Skip to content
Technology How it works Breeder — Hyperion Burner — Aegis Burner — MetroVolt AI-Native Architecture Magnets Fuel cycle Safety Roadmap
Solutions AI & Data Centers Defense & Government Grid & Baseload Neutron Detection Quantum
Learn Technical Library
Proof Publications Whitepapers Technical Library Open Science & Reproducibility The Honest Gates
Company About / Mission Leadership Environment Health & Safety Investors Careers Press Contact
3D Model
Aegis › The Physics
The Physics
Aegis / MetroVolt burner · design render
Aegis / MetroVolt burner · design render · © 2026 KFE

The Honest Gates, Together

Loss cone in velocity space — the tandem-mirror stability driver.
Loss cone in velocity space — the tandem-mirror stability driver. canon 2026-08-27 · © 2026 KFE

Four physics gates define the burner's design stage: plug coil overstress, an unprecedented plug regime, a helium-3 shortfall, and availability short of Tier III.

The four gates, in one place

The burner's physics is promising and its gates are real. We state all four at full strength, because honest gates are what make the promise credible. None of them is an economic claim; each is a physics-and-engineering fact about a machine that is an engineering program, not yet built.

GateStatementConsequence
Plug coilOverstressed ~3–3.9× at the design boreInfeasible as specified
Plug regime166–830× beyond any deviceCannot be post-dicted today
He-3 supply~400× domestic supply per unitNeeds breeder-bred + lunar He-3
Availability0.86–0.995 vs Tier III 0.9998230–100× short; not a sole source

Two of the four — coil overstress and the plug regime — live in the end plug, the heart of the tandem mirror, and are coupled: the field that the plug needs is the field that breaks the coil, and the regime that field enables has never been tested. The other two are systemic: the fuel the machine burns does not exist at scale, and its uptime, as modelled, trails the most demanding standard.

Why they are shown, not hidden

A design that hides its gates cannot be trusted on anything else. By stating the 3–3.9×, 166–830×, ~400×, and 30–100× figures exactly, we make clear what must be solved before the burner is more than a study — and why the program builds the breeder first and stages the burner through a test machine (~2032) toward a commercial design (~2036).

Presented together, the gates also show a path: two are magnet-and-plasma problems concentrated in the end plug that a staged experimental program can attack directly, one is a fuel-supply problem the breeder is built to solve, and one is a reliability problem that improves with maturity and pairing. None is waved away, and none is presented as already solved — which is the whole point of stating them.