO, Moon rocket.

 An objection raised to the calculations I presented there was that the maximum New Glenn first stage tank size I was using did not include ullage space, i.e., the space left unfilled or filled with gas to account for boiloff. Three possible solutions: first, even with the commonly used estimate of ca. 1,150 tons propellant load it would require just a ca. 10% increase in tank size to get the prop. load in the 1,300 ton range. SpaceX has shown that additional tank rings have been swapped in and out of the Starship to get an additional propellant load increase of this size or more. 

 Second, an announcement from the Texas State Senate has indicated Blue Origin has been assigned a grant to increase the New Glenn prop.load by subcooling, i.e., densifying the propellant. Propellant subcooling typically results in an approx. 10% propellant load increase. 

 Third, New Glenn's, Moon lander uses hydrolox so it must make use of some zero-boil-off tech to not lose too much hydrogen over a mission lasting several days. This same tech might be able to be used on the New Glenn first stage to minimize the need for ullage.

 Therefore we'll work on the basis the New Glenn can be upgraded to get ca. 100 tons to LEO as expendable.

Getting a crewed lander.

 The space industry was pleasantly surprised by Blue Origin's New Glenn being able to reach orbit on its first launch. They were even more surprised by the announcement the next mission planned will take a cargo lander to the Moon as early as March, though more recently they've only said sometime in late Spring.

 The success of Blue Origin reaching orbit on the first launch with New Glenn and the rapidity at which they wish to progress to launching a lunar lander on the Moon shows the importance in having a top notch Chief Engineer such as David Limp making the technical decisions. If SpaceX had taken the route of hiring a true Chief Engineer, they would already be flying the Starship with paying customers at least in expendable mode. Moreover, they would recognize having a launcher as expendable with 250 ton capacity means they could do single launch missions to the Moon or Mars, no SLS, no multiple refueling flights required.

 As it is, SpaceX is in real danger of being lapped by Blue Origin in having a manned Moon rocket or even a Mars rocket.

  Blue Origin has stated their Blue Moon Mk1 cargo lander will have a 21,350 kg fueled mass, and payload of 3,000 kg payload to the Moon one-way.

Blue Moon Mk1 cargo lunar lander.

 Given the delta-v requirements for getting to the Moon we can make estimates of its propellant and dry mass values:

Delta-V budget.
Earth–Moon space.

https://en.wikipedia.org/wiki/Delta-v_budget#Earth%E2%80%93Moon_space%E2%80%94high_thrust

 Reports are the current version of the New Glenn has a payload to LEO of 25 tons. A 21,350 kg fueled mass of the Blue Moon Mk1 lander plus 3 tons cargo would be 24,350 kg, just under the payload capacity of the current New Glenn.

This though means Blue Moon has to provide the delta-v for trans-lunar injection(TLI) and insertion into lunar orbit as well as lunar landing. From the table the total of TLI and insertion into low lunar orbit and landing is 5.93 km/s, 5930 m/s.

 The engine on the lander is supposed to be the BE-7 hydrolox engine upgraded from the BE-3 used on the New Glenn's upper stage. We'll assume the BE-7 has about the same vacuum Isp of the BE-3, of 445 s. Then taking the propellant load of the Blue Moon as 18.35 tons and dry mass as 3 tons allows it to get 3 tons in cargo to the 5,930 m/s delta-v needed to go from LEO to the lunar surface, plus some margin:

445*9.81Ln(1 + 18.35/(3 +3)) = 6,110 m/s.

 The Blue Moon Mk1 is also already developed and paid for by Blue Origin on its own dime. And it is established fact at this point that spaceflight components, rockets or spacecraft, as developed by commercial space, and privately funded saves 90% off the previous governmentally financed approach that is paid for by governmental space agencies such as NASA. 

 A key fact not yet generally recognized is that we are already at the long desired point of having spaceflight being sufficiently low cost that it can be fully financed by commercial space and private funding only, no governmental financing required at all. BUT such low costs hold true only if it is privately funded.

 A majorly important example is the Mars Sample Return mission. There is much hand-wringing at NASA and among space science advocates about the $10 billion price tag estimated by NASA for MSL. But in point of fact this mission and all space science missions going forward can be paid for at 1/100th the costs estimated by NASA by following the commercial space approach. And in fact the costs as privately funded would be so low, such missions could even be mounted as privately financed at a profit. See discussion here:

Low Cost Commercial Mars Sample Return.
https://exoscientist.blogspot.com/2023/07/low-cost-commercial-mars-sample-return.html

 The argument for this is quite simple. SpaceX and now multiple other space startups have confirmed that development costs as privately funded are 1/10th the costs of governmental funded development costs. But then production costs of individual space components rockets or spacecraft are commonly 1/10th or less than their development costs. As a space company paying for a space project on your own dime, rather than paying the large development costs of a new component you would just naturally use ones that already exist, resulting in far smaller outlay on your end. Then taking into account 1/10th cheaper development cost overall as privately financed and 1/10th or lower cost using already existing components, rather than developing them from scratch, the result is 1/100th or less cost than the usual development costs estimated by NASA following the government financed approach.

 So we already have a lander in the Blue Moon Mk1. But could this serve as a crewed lander? Yes, it can because of a key fact being overlooked by NASA: Artemis is not Constellation's Apollo on steroids, It is in fact Apollo 2.0.

 Perhaps NASA didn't want to acknowledge this so that it would continue to get funding. Just saying Artemis is Apollo redone would not sound nearly as impressive or necessary. But it is important to understand this point. 

 The argument for this conclusion is quite elementary. The primary launcher of Constellation was the Ares V. It was intended to have a startling 188 tons to LEO payload capacity. But there was more to Constellation than that still. The crew were intended to be launched separately to LEO by the Ares I. This had the payload capacity to LEO of 25 tons. Then the Constellation plan with its two launchers could get ca. 210 tons to LEO. This is about twice that of Apollo, but more importantly its about twice as much as Artemis. So in point of fact in the key measure of payload mass to orbit Artemis is Apollo. It is far from Constellation was capable of.

 Once, this is understood then it is understood Artemis should not try to get a lander the size of the Altair lander of Constellation at 45 tons. It should try to get one comparable in size to Apollo. 

 Instead, NASA is seeking that Altair sized lander such as the crewed version of the Blue Origin lander, the Blue Moon Mk2 also at 45 tons, 

Blue Moon Mk2 crewed lunar lander.

or, worse seeking to get the 1,200 ton Starship HLS with multiple refuelings to fit in the Artemis architecture.

 Instead we'll show the Mk1 cargo lander can form the lunar lander for single launch crewed lunar mission format based on the New Glenn as launcher. 

Architecture 1: this will be analogous to the Early Lunar Access proposal of NASA, a proposed follow-on to Apollo.

https://web.archive.org/web/20081106190735/https://nss.org/settlement/moon/ELA.html

 The salient feature of this proposal is it used a single crew capsule for the full round trip from Earth orbit, all the way to the lunar surface, and back to Earth, thus no separate lunar module, i.e., no lunar orbit rendezvous(LOR).

 You see from the table of delta-v's the delta-v needed from the lunar surface back to Earth is 2.74 km/s, 2,740 m/s. This would not put you in Earth orbit though but on a ballistic return trajectory to reenter Earth's atmosphere, a la the Apollo command module. 

 The total round-trip delta-v would be 2.74 km/s + 5.93 km/s = 8.67 km/s, 8,670 m/s.

 The extra delta-v could be provided by the Delta IV Heavy's upper stage, now being used for the interim upper stage of the SLS. This stage would be put atop the New Glenn as a 3rd stage performing the role of a "Earth Departure Stage" for the push to translunar injection. Carrying the Mk1 with a 3 ton crew module it could get:

465*9.81Ln(1 + 27.2/(3.5 + 24.35)) = 3,110 km/s, sufficient for translunar injection(TLI) of the 24.35 ton total mass of the Mk1 lander and crew module.

 This 3rd stage plus the Mk1 and crew module would have a total mass of 30.7 + 24.35 = 55.05 tons. The cited 45 ton payload capacity of the New Glenn to LEO was a for a partially reusable version, with the booster landing downrange. Then for an expendable use it should get ca. 60 tons to LEO, sufficient for the purpose. 

 However, the key question is of a crew capsule that would be analogous to the Apollo Command capsule or the Orion capsule or the Dragon capsule but only at ca. 3 tons dry mass. This is only half the dry mass of the Apollo Command capsule but required to play a similar role.

 A research report of Prof. David Akin of the University of Maryland aerospace department suggests this is indeed possible:

Phoenix: A Low-Cost Commercial Approach to the Crew Exploration Vehicle

Abstract: Since the announcement of President Bush’s Vision for Space Exploration (VSE) in early 2004, the architecture of Project Constellation has been selected. The system will be centered around the Crew Exploration Vehicle (CEV), which has been dubbed by NASA administrator Michael Griffin as “Apollo on steroids”. The CEV is to be launched on a new launch vehicle, derived from existing shuttle technology. The development of this new
spacecraft and launch vehicle is a very costly proposition. An alternate approach is proposed in this study. The Phoenix is a smaller spacecraft designed specifically to be launched on the Falcon 5 vehicle under development by SpaceX. Because the SpaceX vehicle will cost only a fraction of today’s launch costs, the Phoenix is estimated to cost less than half of the price of the CEV. This reusable three person capsule utilizes an innovative re-entry concept, which allows for a cylindrical spacecraft with greater interior volume. This extremely cost-effective spacecraft is an attractive option for fulfilling VSE requirements.

 Below is page 3 from this report:

 Since the Cygnus cargo capsule of Orbital Sciences, now a division of Northrop Grumman, of comparable size to the Phoenix proposal, already exits I suggest basing it on the Cygnus just given life support and heat shield. Remember our dictum is, "Use existing resources to save on costs if available."

 The proposed heat shield for the Phoenix was a "parashield", a combined parachute and heat shield:

 And a proposed heat shield of the Cygnus to make it reusable was an inflatable:

  These may indeed work. But to get to an operational system minimizing development work and cost I advise simply making the Cygnus tapered like most manned capsules and using a traditional heat shield beneath it:

 For both the Soyuz and Dragon, they have relatively small taper angle so you would lose a relatively small size in capsule interior volume by giving the Cygnus a similar side taper.

 Quite notable is with this option you can get a crewed Moon mission with only a single launch of a 60 ton to LEO launcher. Then both the New Glenn as expendable or the Falcon Heavy as expendable could do it in a single launch.

 Robert Zubrin had proposed a Moon mission architecture using the Falcon Heavy with his "Moon Direct" proposal but it would require two launches of the Falcon Heavy to do it. This alternative approach could do it in a single launch provided it is indeed possible to produce an Apollo Command module analogue of dry mass only 3 tons.

Architecture 2: an Apollo sized capsule.

 The Apollo architecture that had the Apollo Command Module to carry the astronauts for the in space portion of the trip from LEO to lunar orbit with a separate smaller capsule for the lander, had an advantage in providing backup capability. This was quite fortunate during the Apollo 13 mission when the Apollo LEM had to sustain the crew for a part of the time on the way back to Earth.

 There is still the question of whether you can make the Apollo Command Module analogue only at 3 tons dry mass. So here we'll do the calculations for an analogous architecture to that of Apollo with a main crew capsule for the in-space portion of the flight and a smaller, separate crew module for the lander.

 I estimated above the Blue Moon Mk1 lunar lander has about a 6 to 1 propellant load to dry mass ratio, at 18.35 tons prop load to 3 tons dry mass. But the Mk1 was designed to do all the propulsion from LEO, to translunar injection(TLI), to low lunar orbit insertion, to lunar landing, with a 3 ton cargo. If the only thing required is to go from low lunar orbit to the lunar surface and back with a 3 ton crew module then a much smaller lander can be used. 

 I'll assume you can a smaller lander at 1/3rd the Mk1 size with a 6 ton prop load while maintaining the 6 to 1 prop mass to dry mass ratio, so 1 ton dry mass. First, from the Earth-Moon delta-v table, the delta-v one way from low lunar orbit to the lunar surface is 1,870 m/s. Then the round-trip delta-v is 3,740. Note now, the smaller lunar lander can provide a delta-v of:

 445*9.81Ln(1 + 6/(1 + 3)) = 4,000 m/s, sufficient for the round-trip from lunar orbit to the surface and back to lunar orbit.

 Now we need a propulsive stage to do the burn to insert the 6 ton main crew capsule and 10 ton lander into low lunar orbit, and to do the burn to bring the main capsule back to Earth, a la the Apollo architecture. For this we'll use a stage half-size to the Mk1 at 9 ton prop load and 1.5 ton dry mass.

 The burn to escape low lunar orbit is commonly estimated as 800 m/s to 900 m/s, same as that for the burn to enter into low lunar orbit. Then 2 tons of propellant is required to be left over as reserve for the return of the primary capsule to Earth, the lander being jettisoned a la the Apollo architecture:

445*9.81Ln(1 + 2/(1.5 + 6)) = 1,030 m/s.

 Then 7 tons of propellant out of 9, with the 2 tons left in reserve for the return, is sufficient to put the 6 ton primary capsule and the 10 ton lander into low lunar orbit:

445*9.81Ln(1 + 7/(1.5 + 6 +10 +2)) = 1,340 m/s.

 The rather large margin of 1,340 m/s over the maximum 900 m/s needed to insert into low lunar orbit suggests we might be able to do with a somewhat smaller stage for this purpose, perhaps 7 tons instead of 9 tons prop load.

 Now the total mass that needs to be sent to TLI is 9 + 1.5 + 6 + 10 = 26.5 tons. We'll use again the upper stage of the Delta IV Heavy to do the TLI burn:

465*9.81Ln(1 +27.2/(3.5 + 26.5)) = 2,940 m/s. 

 This is slightly less than the value commonly given for TLI in the range of 3,000 m/s to 3,100 m/s. But the propulsive stage that's used to insert into lunar orbit had so much margin that it could be used to provide the slight extra push to make TLI.

 Or as I mentioned that propulsive stage for the lunar orbit insertion, essentially reprising the role of the Apollo's Service Module, had so much margin we could make it smaller to ca. 7 tons prop load. Then the TLI total mass would be the same as the Architecture 1 case. And the Delta IV Heavy's upper stage could get the total mass to TLI on its own. 

 It's still quite notable that doing it either way we still could launch the full system to orbit on a 60 ton to LEO launcher.

Flights to the Moon at costs similar to costs of flights to the ISS. 

 I said Artemis is really Apollo redone based on its payload size. It is not Constellation. It is not "Apollo on Steroids". Does it have any value then? I am arguing the goal of getting sustainable lunar habitation is important and doable now. It probably can't be done by Artemis though in a sustainable fashion considering that both the Orion capsule and SLS already each, separately cost $2 billion per flight. When you add on the over-large proposed landers the SpaceX HLS or the New Glenn MK2 each costing ca. $2 billion per flight, and the the Boeing EUS, advanced composite casing SRB's, and lunar Gateway, the total per flight would be in the range of $8 billion to $10 billion per flight.

 It is now becoming increasing likely that Artemis will be cancelled. The only question now is will it be cancelled before Artemis II or will Artemis II be allowed to fly and then the program would be cancelled.

 However, the most important fact is sustainable lunar habitation can be done following the commercial space approach making use of already existing space assets. As I mentioned the combined effect of both these factors can cut the costs of such missions by a factor of 1/100. For example both the Falcon Heavy and the New Glenn cost in the range of ca. $100 million. The small size of the additional in-space stages probably can be done for less than $100 million under the commercial space approach.

 And the crew capsules? An unexpected calculation suggests they can be done together for less than $100 million. For instance back in 2009, Orbital Science contracted Thales Alenia  to construct the Cygnus capsule for 180 million euros for 9 capsules, about 20 million euros each.

 A further contract Thales Alenia made with Axiom Space illustrates how low cost such modules can be while illuminating also how much more expensive space systems are when government funded compared to being privately funded. A contract Thales Alenia made to Axiom Space for two space station modules was only $110 million for two:

THALES ALENIA SPACE TO PROVIDE THE FIRST TWO PRESSURIZED MODULES FOR AXIOM SPACE STATION
14 JUL 2021
Rome 15 July, 2021 – Thales Alenia Space, Joint Venture between Thales (67%) and Leonardo (33%), and Axiom Space of Houston, Texas (USA), have signed the final contract for the development of  two key pressurized elements of Axiom Space Station - the world’s first commercial space station. Scheduled for launch in 2024 and 2025 respectively, the two elements will originally be docked to the International Space Station (ISS), marking the birth of the new Axiom Station segment. The value of the contract is 110 Million Euro.

https://www.thalesgroup.com/en/worldwide/space/press_release/thales-alenia-space-provide-first-two-pressurized-modules-axiom-space

 The individual modules have about 75 cubic meters pressurized space for four crew members, and already have life support systems.

 Now compare that to the HALO module Northrop Grumman contracted with NASA to produce at a cost of $935 million:

Northrop charges on lunar Gateway module program reach $100 million.

by Jeff Foust

January 25, 2024

Northrop received a $935 million fixed-price contract from NASA in July 2021 to build the module, which is based on the company’s Cygnus cargo spacecraft. HALO will provide initial living accommodations on the Gateway and includes several docking ports for visiting Orion spacecraft and lunar landers as well as additional modules provided by international partners. It will launch together with the Maxar-built Power and Propulsion Element (PPE) on a Falcon Heavy.

https://spacenews.com/northrop-charges-on-lunar-gateway-module-program-reach-100-million/

Based on the "Super" 4-Segment version of the Cygnus, it might have a volume of ca. 33.5 cubic meters:

 The Axiom Space AxH1 habitation modules at 70 cubic meters have double the space of the HALO modules but, as privately financed, cost less than 1/10th as much as government financed HALO modules.

 The needed crew module would be well cut down in size from the 70 cubic meters of the Axiom space station habitation module, with a comparable reduction in cost. Addition of a heat shield would cost a fraction of the total cost of the crew module itself.

 Then the crew modules for the main capsule or of the lander module might cost in the range of a few 10's of millions of dollars.

Possibilities for a single launch architecture of the Artemis missions, Page 4: lightweight landers from NRHO to the lunar surface.

 Copyright 2024 Robert Clark

 Congress is becoming increasingly concerned that with the continuing delays of the Artemis missions that China may beat the U.S. back to the Moon:

US must beat China back to the moon, Congress tells NASA.
By Mike Wall 
'It's no secret that China has a goal to surpass the United States by 2045 as global leaders in space. We can't allow this to happen.'
https://www.space.com/us-win-moon-race-china-congress-artemis-hearing

 I had previously proposed correcting an error in the design of Orion's service module that instead of making it larger than Apollo's service module because of Orion's twice larger size, instead made it 1/3rd smaller:

Possibilities for a single launch architecture of the Artemis missions, Page 2: using the Boeing Exploration Upper Stage.

https://exoscientist.blogspot.com/2023/08/possibilities-for-single-launch.html

 The proposal was to give an additional approx. 10 tons propellant to the service module. This would allow the Orion capsule/service module stack plus an Apollo-size lander to be carried all the way to low lunar orbit, not just to NRHO(near-rectilinear halo orbit). 

 This though because of the higher payload may require use of the higher thrust J-2X engine on the Boeing EUS(upper stage) rather than the 4 RL-10 engines now planned on the SLS Block 1B. It's higher thrust would result in a greater payload to LEO and TLI, perhaps to ca. 120 tons to LEO rather than the 105 tons planned to LEO.

This approach requires additional propellant tanks be added to the service module and a change in the EUS upper stage engine to the J-2X. As I discussed in that blog post, it may also require an additional Centaur V sized third stage be added atop the Boeing EUS. This is dependent on what is the TLI(trans lunar injection) payload for the Boeing EUS using the J-2X engine. It may be it can perform the needed TLI payload without an additional Centaur V 3rd stage.

 In any case, I'll propose here an alternative approach to a single launch Artemis architecture without increasing the service module propellant load. This again will use a light-weight Apollo-sized lander with all the components of Orion capsule/Service Module/lunar lander all carried on that one single SLS launch. Because of the lower propellant load on the service module though I'll also send it to NRHO instead of to low lunar orbit.

 Note the NRHO was chosen by NASA as the orbital location because it has a lower delta-v requirement to get there than going to low lunar orbit. Here’s the the delta-v requirements:

 The second group of delta-v’s shows the delta-v to NRHO as 0.45 km/s and the delta-v to and from the lunar surface from NRHO as 2.75 km/s, or 5.5 km/s round trip.

 I’ve seen various numbers for the Orion and service module dry mass and propellant mass. I’ll use 16.5 total dry mass for the Orion+service module together, and 9 tons of service module propellant mass, but only 8.6 tons of this as usable propellant because of residuals.

 Then we'll use 6 tons of Service module propellant to get the Orion/Service Module/lunar lander to NRHO after being placed on TLI trajectory by the EUS, for the 16.5 ton Orion/Service Module dry mass, and 15 tons gross mass Apollo-sized lander with 2.6 tons left over for the return trip.

 We'll need every bit of performance to accomplish the mission within these constraints. So we'll assume we can get a 324 s Isp out of the storable propellant engines on the service module. This is higher than specified for the Orion service modules engines but is doable because of the storable propellant Aestus engine on the Ariane 5 EPS storable propellant upper stage which gets this vacuum Isp. We'll assume we can get this increased Isp by using a larger expansion ratio nozzle or even by swapping out the engine on the service module to use the Aestus engine. Then we get:

324*9.81Ln(1 + 6/(16.5 + 15 + 2.6 + 0.4)) = 510 m/s, or 0.51 km/s, sufficient for placing the stack in the NRHO orbit, where the 0.4 in the equation is for the unburnt residuals.

 Then with the 2.6 tons usable propellant left over for the return trip, after the lander is jettisoned, we get:

324*9.81Ln(1 + 2.6/(16.5 + 0.4)) = 450 m/s, 0.45 km/s, sufficient for the Orion return.

 To increase performance even more we may want to switch even to the RS-72 engine. This is a turbopump-fed storable propellant engine with a vacuum Isp of 340s. It achieves this by using a higher chamber pressure of 60 bar and higher nozzle expansion ratio of 300 to 1 than the Aestus engine. A turbopump engine also has lower residuals, typically less than 1%. A disadvantage is that pressure-fed engines are simpler with fewer moving parts, and so higher reliability, important for an engine to place the spacecraft in orbit and for leaving orbit.

 Now for the ca. 15 ton gross mass lander, because of the higher delta- v needed from NRHO we’ll use hydrolox rather than storable propellant stage. The Ariane 4 H10 hydrolox upper stage had a 11.8 ton propellant mass and 1.2 ton dry mass. We’ll use a 2 ton dry mass of the crew module:

ORBITAL PROPOSES FUTURE DEEP SPACE APPLICATIONS FOR CYGNUS.
SPACEFLIGHT INSIDER
MAY 1ST, 2014
Orbital’s proposal, outlined in this PDF, involves docking a Cygnus spacecraft with Orion to serve as a habitation and logistics module on longer flights. For these missions, the re-purposed Cygnus would be called the Exploration Augmentation Module (EAM). With its current life support systems used to transport pressurized cargo and experiments to the ISS, Cygnus is stated as being already suitable for the long term support of a crew. While berthed to Orion, Cygnus could support a crew of four for up to 60 days. Cygnus also has the capability of storing food, water, oxygen, and waste and features its own power and propulsion systems. The EAM would utilize the enhanced configuration Cygnus, which will begin flying larger cargoes to the ISS beginning with CRS-4 in 2015. An even larger version is also being proposed, featuring a 4-segment pressurized cargo module.

https://www.spaceflightinsider.com/missions/commercial/orbital-proposes-future-deep-space-applications-cygnus/

 Note though the phrasing here is ambiguous. The Cygnus capsule as used as a cargo transport to the ISS contains air, as it would have to for the astronauts at the ISS opening it to retrieve the cargo, but not life support systems. I'm inclined to believe for the usage cited in this article it would be taking life support from the Orion capsule. Then the calculations need to be made for how much mass it would take for life support, thermal management, consumables for an independent crew module.

 Now for the delta-v calculation for our hydrolox lander, we'll assume we can match the max 465 s Isp of the RL-10 engine by giving the Ariane 4 upper stage engine a nozzle extension as used on the RL-10, then we get:

465*9.81Ln(1 + 11.8/(1.2 + 2)) = 7,000 m/s, 7 km/s. This is quite a bit higher than the 5.5 km/s needed for the round trip from NRHO to the lunar surface and back again. But it uses hydrolox propellant so needs extra mass for low-boiloff tech. 

 Low boiloff-tech and long duration hydrolox stages are an important enabling technology. ULA engineers and ULA CEO Tory Bruno have written about this extensively in regards to for example the proposed ACES derivative of the Centaur upper stage. Because of the prior research on low-boiloff tech, an operational version to be fielded in a short time frame to be used on the Artemis missions likely can be done. 

 This shows a single launch mission is doable if going to NRHO, but it is not my preferred plan. A complete orbit around the Moon at NRHO altitude takes about a week, and for the Orion capsule being at NRHO and not low lunar orbit, the lander's crew would have to remain on the Moon about a week before they could return to the Orion in the NRHO orbit. The landers crew module would have to be larger with heavier life support and consumables in this scenario.

 If instead the Orion was at low lunar orbit it takes two hours to complete an orbit and the lunar lander could launch every two hours to rendezvous with the Orion.

 Since the Orion's service module being given an insufficient propellant load is such an obvious design mistake, the preferred route to take would be to correct that error, thereby allowing the missions to take place from low lunar orbit instead of from NRHO.

  Robert Clark

SpaceX should withdraw its application for the Starship as an Artemis lunar lander, Page 2: The Raptor is an unreliable engine.

 Copyright 2023 Robert Clark

 I had earlier argued that SpaceX should withdraw the Starship as a lunar lander. The primary basis for this was for safety of the surrounding population in case of an explosion on launch, SpaceX should withdraw its application for the Starship as an Artemis lunar lander.

 However, an additional reason why the Starship should not be used for a lunar lander is for safety of the crew. In the blog post, Did SpaceX throttle down the booster engines on the IFT-2 test launch to prevent engine failures?, I noted two separate methods of calculation suggest the SuperHeavy booster was throttled down to <75%. I also suggested the Starship upper stage was fired at ~90%. Given this difference in thrust power levels, I suggested the booster completed its portion of the ascent because it was throttled down and the upper stage did not because it was at close to full thrust. 

 Even though the booster engines successfully fired during the ascent, the booster exploded during the attempted return. One explanation offered was the engines were damaged by fuel slosh during flip of the booster. However, it should be noted the Starship during tests of the landing procedure, that at least one Raptor always leaked fuel and caught fire.

 Note even in the last two shown here, SN10 and SN15, there were engine fires on landing. For SN10 the engine fire led to the vehicle exploding a few minutes after landing. For SN15 the fire was extinguished before it caused an explosion. SN15 was called  a “successful” landing test because it did not explode. But that a Raptor still caught fire during this test gives further evidence the Raptor is still not a reliable engine. 

 And SN11 experienced a catastrophic explosion after a fuel leak and engine fire: 

 Since relighting the Raptors in flight always resulted in an engine fire, that is the most likely explanation for the IFT-2 booster explosion as well.

SpaceX Misleadingly Characterizes Raptor's Qualification for Flight.

 SpaceX has been using the term "full duration" for their Raptor static fire tests when they might only last 5 seconds. In the rest of the industry other than SpaceX, a full duration static test means firing for the full duration of an actual launch. 

Rocket Factory Augsburg

@rfa_space

280 seconds of glorious hot fire! We are incredibly proud to be the 1st private company in #Europe () to hot fire a staged-combustion upper stage for its full duration. This qualifies our upper stage and Helix engine for flight Enjoy the video and read more in our press release https://bit.ly/3WJY2G4

https://x.com/rfa_space/status/1664683388928655374?s=20

And for the four SSME's on the SLS core stage:

 SpaceX calling their 5 second long test fires "full duration" misleadingly gives the impression that is sufficient to qualify the engines for full mission flight time.

No estimates for Raptor engine reliability publicly provided.

 For engines for a craft intended to carry astronauts and for which billions of dollars of public funds are earmarked there should be provided some indication about the safety and reliability of such engines. For instance this report provides estimates of the reliability of the different components of the SLS:

SLS-RPT-077
VERSION: 1
National Aeronautics and Space Administration
RELEASE DATE: MARCH 8, 2013
SPACE LAUNCH SYSTEM PROGRAM (SLSP)
RELIABILITY ALLOCATION REPORT

https://foia.msfc.nasa.gov/sites/foia.msfc.nasa.gov/files/FOIA%20Docs/42/SLS-RPT-077_SLSP-Reliability-Allocation-Report.pdf

 But no such estimates for the Raptor have been provided. That so many engines have consistently failed in actual flights suggest they have quite low reliability.

 In the scenario of the Merlin engines used for crewed flight, over 80 missions of the Falcon 9 were successfully flown before the first crewed flight. That means over 800 successful firings of the Merlins during that time. And added on after that the many launches since then, over one thousand successful firings of the Merlins have been made.

  Robert Clark

SpaceX should withdraw its application for the Starship as an Artemis lunar lander.

 Copyright 2023 Robert Clark

 In the blog post, "SuperHeavy+Starship have the thermal energy of the Hiroshima bomb. UPDATED". I noted the thermal energy content of both stages is comparable to the explosive force of the Hiroshima bomb, ca. 15 kilotons of TNT. However, it is quite important to keep in mind that NASA uses estimates of the explosive force of a possible rocket explosion that is some fraction of what the total thermal energy might be. Based on this, I estimated the explosive force might actually be in the range of 3.4 to 5.4 kilotons. This is as much as 5 times higher than the explosive force attributed to the famous Soviet N-1 rocket failures at ca. 1.2 kilotons. 

 To get an idea of the enormity of 1.2 kilotons explosive force, and remembering also an SH/ST explosion might be as much as 5 times more powerful, look at the case of the Beirut explosion of 2020. This was not a rocket explosion but of ammonium nitrate but its estimated explosive force was about that of N-1 rocket at ca. 1.1 kilotons.

 

 In the explosion, homes as far away as 10 kilometers were damaged and the terminals at the Beirut airport 10 km away suffered moderate damage with some doors and windows blown out.

 Note now populated areas such as Port Isabel consisting of thousands of residents are within 10 km of the SH/ST launch site, and an explosion of this rocket might be 5 times more powerful than the Beirut explosion.

 In my blog post, I argued that not sufficient attention was being given to the possibility of an explosion by either the FAA or NASA. I wrote to one of NASA's safety offices and was told the safety of commercial launches is not the purview of NASA, but of the FAA. But when NASA is depending on that the commercial rocket to complete the planned flagship space program of NASA they have a responsibility to ensure that rocket is being developed safely as well.

 With the failure of the April 20th test flight of the Superheavy/Starship fortunately now both NASA and the FAA are giving closer scrutiny of the safety of the rocket as it should be.

The Superheavy/Starship actually is the N-1 rocket.

 The explosive force of the N-1 rocket, comparable to that of the devastating Beirut explosion, serves as a cautionary tale for those in the space industry. The comparison has been made of the SpaceX SuperHeavy/Starship approach to the Soviet multiple failed N-1 rocket in that they both wanted to test by actually flying the full rocket until it works, despite the number of failures. 

 This comparison was criticized on the grounds the N-1 engines were not tested individually. Instead, the engineers on the N-1 selected an engine at random from a batch to see if that worked. If it worked the entire batch was chosen. The engines could not be tested individually because the testing was destructive. That engine could not be used if it were first tested.

 The SpaceX Raptor engines on the other hand are tested individually. But here’s the major failing of the Raptor: even if the engine is tested successfully there is still a quite high chance the engine will still fail when used on a flight. That is a major flaw in a rocket engine. No rocket engine would be considered successfully developed with that flaw.

 Because of the numerous failures of the Raptor both on the test stand and in short test hops of the Starship landing methods prior to the April test flight, I estimated the chance of engine failures of the SuperHeavy/Starship test flight was 1 out of 3. SpaceX claimed prior to the April test flight their Raptor 2 was more reliable. The result? Only 1 in 4 of the engines failed. That is still a stunningly high percentage. As a point of comparison it would be like on every flight of the Falcon 9 the expectation would be at least two of the engines would fail during each flight.

 The upshot of this in a very real sense the Super/Starship is just like the failed Soviet N-1 in flying with engines with poor reliability.

 It is my contention the attempt of SpaceX trying to reach a 2025 deadline to have the SH/SS flying and with multiple successful test flights completed puts undue pressure on its normal safety procedures. For that reason my opinion is it should withdraw the Starship for consideration as a lander for the Artemis III lander mission.

   Robert Clark

Possibilities for a single launch architecture of the Artemis missions, Page 3: Saving the lander mission for Artemis III.

 Copyright 2023 Robert Clark

  I discussed a possible single-launch lunar lander architecture here:

Possibilities for a single launch architecture of the Artemis missions, Page 2: using the Boeing Exploration Upper Stage.

http://exoscientist.blogspot.com/2023/08/possibilities-for-single-launch.html

 The SpaceX delay in the Starship HLS development has led to NASA considering that Artemis might not even be a lander mission. This leaves open the possibility to save the lander mission for the Artemis III mission alternative methods for landers should be considered.

 Plus, there is the fact many knowledgeable space aficionados from the old days really do not like the SpaceX plan of using 8 to 16 refueling flights just for one lunar lander mission.

 This plan for a replacement lander could be done rather quickly and at low cost because it would use already existing space assets. Also, it would be done by our European partners so would not require NASA expenditures using all European space components. That would save $3 billion that NASA would have had to pay to SpaceX for their lander.

 This would involve even greater European involvement in successfully accomplishing the Artemis missions than just Orion’s service module so would undoubtedly get enthusiastic support from the ESA.

 It is notable that the ESA has expressed even greater support for lunar colonization plans than even NASA. This ESA produced lunar lander would allow them to further their own plans for a sustained human presence on the Moon.

 About ESA’s ATV-derived service module for Orion, that again would require low cost modifications in this plan. It would need just an addition 10 tons of propellant, which would fit easily within a service module diameter expanded to match the Orion’s diameter. Again this cost would be covered by our European partners, with no expenditure by NASA.

 Those two factors would be the easiest aspects of the plan. It might be difficult to believe a lunar lander would be among the “easiest” parts of the plan. But keep it mind it would be derived from already existing space assets.

 The trickiest aspects of the plan would be the fact the SLS would require higher payload capability to allow for the higher propellant load of the service module of 10 tons and a ca. 15 ton mass lunar lander.

 One possibility, keeping the Boeing EUS, is to put atop it a third stage consisting of the 50-ton Centaur V, as discussed in, "Possibilities for a single launch architecture of the Artemis missions, Page 2: using the Boeing Exploration Upper Stage."

 However,  I’m still not convinced the Boeing EUS is the best way to go because of its expense and its small size. If a cryogenic upper stage at a propellant load of ca. 200 ton size instead of the Boeing EUS ca. 125 tons were used, then this larger upper stage itself could do TLI burn carrying the Orion, larger SM, and ca. 15 tons lunar lander.

 I discuss here how such a larger cryogenic stage could be done in a much cheaper fashion than the Boeing EUS approach:  

Why does the Boeing Exploration Upper Stage(EUS) cost so much?  

http://exoscientist.blogspot.com/2022/11/why-does-boeing-exploration-upper.html

 I call this the trickiest aspect of the plan because Boeing has shown repeated delays in getting the core stage ready so the same might happen with their EUS stage, especially when it would have to be moved up to be ready by the 2025 Artemis III launch date, instead of on Artemis IV in 2028. As for the extra Centaur V third stage, since it is expected to first launch this year, likely it will have several launches under its belt by a 2025 Artemis III launch date.

 Note though a MAJOR reason why the development of the different versions of the SLS was arranged as it was was because of cost reasons. The development of the Boeing EUS was pushed back to delay paying for its wildly overpriced development costs. Note too not having to pay for the SpaceX Starship lander would save NASA $3 billion.

 Boeing’s charge to NASA for the EUS is a key reason why I prefer the simpler approach for an upper stage of just basing it on the core with fewer barrel rings. 

This would also give us the stage more cheaply and more quickly since it involves just using fewer rings on the tanks. If you have ever watched the video of construction taking place at the SpaceX development site, barrel rings of the tanks on the Starship and SuperHeavy are swapped out, replaced, taken-off and put back routinely.

 You’ve heard the mantra of former NASA administrator Dan Goldin, “faster, better, cheaper”? This would be "faster, better, cheaper, and simpler".

 Since this is of different design though that also brings into question its availability by a 2025 launch date. Note that this plan involves several new components. For that reason, we might want to use Artemis III as an unmanned test lander mission. We might even have it be “manned” by human-like robots, with their operation controlled from the ground on Earth.

 In this regard it is notable that several lines of evidence suggest that there might be valuable metals at the lunar South Pole, the planned location for the Artemis III landing. Indeed, the untold trillions of dollars of valuable metals speculated to exist in the main-asteroid belt on 16 Psyche might already exist just next door at the Moon's south pole! I discuss this here:

U.S. will lag behind in utilization of resources on the Moon.

http://exoscientist.blogspot.com/2023/08/us-will-lag-behind-in-utilization-of.html

 By the way, the title there stems from my dismay that the U.S. rovers to the South Polar location won’t have instruments for detecting heavy metals but the rovers from other countries will. This is such an obvious thing to include, especially when other countries will include them, that it’s mystifying why the U.S. chose not to include them.

 In any case, it would be pretty cool seeing human-like robotic astronauts prospecting for valuable metals at their landing site.

The ESA produced lunar lander I suggest using is of Apollo-like size at ca. 15 tons. But its crew module volume would be much larger as it is based on the Cygnus capsule, given life support. The Apollo lunar lander had a 6.7 cubic meter internal volume. But the Cygnus has an 18.9 internal volume, nearly 3 times that of the Apollo LEM and the expanded version of the Cygnus has a 27 cubic meter internal volume.

 But as for sending cargo or habitats to the Moon, the SLS is far too expensive, $2 billion+ per launch, and at too low flight cadence, at best 1 once per year, for that purpose. 

 Better to use lower cost launchers such as the Falcon Heavy for the purpose. I estimate using all hydrolox in-space stages, given low-boiloff tech, the FH could get 15 tons one-way to the lunar surface. 

 Using hydrolox only for the TLI burn but a storable propellant lander stage (so no low-boiloff tech needed), the FH could get 10 tons to the lunar surface.

Robert Clark