20 of >500
SDA optical terminals delivered by Jan 2025
Tranche 1 Transport (GAO); one of four qualified suppliers since acquired
Setcoin Intelligence · Space Infrastructure · Supply Chain · Updated October 2026
Space Supply Chain Bottlenecks: Components, Minerals and Capacity
Which parts cap how fast the space economy can grow, who supplies them, and where must capacity be built? Fifty-nine chapters on eighteen critical parts, the minerals beneath them, factories, test, launch, workforce, export control and defence demand, built on filings, government data and Setcoin's own satellite-catalogue analysis.
Open to everyone: the first 3 chapters in full, with exhibits and models; a free account adds their Excel models. Space subscription: all 62 chapters, every model and Excel file, 7-day free trial, then €199 a month. Already a reader? Sign in.
- Chapters
- 59 + 3 appendices · 180 min
- Interactive Models
- 5, with Excel
- Data
- Satellite catalogue to 18 Oct 2026; USGS; GAO; filings
- Figures As Of
- 22 October 2026
- 4,492
- Payloads launched in 2025
- 83/100
- Optical terminals, top bottleneck
- 99%
- China's share of primary gallium
- 180 min
- Read, 59 chapters
In short
Satellites are no longer scarce: the world launched 4,492 payloads in 2025, ten times the 2019 figure, and SpaceX builds about 70 a week. What is scarce is a short list of parts with few qualified suppliers. Setcoin's bottleneck scorecard ranks five as critical: merchant optical terminals (83 of 100), solid rocket motors, launch slots, rad-hard processors and III-V solar cells. The US defence constellation needed more than 500 optical terminals and had 20 by January 2025. China produces 99% of primary gallium, and its suspension of a ban on gallium and germanium exports to the US ends on 27 November 2026. Closing the gaps to 2030 needs roughly $2–5 billion of targeted capacity.
01 · The answer
Satellites are plentiful. A short list of parts is not.
The world launched 4,492 payloads in 2025, against 453 in 2019 (Setcoin catalogue analysis). SpaceX's factory averaged about 70 Starlink satellites a week in early 2026; Amazon says it builds about 30 a week. Satellite assembly is no longer the constraint: Amazon had about 300 satellites in orbit in June 2026 because it could not launch them faster.
The constraint is a short list of parts made by a handful of qualified suppliers, with lead times of a year or more and no drop-in substitutes. Setcoin's scorecard ranks five as critical: optical terminals bought on the open market, solid rocket motors, launch slots, radiation-hardened processors and III-V solar cells. The US Space Development Agency needed more than 500 optical terminals for one layer and had 20 by January 2025; tracking satellites later flew with three terminals instead of four.
Beneath the solar cells and radio chips sit two minerals. China produces 99% of the world's primary gallium and leads in germanium; germanium's average price roughly doubled in 2025. China's suspension of its ban on exports to the United States ends on 27 November 2026. On Setcoin's arithmetic, the price effect on cells is a few per cent; the availability effect is the risk.
The largest builders escaped most limits by making parts themselves, which moves the bottleneck onto defence, sovereign and smaller programmes. Setcoin estimates that roughly $2–5 billion of targeted capacity, mostly in terminals, rad-hard electronics, cells and refining outside China, would close the gaps to 2030.
The study, chapter by chapter
59 chapters + 3 appendices · 180 min · updated 2026-10-22
- Executive Summary and Reading Guide
- What a Satellite Is Made Of: the Bill of Materials
- Demand: How Many Satellites the World Is Building
- Method: from Bill of Materials to Bottleneck Score
- A Satellite Traced Part by Part
- The Supply Chain in Tiers: Primes, Subsystems, Parts and Materials
- How the Bottlenecks Formed: a Timeline, 2018–2026
- Vertical Integration: Why the Largest Builders Make Their Own Parts
- Radiation-Hardened Processors and Space-Grade Electronics
- The EEE Parts Problem: Memories, Power Parts and Obsolescence
- Space Solar Cells: III-V Multi-Junction
- Silicon, Perovskites and the Challengers to III-V Cells
- Germanium and Gallium: the Minerals Beneath the Solar Cell
- Optical Inter-Satellite Terminals
- Case Study: How Optical Terminals Delayed a Defence Constellation
- Electric Propulsion: Hall Thrusters and Their Makers
- Propellants for Electric Propulsion: Xenon, Krypton, Argon and Iodine
- Chemical Propulsion and Hydrazine
- Solid Rocket Motors: Shared with Missiles
- Reaction Wheels, Star Trackers and Attitude Control
- Batteries and Power Electronics
- Switchgear, Connectors and the Small Parts Nobody Plans For
- RF Components, GaN and Phased Arrays
- Structures, Composites and Carbon Fibre
- Optics and Sensors: Infrared Focal Planes and Imaging Payloads
- User Terminals and the Ground Segment
- Satellite Factories: How Many Satellites Can the World Build?
- Test Facilities: Thermal Vacuum, Vibration and EMC
- Launch as the Binding Constraint
- Case Study: Amazon Leo, Producing Faster than It Can Launch
- Workforce: the Slowest Constraint to Fix
- Qualification and Heritage: Why New Suppliers Take Years
- Export Control: ITAR, EAR and Europe's ITAR-Free Drive
- Critical Minerals Policy and the November 2026 Cliff
- Tariffs and Trade
- The Geography of Supply and Single Points of Failure
- The United States: Industrial Policy for the Space Supply Chain
- Europe: Non-Dependence, IRIS² and the Supply Base
- China's Supply Chain: Scale and Constraints
- Japan, India, the United Kingdom and Other Suppliers
- Defence Demand: SDA Tranches and Golden Dome
- Mega-Constellations: Starlink, Amazon Leo, Guowang and Qianfan
- Direct-to-Device and Large Arrays
- The Bottleneck Scorecard: the Full Ranking
- Lessons from Other Industries
- Constellation Bill of Materials: Demand against Supply to 2030
- Lead-Time Schedule Risk: How One Part Sets the Launch Date
- Mineral Shock: What a Germanium Spike Does to Solar Cell Cost
- Capacity Build Needed to 2030
- Financing Supplier Capacity
- Consolidation and M&A in the Supply Chain
- The Case For and Against a Supply Crunch
- Vendor Map: Who Supplies What
- Due-Diligence Checklist for a Supplier or Programme
- Implications by Reader
- Risk Register
- Signals to Watch and What Would Change Our View
- Twenty Questions This Study Answers
- Methodology and Sources
- Appendix A: Component Reference Table
- Appendix B: Glossary
- Appendix C: Setcoin Satellite Catalogue Dataset
02 · The evidence
Four Numbers Behind the Bottleneck
27 months
Longest reported lead time
Space-grade switchgear (PwC/AIA study, March 2026, as reported)
$4,100/kg
Germanium, 2025 average
Up from $1,991 in 2024 (USGS); China's exports fell by four-fifths
2×
Solid motor capacity at one new plant
L3Harris Camden, opened 20 Oct 2026; missiles first in line
03 · The scorecard
Ten Parts That Cap the Space Economy
Setcoin scores each part 1–5 on lead time, supplier concentration, demand growth, difficulty of substitution and geographic exposure, weighted 25/25/20/15/15 and scaled to 100. Seventy-five and above is critical.
Bottleneck score, top ten of eighteen parts
score, 0–100
Setcoin scorecard, October 2026; evidence in the component chapters.
Try the model
Rank the chokepoints your way
Change the five weights and watch the ranking move. Put exposure first and III-V solar cells lead; put lead time first and solid rocket motors lead. Optical terminals and rad-hard processors stay in the top five under every weighting Setcoin tested.
Interactive model
Bottleneck score: which parts are the chokepoints?
How do eighteen critical parts rank when you weight lead time, concentration, demand growth, substitution and exposure your way?
Scenarios
Optical inter-satellite terminals (merchant)Top bottleneck
83Its score (0–100)
5Critical parts (75+)
5Serious parts (64–74)
Optical inter-satellite terminals (merchant) Critical
Solid rocket motors Critical
Launch slots Critical
Rad-hard processors and FPGAs Critical
III-V multi-junction solar cells Critical
Space-grade power switchgear Serious
Infrared focal plane arrays Serious
Connectors and harnesses Serious
GaN and GaAs RF parts Serious
Actuators and nozzles Serious
Carbon fibre and composites Watch
Space-qualified Li-ion cells Watch
Reaction wheels Watch
Star trackers Watch
Hall thrusters Watch
Xenon propellant Watch
Test facility time Watch
Hydrazine Watch
Bottleneck score, 0–100 (critical 75+, serious 64–74)
Show the calculation table
| Rank | Part | Lead time | Concentration | Demand growth | Substitution | Exposure | Score | Tier |
|---|---|---|---|---|---|---|---|---|
| 1 | Optical inter-satellite terminals (merchant) | 4 | 5 | 5 | 4 | 2 | 83 | Critical |
| 2 | Solid rocket motors | 5 | 5 | 4 | 4 | 1 | 81 | Critical |
| 3 | Launch slots | 4 | 5 | 5 | 3 | 1 | 77 | Critical |
| 4 | Rad-hard processors and FPGAs | 5 | 4 | 3 | 3 | 3 | 75 | Critical |
| 5 | III-V multi-junction solar cells | 4 | 3 | 4 | 3 | 5 | 75 | Critical |
| 6 | Space-grade power switchgear | 5 | 3 | 4 | 3 | 2 | 71 | Serious |
| 7 | Infrared focal plane arrays | 4 | 4 | 3 | 4 | 2 | 70 | Serious |
| 8 | Connectors and harnesses | 4 | 4 | 4 | 3 | 1 | 68 | Serious |
| 9 | GaN and GaAs RF parts | 3 | 3 | 4 | 3 | 4 | 67 | Serious |
| 10 | Actuators and nozzles | 4 | 4 | 3 | 3 | 1 | 64 | Serious |
| 11 | Carbon fibre and composites | 3 | 4 | 3 | 3 | 2 | 62 | Watch |
| 12 | Space-qualified Li-ion cells | 3 | 3 | 3 | 2 | 4 | 60 | Watch |
| 13 | Reaction wheels | 3 | 3 | 4 | 2 | 2 | 58 | Watch |
| 14 | Star trackers | 3 | 3 | 4 | 2 | 2 | 58 | Watch |
| 15 | Hall thrusters | 3 | 3 | 4 | 2 | 2 | 58 | Watch |
| 16 | Xenon propellant | 2 | 4 | 3 | 1 | 4 | 57 | Watch |
| 17 | Test facility time | 3 | 3 | 4 | 2 | 1 | 55 | Watch |
| 18 | Hydrazine | 2 | 5 | 1 | 2 | 2 | 51 | Watch |
Sensitivity: which input moves the answer most
What moves “Its score (0–100)” most: each input 20% lower (left of a pair) and 20% higher
Weight: geographic or political exposure
Weight: supplier concentration
Weight: demand growth
Weight: lead time
Weight: difficulty of substitution
This is one of 5 interactive models in the report. The first 3 chapters of the report are open to everyone; a Space subscription, with a 7-day free trial, opens every model and its Excel file.
04 · Three findings
Where the Constraint Really Sits
Counts from Setcoin's catalogue analysis; capacity and lead times as cited in the study.
~300 vs 1,616
Launch, not factories
Amazon builds about 30 satellites a week but had about 300 in orbit in June 2026 against a milestone of 1,616. Only SpaceX launches roughly what it builds.
Up to 27 months
Small parts, long waits
Switchgear and connectors share factories with grids and data centres. Test chambers cap component output. The cheapest parts often set the launch date.
Rocket Lab: 6 deals
Integration moves the bottleneck
SpaceX makes most of its parts; Rocket Lab bought cells, wheels, trackers, terminals and agreed to buy Iridium. Everyone else has fewer independent suppliers.
Try the model
Which part sets your launch date?
Start 30 months from launch with eight months of integration and test. With switchgear at 24 months, the programme is four months late on day one. Pre-order it, add a new terminal supplier, or go back to 2023 lead times.
Interactive model
Lead-time schedule risk: which part sets the launch date?
Working back from launch, when must each long-lead part be ordered, and does any of them push the launch out?
Scenarios
Space-grade switchgearPart that sets the date
-4 monthsSlack on that part
month 34Achievable launch
4 monthsSlip against plan
Space-grade switchgear Late
Rad-hard processors
Optical terminals
III-V solar cells
Connectors
Reaction wheels
Test queue, integration and test Programme
-2002040TodayLaunch, month 34
Months from today · order windows; the part that sets the date highlighted
Show the calculation table
| Part | Lead time (months) | Latest order (month) | Slack (months) |
|---|---|---|---|
| Space-grade switchgear | 24 | -4 | -4 |
| Rad-hard processors | 18 | 2 | 2 |
| Optical terminals | 15 | 5 | 5 |
| III-V solar cells | 12 | 8 | 8 |
| Connectors | 9 | 11 | 11 |
| Reaction wheels | 9 | 11 | 11 |
| Parts needed by (month) | 20 | ||
| Achievable launch (month) | 34 |
Sensitivity: which input moves the answer most
What moves “Achievable launch” most: each input 20% lower (left of a pair) and 20% higher
Lead time: space-grade switchgear
Planned launch (months from today)
Integration and test
This is one of 5 interactive models in the report. The first 3 chapters of the report are open to everyone; a Space subscription, with a 7-day free trial, opens every model and its Excel file.
05 · Geography
Where Scarce Parts Are Made
Examples of the sites behind the scorecard, by part family. Ringed: single points of failure, where a six-month loss would delay programmes across the industry. Switch between world, North America, Europe and East Asia.
Single point of failure (Setcoin) · hover or tap a site for detail
All 29 sites, by category
Electronics, optics and sensors
- BAE Systems (Manassas, VA) ◎ — Rad-hard processors (RAD750, RAD5545)
- Microchip (Chandler, AZ) — Rad-tolerant FPGAs; PIC64-HPSC
- Frontgrade (Colorado Springs, CO) — Rad-hard standard parts
- Tesat-Spacecom (Backnang) ◎ — Optical terminals; longest heritage
- Rocket Lab / Mynaric (Gilching) — Optical terminals (acquired Apr 2026)
- CACI optical (California) — CrossBeam terminals
- Teledyne (Camarillo, CA) — Infrared focal planes
- Sodern (Limeil-Brévannes) — Star trackers
- Jena-Optronik (Jena) — Star trackers
- Qorvo (Greensboro, NC) — GaN/GaAs RF, beamformers
- Sumitomo Electric (Yokohama) — GaN RF
Power, propulsion and motors
- Azur Space (Heilbronn) ◎ — III-V solar cells; +25% capacity 2026
- Spectrolab (Sylmar, CA) — III-V solar cells (Boeing)
- Rocket Lab SolAero (Albuquerque, NM) — III-V cells; CHIPS award
- CESI (Milan) — III-V cells
- Northrop Grumman (Promontory, UT) ◎ — Large solid boosters
- L3Harris (Camden, AR) — Solid motors; capacity doubled Oct 2026
- Anduril (McHenry, MS) — Tactical solid motors
- Avio (Colleferro) — P120C solid motors
- Safran (Vernon) — Hall thrusters
- Busek (Natick, MA) — Hall thrusters
- Saft (Poitiers) — Space Li-ion cells
- GS Yuasa (Kyoto) — Space Li-ion cells
Materials and minerals
- Umicore (Olen) — Germanium refining and substrates
- Germanium wafers (St George, UT) ◎ — Wafers mostly for satellite solar cells (USGS)
- China gallium refining ◎ — 99% of primary gallium (USGS)
- China germanium (Yunnan) — Leading germanium producer
- Toray (Japan) — Aerospace carbon fibre
- Hexcel (Salt Lake City, UT) — Carbon fibre; line restart
Company statements, USGS and reporting; examples, not a complete list.
Try the model
Is there enough supply for 2030?
Satellites that buy parts on the open market, times parts per satellite, against Setcoin's capacity estimates. In 2026 optical terminals run at about 130% of capacity; at 3,000 satellites a year in 2030 every part is over.
Interactive model
Constellation bill of materials: is there enough supply?
For the satellites that buy parts on the open market, does component capacity cover demand?
Scenarios
Optical terminalsTightest part
130.0%Its utilisation
1Parts over capacity
1,000Satellites the tightest part supports
Optical terminals Short
Rad-hard processors Tight
Reaction wheels OK
Star trackers OK
III-V solar cells OK
Demand as % of merchant capacity (Short above 100%, Tight above 80%)
Show the calculation table
| Part | Demand a year | Capacity a year | Utilisation |
|---|---|---|---|
| Optical terminals | 1,170 | 900 | 130% |
| Rad-hard processors | 650 | 800 | 81% |
| Reaction wheels | 4,160 | 6,000 | 69% |
| Star trackers | 2,080 | 3,000 | 69% |
| III-V solar cells | 1.95 | 3 | 65% |
Sensitivity: which input moves the answer most
What moves “Its utilisation” most: each input 20% lower (left of a pair) and 20% higher
Optical terminals per satellite (when fitted)
Merchant optical terminal capacity
Satellites buying merchant parts
Share carrying optical terminals
This is one of 5 interactive models in the report. The first 3 chapters of the report are open to everyone; a Space subscription, with a 7-day free trial, opens every model and its Excel file.
06 · Signals to watch
The Next 24 Months
- ~27 Nov 2026China's mineral decisionExtend, lapse or widen the gallium and germanium controls on exports to the US.
- 2026–2027Terminals per satellite at SDA launchesFull loads on schedule would show the top bottleneck easing.
- 2026–2027New Glenn back in flightThe key to Amazon's deployment and to launch for non-SpaceX constellations.
- 2027–2028Starship carries external satellitesThe step change that would move the bottleneck back to parts and test.
07 · Check it yourself
Test This Study Against the Live Record
Every figure in this study comes from the public record. These Setcoin services let you check the companies, contracts and constraints behind it yourself.
Setcoin Sectors
Who is being paid to build
Cited R&D awards and prime contracts in space and aerospace, by organisation and readiness gate, from public procurement and grant records.
Open the space sector record →Equipment Map
Which equipment is short
The switchgear and electrical shortages that slow satellites also slow grids and data centres. See categories, suppliers and pressure across sectors.
Open the Equipment Map →Setcoin Verify
Know who you depend on
Registrations, ownership, sanctions listings and court records on a supplier, cross-referenced and scored in one report the same day.
See what Verify checks →08 · The full study
What's Inside
Foundations and parts
- The bill of materials and a satellite traced part by part
- Demand from the satellite catalogue, 2015–2026
- Scoring method and supply tiers; vertical integration
- Rad-hard electronics, solar cells, minerals, optical terminals
- Propulsion and propellants, solid motors, attitude control, power, small parts, RF, structures, sensors
Capacity and policy
- Factories, test facilities, launch, workforce, qualification
- Export control, critical minerals and the November 2026 cliff, tariffs
- Single points of failure; the US, Europe, China, Japan and others
- Defence, mega-constellation and direct-to-device demand
Synthesis and tools
- The full scorecard; bill of materials to 2030; lead-time risk
- Mineral shock; capacity build and its financing
- Consolidation, vendor map, due-diligence checklist
- Risk register, signals, twenty questions answered
Ranking
Bottleneck score
Eighteen critical parts scored on lead time, supplier concentration, demand growth, substitution and exposure, with weights you can change.
Schedule
Lead-time schedule risk
Working back from launch to the latest order date for each long-lead part, and the part that sets the launch date.
Capacity
Constellation bill of materials
Merchant demand for terminals, processors, wheels, trackers and III-V cells against estimated capacity, 2026 and 2030.
Free data tools
Check the live record yourself
The study's findings move with launches, contracts and filings. These free Setcoin tools show the current public record behind them, refreshed from the catalogues and procurement registers.
Free tool
Space Contracts & Tender Radar
607 new US federal space awards and European tenders from the last twelve months, by part or service bought, buyer, country and stage, with the tenders still open.
See which parts and services agencies are buying, from propulsion and power to avionics, and who the buyers are.
Open the radar →Free tool
Constellation Deployment Monitor
8 large constellations against what they are licensed or filed for: satellites launched, re-entered and net pace over twelve months, and the date each milestone is reached at that pace.
Constellation pace drives component demand: see how many satellites each fleet still needs and how fast it is launching.
Open the monitor →Free tool
Space Claim Check
Type an operator’s name and see what the public satellite catalogues record: satellites launched, in orbit, operating and re-entered. The full check adds US contracts, SBIR awards, UK and NOAA licences and list screening.
Check a supplier’s or customer’s satellite record before you sign.
Check an operator →Questions
Questions About the Study, Answered
What are the main bottlenecks in the space supply chain?
Setcoin's scorecard ranks five parts as critical in October 2026: optical inter-satellite terminals bought on the open market (83 of 100), solid rocket motors (81), launch slots (77), radiation-hardened processors and FPGAs (75) and III-V multi-junction solar cells (75). Space-grade switchgear, infrared focal planes, connectors, GaN and GaAs RF parts and actuators follow as serious.
Is there a shortage of satellites?
No. The world launched 4,492 payloads in 2025, ten times the 2019 figure, on Setcoin's analysis of the public satellite catalogue. SpaceX built about 70 Starlink satellites a week in early 2026 and Amazon says it builds about 30 a week. Launch and a short list of parts, not satellite factories, limit deployment.
Why are optical inter-satellite terminals scarce?
Demand jumped from dozens to hundreds a year when the US Space Development Agency built its proliferated constellation around laser links, while only four merchant suppliers were qualified. GAO reported that Tranche 1 Transport needed more than 500 terminals and had 20 by January 2025. SpaceX makes its own and does not sell them.
How exposed are satellites to China's gallium and germanium controls?
Highly in materials, modestly in cost. China produces 99% of primary gallium and leads in germanium; both sit beneath III-V solar cells and RF chips. Setcoin estimates germanium adds about 2–10% to III-V cell cost across 2024 prices and a doubling shock. The risk is availability: China's suspension of its US export ban ends on 27 November 2026.
Why do large constellations use silicon solar cells?
Because III-V cell capacity is counted in megawatts a year worldwide, while a constellation launching thousands of satellites a year needs tens of megawatts. Silicon is cheaper, made at vast scale and adequate for five-year missions in low orbit. Defence, geostationary and long-life satellites still need III-V cells.
What are typical lead times for space components?
Reported lead times run up to 27 months for space-grade switchgear, one to two years for rad-hard processors and a year or more for III-V solar cells; the median stated lead time in one small-satellite parts catalogue was six months. Setcoin's lead-time model shows how a cheap part can set a programme's launch date.
Is xenon a constraint for electric propulsion?
For a large constellation using xenon alone, yes: world output is a few tens of tonnes a year and a fleet of 1,500 satellites at 10 kg each would take about a quarter. The largest builders switched to krypton and argon, which cost far less but need more power and larger tanks.
Why are solid rocket motors short?
Two US incumbents produce most motors, and missile and interceptor demand after the wars in Ukraine and the Middle East takes priority. Capacity is expanding: L3Harris opened a plant in Arkansas on 20 October 2026 that it says doubles capacity, and Northrop Grumman reports large increases. Space launch is a small, lower-priority customer.
How much would it cost to remove the bottlenecks?
Setcoin estimates roughly $2–5 billion of targeted capacity to 2030, excluding solid motors and launch: optical terminal lines, rad-hard electronics, III-V cells, gallium and germanium refining outside China, attitude-control lines, test facilities and qualified small parts. Multi-year purchase commitments matter as much as capital.
How do I read the full study?
The first three chapters are open to everyone in full, with no account; the rest can be previewed. A space sector subscription, with a free trial, opens all 59 chapters, the five interactive models and their Excel files.
Setcoin Research analyses public markets, filings and programmes for information only. It is not investment advice and does not recommend any security.
Read Space Supply Chain
1. Read the first 3 chapters in full, no account needed. 2. Create a free account (two minutes, no card) to download their Excel models. 3. Subscribe to Space for every chapter, model and Excel download: 7-day free trial, then €199 a month or €1,490 a year.
Space Supply Chain · first 3 chapters open to allCreate a free account