Redundancy and N+1

Redundancy keeps critical loads powered when a unit fails; N+1 and higher configurations trade added units for tolerance of failures.

Designing for failure

No single generator is perfectly reliable, so resilient sites deploy redundant units. N+1 means enough capacity to serve the load (N units) plus one spare, so any single unit can fail or be maintained without dropping the load. Higher schemes (N+2, 2N) tolerate more concurrent failures at the cost of more units and more fuel or fuel-equivalent.

Redundancy interacts with availability. If each unit has modest availability, redundancy is how a site reaches high aggregate availability. This matters for the burner, whose single-unit availability at the design stage is estimated at 0.86-0.995, well short of the 0.99982 that hyperscale Tier III facilities expect. Redundancy is one lever to close part of that gap, but it does not erase it.

Redundancy is not free

Each redundant unit adds capital, footprint, maintenance burden, and, for combustion sources, fuel. For a fusion generator it would also add fuel demand, which for the burner is already gated by helium-3 supply. Redundancy is a genuine resilience tool but must be reasoned about against these constraints.

• N+1 tolerates one failure; 2N is a full duplicate
• Redundancy raises aggregate availability from per-unit availability
• Burner per-unit availability 0.86-0.995 vs Tier III 0.99982
• More units mean more footprint, maintenance, and fuel demand

Redundancy is only effective if failures are independent. Two units that share a common fuel supply, control system, or cooling loop can fail together, defeating the redundancy on paper. Robust N+1 design therefore requires independent support systems, not just duplicate generators. For a fusion deployment this means independent fuel handling and controls per unit, a real design requirement layered on the underlying machine gates.

Availability figures here are design-and-simulation estimates, stated plainly.