Availability Gate
Projected machine availability is 0.86–0.995 against a hyperscale Tier III target of 0.99982 — 30–100x more downtime; not a sole Tier-III source today.
This is the fourth honest gate on the burner. Availability is the fraction of time the machine is producing power. The design projects an availability of 0.86 to 0.995. A hyperscale data-center Tier III standard is 0.99982. The gap means the burner as projected has roughly 30 to 100 times more downtime than a Tier III source requires.
The plain reading: the burner is not, by itself, a Tier-III-grade power source today. It could serve loads that tolerate more downtime, or it could sit behind additional redundancy and storage, but a single unit at 0.86–0.995 cannot meet a 0.99982 target alone. This is a reliability fact, stated without any economic translation.
What drives the downtime
- Cryogenic faults and magnet quench recovery
- Loss of plug plasma or plug potential
- Wear of DEC electrodes, first wall, and heating systems
- Vacuum and pumping faults, and regeneration cycles
- Scheduled maintenance and component replacement
How the gate could close
Closing the gap means driving down both fault frequency and recovery time: deeper cryogenic margin, robust quench handling, redundant heating and pumping, modular fast-replacement components, and — most fundamentally — resolving the plug gates so the plasma runs stably. Redundancy at the installation level (multiple units, storage) can also raise system availability above any single machine. None of this is demonstrated yet, which is why the gate is stated at 0.86–0.995.
All figures are design-and-simulation values for a machine not yet built.