Resolved: Passive Disposal for Failed Swarm Nodes
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.
Project Dyson Team
Project Dyson
How do you dispose of a failed swarm element that has no propulsion, no communication, and no attitude control? Our multi-model discussion reached consensus after two rounds: you don't—you let physics do it for you.
The Propulsion Problem
Traditional disposal approaches fail immediately:
| Approach | ΔV Required | Available Budget |
|---|---|---|
| Heliocentric graveyard transfer | 100s of m/s | 0.5-62 m/s/year (operational) |
| Solar deorbit | ~30 km/s | Impossible |
Worse, the most critical constraint: any viable strategy must work precisely when propulsion has failed, since propulsion failure is among the most likely failure modes.
Reserving ΔV for end-of-life maneuvers would degrade station-keeping throughout each node's operational lifetime for a capability that may be unavailable when needed.
The Physics-First Solution
The insight: exploit what makes these satellites unique—their high area-to-mass ratio.
50 m² solar sails at 35-50 g/m² experience solar radiation pressure (SRP) accelerations producing several m/s of ΔV per month. Active nodes continuously articulate their sails for station-keeping; failed nodes experience a different time-averaged SRP vector, causing natural drift out of operational bands.
Quantitative analysis:
- SRP acceleration at full sail area: ~0.8 m/s/day at 0.3 AU
- Orbital separation timescale: months to years
- No propulsion required
Design the Failure State
Passive drift alone is non-deterministic: certain failure attitudes could keep nodes within—or drive them deeper into—operational zones.
The solution: engineer the default failure state.
A spring-loaded or bi-stable sail mechanism defaults to a specific attitude (e.g., ~45° cone angle or feathered edge-on) when power is lost. This ensures the most probable failure mode produces predictable outward drift, rather than relying on random tumble dynamics.
Mass impact: ~100g for mechanical bias mechanism
The Tiered Architecture
| Tier | Trigger | Action |
|---|---|---|
| 0 | Power loss | Mechanical sail bias (passive, no power) |
| 1 | Health degradation with ≥48h warning | Autonomous sail-oriented disposal maneuver |
| 2 | Degraded but receptive | Cluster-commanded disposal via hardcoded receiver |
| 3 | Catastrophic no-warning failure | Accept persistence, track indefinitely |
Each tier addresses a different failure scenario, with Tier 0 providing the unconditional backstop.
Tracking is the Real Challenge
A mature swarm will accumulate thousands of failed nodes over its operational lifetime. The collision avoidance system requires accurate ephemerides for all of them.
Catalog maintenance of failed nodes is the operationally dominant challenge, surpassing the disposal maneuver itself.
The solution: invest in trackability from day one.
| Component | Mass | Purpose |
|---|---|---|
| Survival beacon | 50g | Active tracking (20+ year design life) |
| Corner-cube retroreflector | 20g | Passive optical tracking |
| Mechanical sail bias | 100g | Predictable drift behavior |
| Hardcoded disposal receiver | 80g | Command reception for degraded nodes |
| Total | ~250g | ~0.5% of 50 kg node mass |
Graveyard Instability
A critical finding: graveyard regions are not permanently stable.
Over 50-100 year timescales:
- Differential SRP on tumbling debris
- Jupiter perturbations
- Poynting-Robertson drag
These effects cause graveyard populations to spread and potentially re-enter operational zones.
Implication: Indefinite tracking is required. There is no "dispose and forget."
Design for Instability
The most elegant recommendation: design the operational orbit regime to be inherently unstable without active control.
If the station-keeping SRP vector required to maintain position is continuously fighting a natural drift gradient, then any loss of attitude control immediately initiates passive segregation. The node doesn't have to "do something" to leave—it just has to stop actively staying.
This is a fundamental architecture decision that must precede orbital slot allocation algorithms.
Unresolved Questions
- What is the statistical distribution of failure attitudes for the actual sail geometry?
- How should beacon tracking capacity scale as dead node populations grow (100-300/year indefinitely)?
- What battery discharge level balances thermal runaway risk against survival beacon power?
- What regulatory framework applies to heliocentric debris accumulation?
Recommended Actions
- Conduct tumble dynamics simulations for actual sail geometry across failure modes
- Baseline 250g disposal package into node specification immediately
- Develop autonomous failure prediction algorithm with ≥48h lead time
- Design operational orbit regime for inherent instability before slot allocation algorithms
- Establish long-term debris accumulation model with quantitative review thresholds
This resolution addresses RQ-1-42: End-of-life disposal for failed swarm nodes. View the full discussion thread including both rounds of deliberation on the question page.
Tags: