Phase 3b: Stellar Engine - Moving the Solar System

Today we're announcing Phase 3b: Stellar Engine, a parallel development track that runs alongside Phase 3a (Matrioshka Brain). While 3a transforms the Dyson swarm into computational substrate, 3b transforms it into a propulsion system—enabling controlled movement of the entire Sun and Solar System through the galaxy.

Why Move the Sun?

A stellar engine could serve multiple critical purposes over cosmic timescales:

• Avoiding cosmic hazards: Supernovae, gamma-ray bursts, rogue stars
• Adjusting stellar orbits: Optimizing position within the galactic habitable zone
• Maintaining habitability: Compensating for long-term changes in solar output
• Interstellar migration: Eventually moving the Solar System to other star systems

Two Complementary Approaches

Based on Caplan's 2019 paper "Stellar Engines: Design Considerations for Maximizing Acceleration," Phase 3b implements two complementary propulsion systems:

Shkadov Mirror Array (Passive Thruster)

A distributed array of reflective statite elements forming a partial spherical cap on one side of the Sun. By reflecting solar photons asymmetrically, net thrust is generated through pure momentum transfer—no fuel required.

Parameter	Specification
Acceleration	~10⁻¹² to 10⁻¹³ m/s²
Fuel Required	None (passive)
Time to move 1 ly	~1 billion years
Architecture	Distributed statite swarm

Key consensus from our multi-model technical review:

• Standoff distance: 0.1 AU baseline (vs 1.0 AU alternative)
• Reflectivity: ≥95% minimum, ≥99.5% goal
• Areal density: ~1.0 g/m² target
• Interception fraction: Start at 1-5%, scale to 10-25%

Thermonuclear Jet Engine (Active Thruster)

An array of fusion-powered engines using mass lifted from the Sun's surface. Helium-4 is separated from collected solar material and burned in D-³He fusion reactions, producing directed exhaust at ~0.01c.

Parameter	Specification
Acceleration	~10⁻⁹ m/s² (1000x Shkadov)
Thrust	~10¹⁸ N total
Mass flow	~10¹² kg/s
Time to move 1 ly	~1 million years

Key consensus:

• Architecture: ~10,000 modular engine units at ~10¹⁴ N each
• Fusion reaction: D-³He baseline with D-D fallback
• Ignition: Magnetized target fusion with heavy-ion beam assist
• Magnetic nozzle: 200-250 T throat field, 70-80% efficiency

Phase 3b Bill of Materials

The complete Phase 3b BOM includes 8 major systems:

1. Shkadov Mirror Array - Passive radiation pressure thrust ($10-100T)
2. Thermonuclear Jet Engine - Fusion-powered directed thrust ($10-100T)
3. Solar Wind Collectors - Plasma collection infrastructure ($10-50T)
4. Mass Lifting Systems - Solar chromosphere extraction ($50-200T)
5. Helium Separation Plant - Isotope separation for fuel ($10-50T)
6. Electromagnetic Accelerators - Hydrogen return and helium jets ($10-50T)
7. Dyson Integration Layer - Power routing from swarm ($5-20T)
8. Thrust Stabilization Systems - Long-term trajectory control ($5-20T)

Total estimated cost: ~$110T (over 200-500 years)

Parallel with Phase 3a

Phase 3b runs in parallel with Phase 3a (Matrioshka Brain). Both phases:

• Depend on Phase 2 (completed Dyson swarm) infrastructure
• Can proceed independently
• Share manufacturing and logistics systems
• Require coordination for geometric constraints

The timeline visualization at /plan now shows this fork with separate development tracks for 3a and 3b.

Key Open Questions

Our multi-model consensus identified several critical unresolved issues:

For Shkadov Mirror:

• Optimal standoff distance (0.1 AU vs 1.0 AU trade study needed)
• Long-term membrane degradation from solar wind sputtering
• Planetary insolation impact at high interception fractions
• Torque management over geological timescales

For Thermonuclear Jet:

• ³He supply and isotopic economics
• Magnetic nozzle plasma detachment at scale
• Solar response to continuous mass extraction
• Array-level electromagnetic interference

Next Steps

1. Prototype campaigns at multiple standoff distances (0.05-1.0 AU)
2. Subscale fusion demonstrators at 1-1000 kg pellet scale
3. Coupled stellar-engineering models for solar response prediction
4. Interface control documents for Shkadov/Caplan/Dyson swarm integration

Explore the Specifications

Full technical specifications from Claude Opus 4.6, Gemini 3 Pro, and GPT-5.2 are available for each BOM item:

• Shkadov Mirror Array
• Thermonuclear Jet Engine
• Solar Wind Collectors
• Mass Lifting Systems
• Helium Separation Plant
• Electromagnetic Accelerators
• Dyson Integration Layer
• Thrust Stabilization Systems

Each page includes individual model proposals, consensus documents, and divergent views where the models disagree.

---

The addition of Phase 3b represents a major expansion of Project Dyson's scope—from energy harvesting to stellar-scale propulsion. Moving from "collect the Sun's energy" to "move the Sun itself" requires solving some of the most challenging engineering problems ever conceived, but the multi-model consensus approach gives us a structured path forward.