Where Should Japan's Satellite Supply Chain Compete Before On-Orbit Data Centers Arrive?

In my previous article, I explained SpaceX's on-orbit data center (hereinafter "on-orbit DC") concept. It is a grand vision, but one that also carries significant risks.

In this article, I will shift the perspective. What does this concept mean for Japan's satellite supply chain? Where do the winning opportunities lie?

I will state the conclusion first: the technology is there. The problem is the organizational structure and speed.

What SpaceX cannot do on its own

First, let's clarify the premise.

SpaceX builds rockets, builds satellites, and operates communication networks. Its capabilities are world-class. However, "cooling massive amounts of computing chips in space" and "mass-producing high-power, lightweight solar cells for space" are things it cannot currently achieve on its own.

I have already mentioned solar cells. SpaceX uses silicon cells for Starlink, but that is a desperate measure to avoid the supply constraints of high-efficiency GaAs cells.

Thermal management is a more fundamental issue. Current Starlink V2 satellites have about 120 square meters of solar panels and use several tens of kW of power generation for communication, propulsion, and control. On the other hand, the estimate SpaceX provided in its FCC application was "100 kW of compute power per ton," which, when converted to a single satellite, implies a scale requiring several tens to over 100 kW of power just for computation*. Moreover, since computing chips release almost all their power as heat, all that heat must be rejected into space. This is an order-of-magnitude requirement that cannot be met by "slightly improving the Starlink design," as it involves rejecting heat equal to or exceeding the total power generation of current Starlink satellites.

* The specifications for the first "AI1" satellite released by SpaceX in June 2026 indicate a compute payload of 120 kW (150 kW peak) and a deployment span of 70m, which is generally consistent with this estimate.

There are three major technical areas specifically required.

• Thermal management systems

• High-power solar cells for space

• Optical Communication Terminals (OCT) for inter-satellite links

These three areas overlap with fields where Japanese private companies possess world-class technology and track records.

Thermal management: A field where Japan has strengths and the biggest technical challenge for on-orbit DCs

I wrote in the previous article that only radiative cooling can be used in space. One of the most difficult challenges when operating computing chips in space is thermal management, specifically the problem of having to mount ISS-level heat dissipation equipment on a single satellite.

Thermal management systems consist of two main elements: "heat transport devices" that carry heat generated by chips to the outside of the satellite, and "radiators (heat dissipation panels)" that release heat into space.

Regarding radiators, major satellite manufacturers, including SpaceX, already have the capability to design and manufacture them in-house, so this is not an area where Japanese companies can claim a distinct advantage at this time. On the other hand, Japanese private companies have competitive strength in heat transport devices due to many years of accumulated research and development.

A "Loop Heat Pipe (LHP)" is a device that uses no electricity and transports heat over long distances through the evaporation and condensation of a working fluid. It has no pumps or moving parts, making it suitable for the harsh environment of space. In October 2018, it was brought to the ISS on the Kounotori 7 (HTV-7) and underwent an on-orbit demonstration experiment for about one month while held by the tip of the Kibo module's robotic arm. Long-term ground operation tests have also been ongoing since 2017, and no degradation has been observed for over seven years.

Another is the "Flat Heat Pipe (FHP)." It has a structure where a check valve is added to an oscillating heat pipe (OHP) and sandwiched between thin aluminum plates; because it does not require a wick (capillary structure), mass production costs are low. It is suitable for spreading heat generated at a heat spot across the entire plate to equalize the temperature, making it effective in situations dealing with localized high heat density, such as directly under a computing chip. It was mounted on the small demonstration satellite SDS-4 in 2012, and stable operation without degradation was confirmed for nearly four years.

Regarding OHPs themselves, engineering researchers at the Institute of Space and Astronautical Science (ISAS) are conducting research as an independent technical lineage, and recently technical manuals have also been self-published. This is a technology that has developed differently from FHPs.

LHPs and FHPs/OHPs are technologies with different applications, and which one to adopt depends on the system design. In the thermal design of on-orbit DCs, each may play a different role.

Furukawa Electric declared its full-scale entry into the space business in 2023, in collaboration with the University of Tokyo, to repurpose technologies cultivated in optical fibers and terrestrial heat dissipation devices for space. In October 2026, it plans to launch its self-developed demonstration satellite, "Funade," to conduct an on-orbit demonstration of heat pipe modules.

Japanese-made solar cells that operated on the moon

In January 2024, JAXA's Smart Lander for Investigating Moon (SLIM) succeeded in a high-precision landing on the lunar surface. The power for this was provided by thin-film triple-junction compound solar cells (IMM3J: Inverted Metamorphic Multi-Junction) developed by Sharp Energy Solutions. They are 0.25mm thick, weigh about 41g each, and have a total output of about 540W. They have achieved one of the world's highest levels of module conversion efficiency at 32.65% at the research level. They are light, thin, and flexible. They have high power-to-weight ratios. These are characteristics that precisely meet the requirements for space solar cells needed by on-orbit DCs.

It is not just Sharp Energy Solutions that has been in the spotlight with SLIM. Organizing the approaches of each company from public information, the situation is as follows, and both the technical lineages and the target markets are diverse.

The fact that they are being successively selected for the Space Strategy Fund conveys the seriousness of both the public and private sectors in this area.

An often overlooked fact: Space solar cells are currently in short supply

Space solar cells are currently in a state of structural shortage.

The conventional mainstream is GaAs-based III-V multi-junction cells. While they are highly efficient and radiation-resistant, they require advanced epitaxial growth technology for manufacturing, making it difficult to expand production capacity. Due to the rapid expansion of constellation demand, "difficulty in procurement, schedule constraints, and high costs" have become the norm.

According to a survey by ESA and NovaSpace ("ESA Technology Market Assessments: understanding the foreseen demand for Solar Generators and Solar Cells", space-economy.esa.int), Starlink alone accounts for 83% of the global demand for space solar cells from 2024 to 2033. Of the remaining demand, the market that Western and Canadian suppliers other than the US can actually enter is only 3.5%. This means that massive constellations like Starlink are effectively self-contained within their own national ecosystems, and there is structurally limited room for external suppliers to enter.

This shortage structure is also an entry opportunity for Japan. Being able to "supply" is becoming the core of competitiveness, not just "having high efficiency."

Optical communication terminals and radiation-hardened semiconductors

In addition to thermal management and solar cells, I will also touch on two adjacent areas.

Regarding inter-satellite Optical Communication Terminals (OCT), NEC has already been selected for the Space Strategy Fund for its technical development of building optical communication satellite constellations and is proceeding with a full-scale entry. NTT and JAXA have also begun joint research on ultra-high-speed optical and wireless communication infrastructure that seamlessly connects space and the ground.

Regarding radiation-hardened semiconductors, there is also one fact worth noting. Renesas Electronics is actively commercializing radiation-hardened ICs for space and defense, and they are used in many satellites. However, this business foundation originated from Intersil, a US company that Renesas acquired in 2017. It is a long-established company for space and defense semiconductors founded in 1950 as "Radiation, Inc." Because it is a technology derived from the US, it is subject to US export controls, but it is in a position where it can continue to be utilized as an accessible technology and product line under the Renesas umbrella.

The biggest challenge: No "space heritage"

I must talk about the harsh reality here.

The space industry has an implicit barrier to entry. It is "space heritage." Space heritage here refers not to technical demonstrations, but to a track record of use in actual missions.

In NASA's TRL (Technology Readiness Level) definition, a technical demonstration in a space environment only reaches TRL 7, and it only reaches TRL 9 (flight proven) after being operated in an actual mission. What space procurement officers are looking for is this TRL 9. No matter how excellent the ground testing or on-orbit demo is, unless it is adopted in an actual mission, it is difficult to lead to adoption in the next mission.

The technical capabilities of Japan's satellite supply chain are genuine. However, the manufacturing system is specialized for 'small-batch, high-quality, one-off' production. There are no mass-production lines capable of supplying tens of thousands of constellation satellites per year. Furthermore, there is a lack of track record in adoption by commercial constellations. When SpaceX's procurement system is established, if Japanese companies are not registered as suppliers, they will be unable to enter the market regardless of their technology.

A strategy can succeed even without on-orbit DCs

As I wrote in the previous article, there are significant realization risks for on-orbit DCs. Starship might be delayed. Regulatory coordination might drag on. Ground-based power issues might be solved in other ways.

That is precisely why strategic design must not be 'on-orbit DC exclusive'.

Thermal management, space-grade solar cells, and optical communication terminals are technologies that will be needed even if on-orbit DCs do not arrive. Starlink has already exceeded 10,000 satellites. Mega-constellations by overseas operators, such as Amazon's Project Kuiper, Eutelsat's OneWeb, and national security constellations, are being deployed one after another. This demand alone requires massive quantities of scarce space-grade solar cells and high-performance thermal management components.

The correct strategic design is as follows.

If mass-production systems and space heritage are established in the base scenario, you will be in the front row when on-orbit DCs approach reality. Conversely, you will not lose out if on-orbit DCs are delayed.

The reason why we must act now

The window of a SpaceX IPO is currently open. There is little time left for Japanese companies to be recognized as suppliers before the procurement system is solidified.

Solar cells that operated on the moon, thermal management technology demonstrated on the ISS, and CIGS and perovskite cells adopted by the Space Strategy Fund. These are genuine achievements. All that remains is the will to accumulate 'space heritage' by being adopted in actual missions, and the speed at which you knock on the door.

Mamochichi @ A Corner of the Space Industry, June 2026

The analysis in this article is based on the SpaceX prospectus (filed with EDINET, May 2026), FCC application documents (January 2026), ESA/NovaSpace "ESA Technology Market Assessments: understanding the foreseen demand for Solar Generators and Solar Cells" (space-economy.esa.int), and publicly available information from each company. These are personal views and not the views of the organization I belong to.