The Space Reviewin association with SpaceNews

ISDC 2024

Gateway chart
A chart from a NASA presentation shows a design for a Gateway far more complex than what the agency has been publicizing, taking into account international contributions. (credit: NASA)


Bookmark and Share

In my previous article (see “Does the Gateway make sense?”, The Space Review, May 29, 2018), I reviewed the history of how we got to the point where the LOP-G station is our next step and destination in human spaceflight (see “Does the Gateway make sense?”, The Space Review, May 29, 2018). I somewhat reluctantly agreed that it is the logical next step. That was written before I saw on YouTube the Marshall Spaceflight Center Roadshow from May 24. NASA has been meeting with the space agencies of the International Space Exploration Cooperation Group (ISECG). The image above is from that presentation and is the latest proposed configuration of the LOP-G station, which includes international partners.

Until recently NASA was showing concepts similar to the model displayed last summer at the EAA AirVenture in Oshkosh in the following image:


My first thought after seeing the changes was the old saying that a camel is a horse designed by committee. My second thought was this is going to cost twice as much and take three to four years longer than the prior concept, but also that getting back to the surface of the Moon is not a high priority for these people. Political considerations, including a lack of trust between partners, are far outweighing technical considerations. If we want to get back to the surface of the Moon using a gateway station sooner rather than later, this is not the way to do it.

A panelist at the National Space Council meeting in the White House June 18 said the purpose of international partners is to “cancellation-proof” the project, and this does not save money. It is time for an attitude change.

If you look at the new configuration, several changes have been made to spread the work around. Originally the Power Propulsion Element (PPE) was supposed to provide power, propulsion and communications with Earth. The new PPE is smaller and has the telecommunications capabilities reduced. A new international-supplied module called Esprit supplies the communications, has a science airlock, and additional propellent storage for the PPE. The third element now on the configuration is the Utilization Element, which is a small pressurized volume for “additional habitation” provided by the US. The fourth and fifth elements inline with the first three are habitation modules, with one provided by the US and the other provided by international partners. The sixth element inline is an airlock module. In addition, the diagram shows two logistics modules attached, one of which has a robot arm on it. There is also a medium-sized robotic lander shown attached to the station.

One issue I have with the new design is that there are probably going to be two or three habitation modules of different designs. No one can convince me that this will cost anywhere close to what one design, optimized in size for a minimal number of launches, will cost. Interconnecting more but smaller modules will be an inefficient use of volume versus mass. Another issue I have is with separating the PPE into two modules: it adds complexity, cost, and time to ensure that they work together. It is not efficiently using limited budgets. Build one design to save money. A panelist at the National Space Council meeting in the White House June 18 said the purpose of international partners is to “cancellation-proof” the project, and this does not save money. It is time for an attitude change.

Many years ago, a friend of mine told me a story that seems to apply to what is going on. He owns a small company that was developing software for scheduling machine tools on the factory floor. If anyone knows the challenges of efficiently using machine tools and raw materials, they know that scheduling can be a very complicated challenge. They were designing the software from the starting point and going through, step by step, until they were to have schedules and nests that would give them the high productivity results they wanted. They were struggling for a long time to get it right. If they optimized for efficient material usage they were getting schedules that were missing delivery dates. If they optimized for hitting delivery dates, they were getting horrible material yields.

My friend’s company at the time was selling Unix computer systems from Sun Microsystems. He was at a golf outing put on by Sun when he started explaining his problem to his golf partner he had been assigned to. His partner said he was looking at the problem all wrong: he should start with the output part, focusing on the end results he wanted, and work backwards to the beginning. When he got back to the office he looked at it from that perspective and he and his team quickly resolved their issues. They now have a highly capable scheduling system that does a great job of producing schedules and part nestings for machine tools that hits most delivery dates and efficiently uses material and is running in hundreds of factories around the world. My friend’s golf partner that day was Bill Joy, one of the founders of Sun Microsystems and the creator of the Java programming language. It was great advice.

This is what I think NASA and the international partners are doing wrong on this project. They are not looking at an outcome and trying figure out the best way to get there. They are looking at just the next step and trying to figure out how to fit it within budgets and satisfy political whims of several governments to cancellation-proof the project. This comes from a lack of trust that partners will follow through on commitments if they don’t build multinational dependencies for practically every major component and step of the process. Partners could bail out leaving a mess for the rest.

If we start with an outcome that the LOP-G station is to support and work backwards from that, the results will probably be much better. The outcome I think they should be looking at is an outpost on the Moon (a Moon Village, if you want to call it that) that has a basic mission to explore, support research, and potentially offer commercial opportunities. The kind of research space agencies want to accomplish at LOP-G should also be considered. If this outpost could be defined within a reasonable set of size and capabilities, we would have the end point that we want to design to achieve. The path being taken by adding step-by-step capabilities without a goal in mind is why the Orion service module is undersized, requiring LOP-G to be in the near rectilinear halo orbit (NRHO) Orion can reach and return from instead of instead of an Earth-Moon L-2 halo orbit that would allow anytime flights by landers to and from any point on the surface of the Moon. In addition, the L-2 halo orbit is the ideal spot for LOP-G to eventually support crewed and robotic missions to the rest of the solar system.

For various reasons beyond the scope of this article, the Moon’s poles are prime locations for an initial human outpost. We have some basic idea of the topography and resources on much of the lunar surface from spacecraft like the Lunar CRater Observing and Sensing Satellite (LCROSS) mission and the Lunar Reconnaissance Orbiter (LRO). Now is the time to start refining our knowledge including below the surface so we can find the best locations for human landings and the first outpost.

NASA has asked industry for proposals for various sized landers in a public-private partnership context called Commercial Lunar Payload Services (CLPS), where NASA would pay for flights but not fully fund those vehicles’ development. The slow-motion, never fully funded Resource Prospector mission was recently canceled. NASA now wants to send some of the instruments developed for the mission to the Moon on these landers. For the larger landers, I don’t like the idea of instruments stuck in one spot. We need rovers. This is where our international partners could step up and develop rovers to more thoroughly investigate areas within several kilometers of landing sites.

Just like smaller landers could evolve into large crewed landers, small robotic rovers could evolve to a variety of larger rovers for exploration, construction and infrastructure maintenance. Any companies that took part could be positioning themselves for potentially larger business opportunities in the future.

The point is that we should be designing both LOP-G and the lunar surface outpost simultaneously.

NASA and the ISECG space agencies need to decide what they want an initial lunar outpost to be capable of. They need to decide how much each one of them wants to spend on it. The missions I see the outpost supporting starts with exploring and understanding the area around the outpost. This starts giving us two immediate complimentary returns on the investment. It will be a lunar science bonanza, especially in analyzing material from the cold traps of deep craters permanently shadowed from the Sun. There should be great evidence regarding the history of the formation of the inner solar system. The second return is that we’ll learn what resources are available for use to support further exploration and commercial opportunities.

The point is that we should be designing both LOP-G and the lunar surface outpost simultaneously. Starting with the outpost, we need to decide how many people will it initially house. We need to decide what these people will be doing. Considerations should include what kind and size of habitat will be needed; how much power needs to be available; what kind of tools, life support systems, rovers, regolith movers, core drills, and so on are needed; and how much science needs to be done on site and how much sample mass needs to be returned to Earth. The maximum size of individually delivered payloads needs to be determined to set requirements for landers supporting the outpost

From here we can figure out the requirements for LOP-G to support the lunar outpost. We’ll know what we need in terms of landers, rockets, space tugs, crewed spacecraft, and more. This way they can be designed to fit a need instead of scrambling to figure out what we can do with various components that poorly designed without any planning for their end use.

If we split up LOP-G so every partner can play a role in every step, we’ll go slowly and take forever to go anywhere. I fear if we split LOP-G up it won’t be ready to support landing an outpost on the lunar surface until 15 to 20 years from now, and no one will have started designing the components we will need to land and operate on the surface.

The true advantage of world trade is that different countries do different things better and more efficiently than others. By trading what each country does best, we’re all better off with better standards of living than if each country tried to produce everything for themselves. Why not try the same philosophy for space development?

A 2030 design goal

Imagine it is now July of 2030 and a plan with forward looking goals has been executed. Where could we be with a little common sense and planning? The following is a possible future space history from now to 2030.

By 2019 the US and the ISECG agree to a simple LOP-G station and agree to call it the Jules Verne Gateway Station in honor of the great inspiration the author provided generations of space enthusiasts to follow—and because LOP-G is an awkward, uninspiring name. They also agree to start developing a lunar surface outpost to fit a well-defined initial mission. They argue over the name, but eventually settle on Selene Prime. The goal is to have Selene Prime up and running with four astronauts at the facility by 2028. It will slip to July of 2030.

The work is apportioned to accomplish more than the current plan of building LOP-G will deliver. The part of building the Jules Verne Gateway Station will become primarily an American project. It will have a large PPE that will insert itself into the NRHO lunar orbit after being launched on a Falcon Heavy rocket. The PPE is built in a commercial partnership with NASA and as a prototype for future PPEs to be used for other purposes, including exploration of the rest of the solar system. It will arrive in lunar orbit in 2022 and have the capability to support a high-bandwidth communications with Earth and the rest of cislunar space.

The first in a series of landers will start arriving on the Moon with a basic suite of instruments aboard and a panoramic mast camera to start scouting possible outpost locations in late 2021. By 2025 a dozen of these small landers will have covered promising sites at both poles and a few other interesting sites elsewhere on the lunar surface for science and exploration.

By 2020, specs for payload delivery to the lunar surface with intermediate sized landers will be developed. Blue Origin, with a first version of their Blue Moon lander, gets a contract to develop a lander that can place a 2,500-kilogram payload on the surface. Masten Space Systems gets a contract for a smaller lander. ESA goes ahead and starts designing a series of rovers with commercial partners to explore the lunar surface in much greater detail that weigh in around 2000 kilograms. Mitsubishi designs a 750-kilogram rover in partnership with JAXA, with three going to the lunar surface on a Blue Moon lander. Both rovers have instruments and tools aboard from several partners including Australia, Israel, the UAE, Korea, India, and the US.

In 2023, SLS/Orion delivers the habitat module to permanently dock with the PPE waiting in lunar orbit. A crew of four—two Americans, a European, and a Russian—installs and checks out the habitat module. They stay for 30 days, giving the Jules Verne Gateway Station a shakedown test of its systems.

In 2025, an SLS/Orion flight arrives at Jules Verne Gateway with a US-built science module. A crew of four—two Americans, a Canadian, and a Japanese astronaut—stay for 40 days. The trip also brings a Canadian-built roboict arm that can maneuver all around the Jules Verne Gateway Station. It handles berthing of cargo ships, small landers, unloading of unpressurized cargo, and external maintenance, among other tasks.

In 2026, an American built airlock module is delivered and mated to the Jules Verne Gateway Station. The crew stays for three months testing systems. The crew does the first test of controlling exploration rovers on the lunar surface.

Blue Origin develops a larger version of their Blue Moon lander capable of carrying a crewed ascent module to the surface. The descent stage is ready by 2026 and can place 10,000 kilograms on the lunar surface. In late 2026, the first test landing, at the chosen site of the Selene Prime outpost, carries a Japanese-built solar power station. After landing, the solar array tower, permanently attached to the spent descent stage, deploys vertically and starts to rotate following the sun.

After 2030, options are wide open to do creative things because a foundational infrastructure will be in place for cislunar space.

A decision to cancel the SLS after eight missions negates the need to develop Advanced Solid Rocket Motors required for SLS Block 2. Lockheed Martin works with Blue Origin to launch future Orion missions on the New Glenn rocket. The second stage gets modified to be refuellable in orbit and to work as the service module for Orion. The New Glenn second stage now gives Orion the delta-v to reach any orbit around the Moon and return to Earth. This combination of Orion on New Glenn gets a first test flight in 2028 in low Earth orbit. An operational fight to the Jules Verne Gateway Station follows later in the year.

Northrup Grumman Innovation Systems lands a contract to deliver cargo to the Jules Verne station on upgraded Cygnus space vehicles that can deliver three tons of supplies and experiments per flight. The first one flies in 2027 on ULA’s Vulcan/ACES rocket. ULA develops an ACES tanker that will refuel lander ascent and descent stages near the gateway station.

Lockheed and Blue Origin cooperate to design and build the crewed reusable ascent module capable of carrying four astronauts, based on the systems designed for the Orion capsule. Lockheed has proposed using the flight controls, life support system, power system, communications, and more for a Mars lander. The first mission lands in 2027 with an American commander, a Russian, and two Europeans aboard to precisely survey and prepare the location where the first habitat module will set down. A target will be placed for precision landing of the first habitat on a Blue Moon lander.

In 2028, a Blue Moon descent stage with the first habitat module for the Selene Prime outpost, funded by ESA and built primarily by Thales Alenia Space, is precision-landed on a wide crater ridge at the South Pole of the Moon precisely on the target placed by the first crew. After successfully landing, a second mission arrives to connect the solar power system to the habitat, which will stay on top of the spent descent stage. This crew has an American, a Canadian, a Japanese, and a European on board. Their mission is to connect power cables between the habitat and the solar power station. In addition, they setup and precisely aim communications antennas towards Earth. The crew stays seven days and uses a small backup airlock to enter the outpost and test systems.

In 2029 a Russian built airlock with an electrostatic dust cleaning porch for Selene Prime is landed on Blue Moon lander near the outpost. A mission with a crew of two Americans and two Russians lands to hoist the airlock off the lander and move it to and attach it to the habitat. After installing it, they test out the airlock including carrying in cargo from an intermediate size lander sitting nearby.

Boeing works with ULA to put an upgraded CST-100 spacecraft on a Vulcan rocket with an ACES upper stage modified to work as a service module that can take Orion to any cislunar destination and return with enough delta-v capability with margin to spare. This combination is ready by 2030 to give dissimilar redundancy for human flights to cislunar space and back with Orion on New Glenn.

In early 2030, a Thales Alenia-built science module is added to the Selene Prime outpost. It is outfitted with science instruments from all the project partners. A crew lands shortly afterwards and attaches it to the outpost core habitat and outfits it for use.

In 2030, the Jules Verne Gateway station PPE is refueled and moves the station from the NRHO orbit to Earth Moon L-2 halo orbit because both Orion and CST-100 can now service the location. Landers from this orbit can land anywhere on the Moon at any time. In addition to supporting surface missions, the station is now in an ideal position to support crewed and robotic missions to the rest of the solar system.

In spring of 2030, an expandable Bigelow module is added to the Jules Verne Gateway Station as a commercial module with space inside leased back to space agencies and commercial customers. In July 2030, both the Jules Verne Gateway Station and the Selene Prime outpost are declared operational. A regular rotation of crews is started. A lander arrives at Selene Prime in July 2030 with a crew of four. Every three months going forward a lander comes down and rotates out two crew members. Cargo landers will bring supplies, rovers, tools and anything else the people on the surface need.

Beyond 2030

After 2030, options are wide open to do creative things because a foundational infrastructure will be in place for cislunar space. Anyone wanting to mine the Moon, for water or anything else, won’t have to develop their own rocket and landers. They can buy the transportation services from the companies that developed it for NASA and the international partners. The same is true if some company wants to put a factory, a mine, or a hotel on the Moon or anywhere else where rockets and spacecraft are flying. This will be the tipping point where commercial space beyond low Earth orbit can start growing organically. If it becomes practical to produce propellants on the surface from local resources a fully reusable lander becomes much more viable if it can be fueled on the lunar surface for return trips.

As a cislunar infrastructure strengthens, smaller or less developed countries may want to start taking part as their industries mature to the point where it makes sense to join in.

The crewed version of BFS will need systems that have to operate reliably for years. That is why I don’t think the crewed version will be ready before 2030. But, once it is ready, it has the potential to be a serious gamechanger.

The complicating factor in all this is China. I suspect a couple of major disagreements will have to be resolved before the US would be willing to let China take part. If China’s neighbors like Singapore, Japan, Korea, Vietnam, Indonesia, and the Philippines are involved, it might be convince China to negotiate an agreement on the territorial disputes in the South China Sea. They would also have to start respecting global intellectual property laws. If people are constantly on the Moon and in lunar orbit without them, China might decide its worth change their thinking on a few key policies.

During the decade of the 2030s many spectacular achievements could be accomplished with smaller incremental investments because an infrastructure foundation is in place. PPEs could take large probes to places across the solar system and possibly return with samples. A Kilopower reactor could be added to Selene Prime to test before use on Mars. A fuel depot co-orbiting with the Jules Verne Gateway station could be added and supplied with propellants from the lowest bidder. Somebody will think of adding something nobody is considering right now.


I don’t believe SpaceX will have their Big Falcon Rocket (BFR) and Big Falcon Spaceship (BFS) ready in the timeframe they are claiming. Their current claim is that they will be able to launch the cargo version to Mars in 2022 and the crewed version in 2024. I’m not buying it. There will be a learning curve in the design and construction of a much larger rocket. There will be design and build iterations before the cargo version is stable. I’m also predicting that it will take a year or two to refine tanker flights and refueling in orbit after that. I also don’t believe Congress will fund accelerating BFS/BFR development because it would probably take money away from SLS. The timeframe makes BFS/BFR not useful for planning for the Jules Verne Gateway station or the initial support of the Selene Prime lunar surface outpost.

I also think Elon Musk is greatly underestimating the time and money it will take to get a crewed version of BFS flying. He also may not have enough revenue along the way to develop it as fast as he wants. He is counting on revenue from Starlink, his broadband satellite constellation. Revenues may not grow as fast as he expects. He will have competition from other constellations like OneWeb, and terrestrial Internet providers are not standing still.

The crewed version of BFS will need systems that have to operate reliably for years. This is a much bigger challenge than developing the Dragon capsule. That is why I don’t think the crewed version will be ready before 2030. But, once it is ready, it has the potential to be a serious gamechanger.

Prepare for Mars

As many of the components for humans returning to the Moon will be designed with operations to and on Mars in mind. A deep space transport to carry people to and from Mars has been envisioned using a PPE developed and tested on the Gateway station, plus habitat modules tested for durability in deep space. Lunar habitats landers and rovers will be the foundation for the same on Mars, supporting a human landing within a decade after Selene Prime becomes operational.

The wildcard for going to Mars is if BFR/BFS is ready for the task. If it is, I think a landing could happen in my estimation by 2034. If not, 2040 is more likely. I don’t think SpaceX’s Starlink will generate enough money to develop the technology needed to live on Mars by the time BFS is ready. NASA and the international partners do have the resources. Even if BFS was ready to land on Mars in 2024, the habitats, rovers, and a slew of other required items won’t be. If Selene Prime is operational by 2030, many of the items needed on Mars could be ready in just a few more years.


If the latest design of LOP-G is chosen over the simpler designs that have been previously proposed, I’m against it. It will be a slow journey to nowhere. We might as well throw it out and start over. If we use international partners to accelerate getting to the Moon and Mars, then I’m for it. Doing so requires a change of attitude and methods of cooperation and trust. We need to leverage and develop complimentary capabilities.

I understand the lack of trust driving the latest changes to the LOP-G design. If the US takes responsibility for the Gateway station and transportation to the Moon, international partners must trust that we will deliver. If Europe, Russia, Canada, Japan, and any other countries involved focus on the lunar outpost and a future President or Congress backs out, they will have wasted billions on what will be useless technology.

By picking a more advanced goal of having an outpost on the Moon and a station in orbit supporting it, we will put in place the infrastructure to stay bold and really explore, understand, and settle our solar system.

If we trust each other and each focus on building complimentary technology, we can actually move rapidly to conquer the solar system. We will save money and do more. The hypothetical scenario I laid out in this article or something similar could become real. Conquering the solar system in bold, efficient steps is how we need to move forward. With each international partner focusing on their expertise, we can move faster. ESA director general Jan Wörner wants a Moon village. ESA should focus on that. If they can trust us to focus on the Jules Verne Gateway Station and the transportation infrastructure, the Selene Prime Moon Village could be in place before the bulk of the people who work on it retire. If not, the younger generations who are going to do the bulk of the work will end up like my generation who watched the Apollo landing and expected colonies on the Moon and Mars by now. We were seriously disappointed by the pace of progress in space.

It is time to seize the moment. The technology is rapidly advancing. Interest in space seems to be growing somewhat. Our economies are much bigger than they were in the 1960s. By picking a more advanced goal of having an outpost on the Moon and a station in orbit supporting it, we will put in place the infrastructure to stay bold and really explore, understand, and settle our solar system. This has a chance to be the key moment in history for advancing permanently beyond low Earth orbit. History remembers the bold. It doesn’t remember those who decided to wait. It also remembers those who were smart enough to do it right. I’m ready for this. Are you?

Note: we are temporarily moderating all comments subcommitted to deal with a surge in spam.