Technical articles, research summaries, and project updates. Follow our progress as we work toward making a Dyson swarm a reality.
Six months of depth-over-breadth work wraps up: 44 external papers obtained, 4 critical deliberations concluded, Papers 04 and 05 drafted, and Paper 03 strengthened with controlled experiments.
Consensus: yes. A sunshield does 95% of the work, active cooling handles the rest at less than 1% of station power. Zero boil-off is an engineering problem, not a physics barrier.
Consensus: pure microgravity smelting won't work, but a hybrid station with a rotating smelting arm at 0.05-0.15g solves the physics. The 6-8 order magnitude gap is an architecture problem, not a research dead end.
Consensus: MLI will degrade 2-3.5x over 20 years at L4/L5. Design for it with LBMLI, active intermediate shields, and planned replacement cycles rather than hoping it won't happen.
Consensus: a modular 3-layer membrane sunshield, ~60m class, assembled incrementally from gore-shaped segments over 3-4 deliveries. No full-station spin required.
Monte Carlo analysis of 10,000 scenarios shows C-type asteroids deliver water to L4/L5 at $3,333/kg vs $4,845/kg for lunar ice. Asteroids win 90% of the time, driven by electric propulsion's payload fraction advantage.
Systematic literature review, external research, and four multi-model deliberations have materially improved the technical feasibility picture. Microgravity metallurgy goes from TRL 2-3 to 3-4 via a hybrid gravity architecture, and microgravity electrolysis jumps to TRL 4-5 after a Nature Chemistry breakthrough.
Our swarm coordination paper derives closed-form design equations for per-cluster TDMA sizing. Two tests -- byte budget and airtime schedulability -- yield a 35 kbps recommended PHY rate. An independent NS-3 simulation confirms the feasibility boundary within 3-8%.
Our ISRU economic crossover paper achieves 3/3 Accept after 39 rounds of AI peer review. A 10,000-run Monte Carlo with 20 stochastic parameters finds crossover in 85% of draws at r=5%, with ISRU capital cost K explaining 63% of variance.
Project Dyson shifts from expanding its question universe to deepening technical rigor on the critical unknowns. New analysis tools — a feasibility report, TRL dashboard, critical path visualization, and decision gates — establish the framework for publication-quality Phase 0 assessment.
We reviewed 15 arxiv papers on Mercury factories, lunar mining, asteroid economics, and gas giant resources to assess alternative feedstock pathways. The result: five new research questions and a major gap in comparative economics.
Fifteen non-arxiv sources from NASA technical memoranda, AIAA conference proceedings, the journal Cryogenics, and a US patent fill the gaps in our cryogenic boiloff management research. The key finding: the physics works at scale on the ground, but flight cryocoolers are 2-3 orders of magnitude below station requirements.
A review of ~30 external sources across 7 open research questions finds significant new evidence from 2024-2026. Four questions move from 'open' to 'investigating', including SpaceX's operational argon thrusters, a Nature Chemistry magnetic separation breakthrough, and the first orbital semiconductor manufacturing demonstration.
A second wave of literature review tackles electrostatic contamination of mining robot mechanisms, silicon purification without zone refining, propellant boiloff at L4/L5, and machine learning for asteroid composition—with one surprising finding that may retire a question entirely.
A third wave of literature review tackles slag separation in microgravity, the concentrator-vs-flat-plate architecture decision for 100 MW solar arrays, and in-space manufacturing of array structures from asteroid-derived metals. The key finding: electromagnetic separation and ultrasound-assisted additive manufacturing may solve two problems simultaneously.
Lacki (2025) calculates that an uncontrolled megaswarm at 1 AU grinds itself to dust in 41,000 years. We reviewed his findings against our existing plans—and found two gaps we hadn't considered.
A deep dive into published literature on regolith excavation, anchoring reliability, microgravity metallurgy, and silicon purification—the four questions that could make or break Phase 0.
We assess three arxiv papers spanning Dyson swarm engineering, thermodynamic efficiency limits for Matrioshka Brains, and humanity's path to Type II civilization. Wright's thermodynamics paper independently confirms our finding that nested shells hit diminishing returns fast.
A new taxonomy system organizes Project Dyson's research questions, BOM specs, blog posts, and validations into 13 engineering domains and 52 topics — enabling cross-artifact discovery without changing a single tag.
Literature review of 14+ arxiv papers converges on water as the correct first extraction target for asteroid ISRU. Economics, simplicity, and the bootstrap effect all point the same way.
Monte Carlo comparison of sequential, batch, greedy, and NN-guided deployment strategies shows batch deployment wins 100% of runs. Deployment-regime NN retraining (val MSE 0.0005) fixes accuracy but not strategy — uniform-radius slots make NN trajectory optimization structurally irrelevant.
Pre-computed FEA eigenvalue analysis reveals that analytical flutter models were systematically optimistic by 2-4x. Membranes above 400m diameter require higher tension or spin stabilization -- passive boom tensioning at 1 N/m is insufficient for the planned 500m-1km arrays.
Monte Carlo simulation of self-replicating foundries reveals that manufacturing degradation per generation—not closure ratio—is the critical factor. At 5% degradation, 31% of scenarios never reach target scale.
Trade sweep simulation across 0.1-2.0 AU shows all distances are thermally feasible for Shkadov mirrors. Close-in placement at 0.1 AU minimizes mirror mass by 400× compared to 1.0 AU, with identical thrust output.
1D analytical model with Monte Carlo uncertainty shows solar mass extraction up to 10¹³ kg/s is feasible at 5% lifting efficiency. The Sun barely notices—luminosity perturbation is less than 0.001% over 1000 years.
Monte Carlo simulation confirms that thermal warping in thin-film membranes up to 1 million m² is fully suppressed by standard boom tensioning (≥0.5 N/m). No active compensation needed.
Monte Carlo thermodynamic cascade analysis shows 4 nested shells achieve 50.8% energy extraction efficiency. Beyond 4, diminishing returns and enormous cryogenic radiators make additional layers impractical.
Research into upgraded metallurgical-grade silicon, 2D materials, and metamaterials expands material options for solar collector manufacturing, potentially relaxing silicon purity requirements and reducing mass.
A deep dive into asteroid compositional variability and its implications for in-situ resource utilization, synthesizing findings from sample return missions and spectroscopic surveys.
A fundamental revision to Project Dyson cost estimates based on multi-model consensus. Self-replicating ISRU economics reduce Phase 2-3 budgets by 10-1,350x.
Our analysis reveals that terrestrial supplies of tellurium and indium are fundamentally insufficient for Dyson-scale photovoltaic deployment, making alternative cell chemistries or asteroid mining mandatory.
Our research into energy storage technologies for Project Dyson's L4/L5 Processing Station reveals why lithium-ion batteries remain the practical choice despite their limitations, and what the future may hold.
Research into satellite failure patterns reveals mature optimization frameworks but critical data gaps for Dyson swarm maintenance fleet sizing.
Research into silicate-sulfuric acid processing provides a closed-loop alternative to thermal metallurgy for asteroid ISRU, avoiding melt containment challenges in microgravity.
A deep dive into radiation degradation research for III-V solar cells operating outside Earth's magnetosphere at the L4/L5 Lagrange points.
Multi-model consensus on organizational architecture for Project Dyson. A three-layer hybrid governance structure designed for multi-century persistence, volunteer coordination, and capture resistance.
A deep dive into micrometeoroid impacts on optical surfaces reveals that Dyson swarm communication systems can survive decades in interplanetary space with the right design choices.
How do you protect electronics across a fleet of orbital tugs operating from Mercury's orbit to beyond Earth? Our analysis of radiation effects research reveals a stratified approach that balances safety, reliability, and the $200 million question.
Our research into reversible computing reveals a path to 10-100x energy efficiency gains for the Matrioshka brain, but success depends on a temperature-stratified hybrid architecture.
Recent research validates that Graph Neural Networks and mean-field mathematics can scale swarm coordination to billions of units, providing theoretical backing for Project Dyson's Phase 2 architecture.
Introducing Phase 3a, a development track that transforms the Dyson swarm into a nested computational megastructure harvesting waste heat through a thermodynamic cascade.
Introducing our new discussion system where Claude, Gemini, and GPT collaboratively debate policy decisions through structured rounds of responses and voting.
Introducing Phase 3b, a parallel development track that transforms the Dyson swarm into a stellar propulsion system capable of moving the entire Solar System through the galaxy.
Consensus on a Continuous Assurance Architecture for assembly robots: hardware-enforced safety boundaries, runtime verification, and 10 billion simulated robot-hours.
Consensus on a five-tier authority framework where 95% of maintenance is autonomous. The math is brutal: human approval for even routine operations would double fleet size.
Consensus: 50 km minimum separation, three-layer probabilistic certification framework, and formal verification of flocking algorithms. The scaling exponent is the existential unknown.
Consensus establishes a moderate ISRU transition timeline: Earth-supplied feedstock for Years 1-3, bulk metals ISRU by Year 4, 50-60% self-sufficiency by Year 6.
Our multi-model discussion reached consensus on a modular human-rating approach that captures 80% of the cost savings while eliminating catastrophic retrofit risk.
Consensus: use solar radiation pressure for passive orbital segregation. Design the failure state, not just the operational state. Tracking thousands of dead nodes is the real challenge.
Consensus: design for propellant production from Day 1, but deploy it 18-24 months after commissioning. Water capture is non-negotiable; cryogenic storage is the hard part.
Consensus on tiered-authority governance with append-only slot lifecycle, quarantine-first protocol, and Raft consensus. It's a trajectory uncertainty problem, not a distributed systems problem.
Multi-model consensus establishes a phased power architecture: local use first, Mercury beaming second, Earth delivery via relay constellation only at scale.
Consensus on a tiered disposal protocol: salvage at depot (primary), heliocentric graveyard (fallback), passive safety features (baseline). Solar impact is eliminated.
Consensus establishes a four-tier threshold framework: 100 GW for market entry, 1 TW for grid significance, 10 TW for LCOE crossover. Architecture choice spans 3 orders of magnitude in outcomes.
How recent papers on asteroid mining, Dyson sphere detection, and trajectory optimization are shaping our specifications and opening new research questions.
New organizations registry tracks space agencies, research labs, and industry partners whose expertise and data inform Dyson swarm planning.
New validation tracking system connects theoretical claims to experimental evidence, building confidence in Dyson swarm specifications through systematic verification.
Introducing our resolution tracking system that documents how research questions are answered, providing transparency into Project Dyson's decision-making process.
New cost analysis tools help quantify uncertainty in Dyson swarm budgets, from confidence intervals on BOM items to reconciliation analysis across LLM estimates.
Discrete event logistics simulation reveals that 150,000-200,000 km depot spacing achieves <7 day mean time to repair with 85%+ fleet utilization for billion-unit maintenance operations.
Monte Carlo cost modeling identifies the crossover point where in-space manufacturing becomes cheaper than Earth production plus launch—approximately 3,500 units under baseline assumptions.
Pareto frontier analysis comparing 8 orbital locations for Assembly Hub and depot placement. Sun-Earth L4/L5 emerges as optimal for Phase 1, with 0.7 AU heliocentric as a Phase 2 option.
Discrete event simulation demonstrates that hierarchical coordination scales to 1M+ nodes while centralized architectures bottleneck at ~10,000. Optimal coordinator duty cycle: 24-48 hours.
Monte Carlo simulation reveals that solar radiation pressure provides sufficient station-keeping for collectors at ≤0.7 AU, potentially eliminating propellant costs for inner-system swarm operations.
Analysis reveals xenon demand would exceed global production by 15-20x. Alternative propellants and hybrid architectures offer the only viable path forward.
Discrete event simulation reveals the sweet spot for transport vehicle fleet sizing—not too few large vehicles, not too many small ones.
Monte Carlo simulation demonstrates that ground-based spectral analysis achieves only 47% survey efficiency due to decision latency—not bandwidth.
Monte Carlo analysis of 5,500+ simulated missions shows that electric propulsion choice matters more than constellation size for identifying high-value mining targets.
Recent peer-reviewed studies from 2022-2025 provide critical insights on habitability constraints, construction approaches, and material requirements.
How Gemini 3 Pro, GPT-5.2, and Claude Opus 4.5 differ in their analysis of Dyson swarm construction phases.
Analyzing the strengths and blind spots of AI analysis for asteroid mining and resource processing.
Side-by-side comparison of Phase 0 Bill of Materials from Gemini 3 Pro, GPT-5.2, and Claude Opus 4.5.
A review of solar sail technology and its applications for Dyson swarm satellite positioning and station-keeping.
Explaining our methodology for using multiple large language models to analyze complex engineering challenges in Dyson swarm construction.
A deep dive into the first phase of Dyson swarm construction: establishing space-based resource extraction and processing capabilities.
An overview of Project Dyson, our mission to make a Dyson swarm a reality, and how we plan to get there through collaborative planning and research.