Framing the Discussion
There is a great quote, apocryphally attributed to British Prime Minister Winston Churchill that goes:
Americans can always be trusted to do the right thing, once all other possibilities have been exhausted.
No historian has confirmed this Churchill quote from the famous quipper, but it is certain that this is a truism when it comes to our country’s efforts in space. The singular distinction in American space policy regarding expansion beyond low earth orbit since the demise of the Apollo program is one of bad decisions interspersed with indecision. Another quote, this one from Johann Wolfgang von Goethe is appropriate here.
Each indecision brings its own delays and days are lost lamenting over lost days. What you can do or think you can do, begin it. For boldness has magic, power, and genius in it.
Indecision and a history of not doing the right thing when presented with opportunity is illustrated in the following chart, from a NASA briefing given October 30, 2019 by Marshall Smith, the Director of NASA’s Human Exploration Program.
This is really not a comprehensive list and indeed gives a mistaken impression that these architectures are focused on getting to Mars, which at the end of the day, is part of our problem, one of the insane and divisive debate on whether to return to the Moon or “turn and burn” to Mars. This bias in NASA has put more weight on a presidential tweet than on multiple presidential space policy directives and budgets. On February 19, 2020 Vice President Mike Pence gave a remarkable speech at NASA Langley. The key portion of that speech is here:
“The President has directed NASA and the Administrator to accomplish our goal to return to the Moon and then on to Mars not only within five years, but let me be clear: The President has made it clear that we’re going to accomplish this goal by any means necessary. In order to succeed, we are going to continue to focus on the mission over the means. We want to challenge each one of you here at Langley: Consider every available option and platform to meet our goals, including industry, government, and the entire American space enterprise. It’s the reason why we’re cutting out the underbrush of needless regulations and barriers to innovation, because we want you all to be able to reach, to engage, and to draw on the best ideas in America to get us where we’re going by the time we set ourselves to get there. Our administration is absolutely committed to this goal and we want you all to have the same determination and resolve to get there. And this President and this administration and the American people are committed to achieving this goal through NASA and through the Langley Research Center.
So let me at least give you one word of admonition on behalf of your President and on behalf of the entire National Space Council: More than ever before, we want you to engage your imaginations, your creativity. Challenge one another. You know, there’s that old proverb that says, “Iron sharpens iron.” So I encourage you to come in every day with that same impatience and energy that, frankly, I heard in the voices of everybody that Director Turner introduced me to today. The enthusiasm as we walked through the Center, the fire in their eyes — just let that be in your eyes.”
A great speech, but as always, it is the follow through that counts. Thirty years ago as a student I stood up for my first public speaking to castigate NASA’s management at the time to quit talking about going back to the Moon and Mars and do it. That is still the point, but at the end of the day, why are we going and what are we going to do when we get there?
NASA’s Current Rationale?
This is what the Vice President said last year, in April, regarding the Moon.
Here is the language from the president’s Executive Order, SPD-1 from 2017.
In the same NASA presentation listed above, this is what the agency says about going to the Moon.
The above is a pale shadow of what the Vice President has said in his speeches and the intent of Space Policy Directive #1. Finally, from the NASA 2021 budget roll out, is the plan through 2030 for human launches and other missions to the Moon (link).
This is the current NASA plan for missions, human and robotic, to the Moon through 2030. There is already a caveat to this that did not exist in previous versions for the forward 2024 landing with the addition of a large cargo scale cargo lander that will carry the Lunar Terrain Vehicle (LTV). This inclusion is shown in the next chart from the Same NASA presentation.
The new inclusion of a rover is a great step, to be applauded. This would be delivered on a large scale cargo lander. At this time there are only two possibilities for this, the five metric ton lander (payload), by Blue Origin (link) and the SpaceX Starship (link). The most interesting thing to me about this is that the mass of the rover (My company is on one of the rover teams), is not expected to be over 500 kilograms (the NASA Apollo lunar rover was 209 kg), leaving several thousand (over 4,000 kg minimum on a Blue Moon) kg of payload for other things. This is encouraging, and hopefully represents a work in progress.
However, there is a systemic issue with the current plan, and that is the amount of time spent on the surface. In the current plan, the stay time on the surface approximately 6 days or 144 hours. For comparison, the Apollo 17 mission, the last of that series, lasted a bit over three days, or 74 hours, slightly over half the time of an Artemis mission. The most important aspect here is the cost. The Apollo program cost, which really and truly was an effort on the scale of the Egyptians building the Great Pyramid 4,500 years ago is not a good comparison. Today, a cost per minute is an interesting metric to examine, the cost per minute of operations on the surface, holding the overall mission time and cost invariant.
Not including the cost of development, a conservative estimate is about $2-3 billion dollars per launch for the SLS/Orion/Artemis mission. For a 144 hour stay on the surface, or about 8,640 minutes of time on the surface, that works out to a cost of about $231,481-$347,222 per minute of time on the surface. With four missions between 2024 and 2028 that is only 566 hours on the surface, for $8-12 billion dollars, and that is a very low end estimate. With the overhead of the human spaceflight beyond Low Earth Orbit (LEO), (not including ISS operations), we are looking at $20-25 billion dollars in cost for those four years, forgetting development costs.
The thought experiment here is, how can we make the stay time more cost effective?; by obviously increasing the number of minutes on the surface. By increasing the number of minutes on the surface per mission we multiply the effectiveness of the crew, and what can be done there in terms of exploration, in terms of building an operational capability on the surface, and experimenting with In Situ resources. Remember that these are the goals, as laid out by the Space Policy Directive #1. That is what I want to explore here. We have to do this. We have to make this more cost effective, well before 2028 because even then there is no plan for long term surface operations as of yet, and thus every minute more that can be spent on the surface before 2028, the higher the leverage is and the lower the cost to the American and European taxpayers. The goal is for lunar development on the Moon make getting to Mars less risky, more cost effective, and begin the move of humanity in to the great economic enterprise of utilizing the resources of the solar system for the betterment of humanity.
Details of the Current Plan
Figure 6 here shows an overlay of the timing of each mission segment of the Artemis missions on a NASA chart from the October 2019 presentation by Marshall Smith.
The timeline above comes from a paper [Condon, et al, 2020] which gives the details of the four Artemis sortie missions between 2024-27. There are some variations in this timeline due to exact arrival and departure time to and from the Gateway to and from the Moon and to and from the Earth, but the numbers in figure 6 reliably bracket current NASA mission planning.
The first thing that is striking is the number of hours in space vs the number of hours on the lunar surface. Leaving out the Earth transit time, the number of hours from Gateway arrival outbound and Gateway departure inbound, you have a total time of 419.6 to 477.6 hours in the vicinity of the Moon. However, out of that you only have 134-144 hours on the ground. Table 1 Gives a summary in hours, minutes and dollars per minute for the missions.
To paraphrase something that famous NASA astro-engineer John Hubolt once said; “Do we want to spend time on the Moon, or not?” It would seem that it should be our goal to raise the number of hours, if possible, at little or no increase in cost. The most obvious is to increase the stay time, but there is a problem with that in the current NRHO Gateway orbit. The current Gateway 9:2 resonant NRHO orbit has a period of ~6.56 days. This creates a an intrinsic time constraint regarding surface stay times. The Human Landing System (HLS), can only leave the Gateway orbit to transit to the surface near periapsis on the inbound leg of the 70,000 km apoapsis. The HLS vehicle can only leave the surface when the Gateway has passed periapsis and is headed outbound toward the 70,000 km apoapsis. This is shown in figure 7, which is an annotated inset from figure 6.
This orbital geometry creates many timing constraints. Coming from the Earth to the Gateway is like leaving the Moon in that you catch it on its outbound leg from periapsis. This minimizes the time spent in Orion as crew consumables are a major constraint. This means that the crew has to spend 120-144 hours in the Gateway prior to departure to the surface, which obviously is not time spent on the surface. Then, as shown in figure 6, the total time to get to the surface is about 22.8 hours. This creates a crew fatigue level concern, as it would probably be pretty hard to sleep during that time. Coming back from the Moon to the point of departure for Trans Earth Injection (TEI), has the same issue.
As shown in table 1 its about 22.8 hours from the surface back to the Gateway, and then about the same 12-144 hours until TEI departure. Looking at Table 1, that is between 77-78% of a lunar mission, not on the Moon, but in space, not even counting the transit from the Earth to Moon and back. If NASA Just has to have their gateway, why can’t we put it in Low Lunar Orbit (LLO)?
Gateway in Low Lunar Orbit
If NASA just simply has to have a Gateway, can it be put into LLO? Another figure from the October 2019 presentation by Marshall Smith gives an inkling of the thinking from that corner. This is figure 8.
The orbits to the left are those that were/are under consideration. The current orbit the Near Rectilinear Halo Orbit (NRHO) as a family is shown at the lower left. A feature of the NRHO orbits is low delta V to get into them from the Earth, but with a very large semi major axis (distance) from the Moon. This creates the problem just described. The family above that which are the Distant Retrograde Orbits (DRO) are also low dV and large semi major axis orbits and have the same benefits and issues regarding lunar surface access. Now the third is Low Lunar Orbit (LLO), but as shown above, is an orbit that requires a lot of delta V to stay there. Or does it?
Frozen LLO Orbits
When a spacecraft is in LLO there are three major influences on it in descending order. For very low orbits the mass discontinuities (called Mascons) perturb its orbit. This was found during the Lunar Orbiter missions in 1966-67 and Apollo missions. Since these were short missions it was easy to compensate for those perturbations. As the orbit gets higher, above 500 km, the influence of the Earth and its gravity field starts to really influence the shape of the orbit and thus orbital lifetime. Finally, the Sun also has a significant influence on spacecraft orbits with its gravity. However, there are a class of low lunar orbits, called “frozen orbits” that have been found over the years since we have grown familiar with operating in LLO with the Apollo, Lunar Orbiter, Lunar Prospector, and Clementine missions. Orbital simulations were carried out by multiple NASA astro-dynamic experts [Ely, 2005] and [Folta & Quinn, 2006], to determine the class of frozen orbits that would make operating a spacecraft or a constellation of them in LLO feasible with minimal propellant use for station keeping.
The definition of a frozen orbit is (from Folta and Quinn)
A more general definition of the term frozen orbit could be applied to any orbit which seeks to keep one or more of its orbital parameters constant.
Different classes of frozen orbits around the Earth are the Molniya, (12 hour high eccentricity, high inclination) and the Tundra, ( 24 hour high inclination, high eccentricity) both used by Russian spacecraft. In lunar orbit the DRO, NRHO, as well as LLO frozen orbits fall into this classification. The unique thing about a LLO frozen orbit is that they are possible for any inclination greater than 39.3 degrees, depending on eccentricity and other orbital parameters. Suffice to say, that for a polar orbit of 90 degrees there are frozen polar orbits where the requirement for orbit maintenance is basically zero in delta V terms. This is shown in figure 9 for 90 degrees from the Folta and Quinn paper.
The above charts show that the periapsis and apoapsis varies only within certain secular limits. The chart to the right shows this same variation in the form of a polar orbit plot. With the circle closed, it means that the orbit variance you see on the left is encompassed within the blue area on the right. This is an exceptionally important finding that is not just theoretical. The NASA Lunar Reconnaissance Orbiter (LRO), has operated in this frozen orbit with zero delta V since 2014, and I have been told that the orbit is stable out past 2040 [Mersarch, 2020 personal communication]. Lunar Prospector in 1998 also followed a modified form of this orbit for its mission.
Thus I think that we can definitively put to bed the notion that LLO orbits are unstable and unusable as staging orbits for lunar missions.
What Gateway in LLO Brings to Exploration
Leaving aside for a moment the issue of Orion’s Service Propulsion System (SPS), here are some of the advantages that the Gateway in LLO brings to lunar exploration and development. Following is Table 2, which recasts Table 1 to show how the time on the surface changes with Gateway in LLO.
The difference is dramatic. Surface stay time increases from 134 hours to ~369.5 hours for the minimum cast to ~427.5 hours for the max case. that is 2.76 to 3.19 times as much time on the surface, bringing the total hours for Artemis III, IV, V, and VI to between 1478 to 1710 hours on the surface. That would triple the exploration productivity for either the same or even less cost. Also, since we hold cost constant, then the cost per minute drops by a factor of ~3 and well. Why would we not do this? There are caveats to this in the current mission constraints that must be addressed. However, if this can be pulled off, it would drop the cost per minute on the Moon dramatically as well as provide the time for much more surface productivity.
Surface Stay Time Limits
Currently the specification for the Human Landing System (HLS) is 8 days or 192 hours. This is driven by the consumables that can be carried on the putative HLS design. However, NASA is already including a heavy cargo lander with the first mission, and if that was a Blue Origin lander, this could easily provide 20-30 kilowatts of power and other supplies to enable the much longer surface time. Purposefully, I did not change the overall mission time, but there is no reason that this can’t happen as we have reduced the amount of time on the Gateway, and thus the consumable resources there as well as the ability to create a supply chain that could allow the crew to remain on the surface for far longer than current plans or even this thought experiment.
Human Landing System Changes
Figure 10 shows the current quasi-baseline for the Artemis mission scenario.
The word nominal is used here as there is considerable flexibility to this architecture. The HLS procurement document indicate that this can either be a two or three vehicle option. The “Dream Team” of Blue Origin, Lockheed Martin, Northrup Grumman, and Draper labs, have delivered to NASA (According to an article by Are Technica [link]) a proposal for a three element lander with the Blue Origin Blue Moon lander as the Decent Element, Lockheed Martin with the Ascent Element, and Northrup Grumman with the Transfer Element. Below is NASA’s full architecture for the three element configuration in figure 11 which is the one that the Dream Team has proposed against.
The complex operational scenario with three heavy commercial launches of these elements to the Gateway, where presumably they dock together, then as a unit proceed to the Moon, is necessary as none of the launchers (except maybe SpaceX Starship), can get more than 15 tons to the NRHO orbit, then to the surface. The mission scenario as shown in figure 10 has the Transfer Element doing the burn from NRHO to LLO where it is then either expended, or later transfer back to the Gateway. If the Gateway is in LLO, this is simplified, and the Transfer Element is redundant and can be discarded from the architecture. This is true as it can be seen from the October 2019 NASA presentation that the delta V is actually slightly less from LLO than from the NRHO orbit. This is seen in figure 12.
NASA has explicitly stated in their HLS procurement documents that an LLO rendezvous is acceptable for the first missions. This is exceptionally curious in that we still have the problem that Orion cannot get into and out of LLO with its current European provided Service Propulsion System. A possible explanation for this is the Boeing proposal for the Human Landing System. Boeing has proposed a two segment (A Descent Element with an Ascent Element) that would launch on an SLS 1B vehicle (link). This would, according to a Wikipedia article (link) eliminate the need for the Gateway entirely as the system would rendezvous and dock with Orion launched on an SLS Block 1 vehicle, and presumably, (as did the Constellation program), with the lander doing the Lunar Orbit Insertion (LOI) burn.
While this is an interesting architecture that eliminates the Gateway entirely, it also eliminates the competition by rendering moot the commercial launchers and the contractors building the Gateway. It is unlikely that physically Boeing could build all of these systems by 2024, even with congressional budget support. A two SLS launch would raise the cost of a single lunar mission to well over $4 billion dollars. It does appear that Boeing has been doing its lobbying work in the House NASA authorization that appears to favor this approach, but beyond the political aspect, it is unlikely that Boeing, with its current track record, could pull this off. Far better to keep as much of the Gateway and commercial approach, putting it instead into LLO as to make the job more manageable and spread the corporate risk.
Gateway in LLO
It is entirely possible to get the Gateway into LLO. Figure 13 is from a presentation by Maxar (the Power and Propulsion Element contractor) at the NASA FISO telecom.
It appears that in step five, that the Maxar PPE goes into a medium lunar orbit then spirals out to the NRHO orbit. After aggregating the other elements at the NRHO orbit for the minimal Gateway, it should have enough fuel to transit to an LLO frozen orbit. After reaching there, it would have no further need for the propulsion system and the solar arrays and power system would be available to support Gateway operations.
What Would the New Architecture Bring?
Having the Gateway in LLO brings many benefits.
- Increased surface operational tempo; ~3X the time on the surface
- Reduced complexity; elimination of the Transfer Element
- Lower Cost; Commercial launchers retained for Ascent and Descent Elements
- Operations; Gateway in LLO enables future propellant depot
- Crew Safety; an 82% reduction in the in-space minutes for radiation risk
- Crew Operations; eliminates the long day for NRHO to surface transfer
- Dramatically increases the mass of returnable samples to the Earth
- Requires the EUS for 2024 with the ULA Integrated Vehicles Fluids for Orion
- Requires relay satellites; Not a big deal with commercial providers
We can only begin to imagine the operational benefits. The stay time can be carefully crafted to be during the sunlit time at the polar landing sites. Also, rather than multiple sorties that go to different areas, infrastructure can be built up. One of the big problems that has become apparent is that of the dust kicked up by the multiple landings that could hang in LLO to damage the Gateway or any other vehicle there. A lightweight landing pad, sent with the cargo lander that carries the Lunar Terrain Vehicle (LTV) and other supplies, could also carry the parts for a landing pad. This would dramatically reduce or eliminate the dust problem.
With a longer stay time, we could work on increasing the stay time as we land cargo landers and build up capability. Especially in the south polar area of the Moon the driving area is quite interesting though limited in comparison with the north. The opportunities to build up In Situ Resource Utilization of the water and metal oxide resources technology, experiments, and operations. It triples taxpayer dollar value per mission.
One factor that is not talked about much with the Gateway architecture is the paucity of samples that can be returned per flight, only 26 kilograms, worse than Apollo 12! This is shown in figure 14 from the Human Landing System procurement documents.
Having the Gateway in LLO should allow for a lot more sample mass to be brought up and transshipped back to the Earth. It may even open a pathway for commercial cislunar transportation. This will certainly be the case should starship become operational. Even if that does not happen, having an EUS with the ULA Integrated Vehicle Fluids located nearby and or attached to the Gateway creates the volume for a propellant depot. There are just so many advantages of having the Gateway in LLO versus the NRHO orbit. This is what Vice President Mike Pence meant in his statement at the NASA Marshall Spaceflight Center in March of 2019. This is what we should be focusing on, and working toward. You want to go to Mars? Get with the lunar program.