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
AI Architecture › L7 · Ecosystem & Strategy
L7 · Ecosystem & Strategy

Deuterium Supply Tracking

Deuterium fuels both machines; L7 tracks its inventory, purity, and consumption rate against burn plans so fuel never becomes the limiting constraint.

THE STACK · click to jumpL7Ecosystem & StrategyL6Experience & VisualizationL5Applications & CopilotsL4OrchestrationL3Twin Modeling & AIL2Data FabricL1Control PlaneL0Foundation▲tlmctl▼L7 · ECOSYSTEM & STRATEGYThe plant in its world — integrated through one unified API.1Unified API Layerone door in/out2Grid Integrationdispatch & firm supply3Supply Chainfuel, parts, isotopes4Maintenanceservice & spares loops5Regulatorycompliance & reporting6Fleet Strategymulti-unit planningMACHINE TIEConnects the machine to grid, suppliers, and regulators — the outermost loop.KRONOS FUSION ENERGYAI-NATIVE S.M.A.R.T. GENERATORECOSYSTEM & STRATEGYSHEET 09REV. 2026-08L7 · AI-NATIVE STACK
L7 · Ecosystem & Strategy — its place in the stack (left, click any layer) and its internal components (right). Telemetry rises; control descends.

The common feedstock

Deuterium is the one feedstock both machines share. The breeder burns deuterium with tritium; the burner burns deuterium with helium-3. Deuterium is abundant and separable from water, so the supply question is not scarcity but continuity, purity, and matching delivery to burn rate. L7 tracks deuterium as an inventory with assay and a consumption model tied to each unit's operating plan.

Purity gating and consumption modeling

Fuel purity matters to plasma performance and to the diagnostics that infer plasma state; contaminants raise effective charge and radiation. Inbound deuterium is assayed and gated before it enters the fuel cycle, and the assay is recorded so a later performance anomaly can be correlated with a fuel batch. Consumption is modeled from the twin's burn state, not from nameplate, so inventory projections track what the machine actually uses.

python
# days-of-supply projection per unit
inv    = deuterium_inventory(unit)          # kg, assayed
rate   = twin_burn_rate(unit)                # kg/day from L3 state
inbound = scheduled_deliveries(unit)         # (day, kg) list
def days_of_supply(horizon):
    stock = inv
    for d in range(horizon):
        stock += delivered(inbound, d) - rate(d)
        if stock < reserve_floor(unit):
            return d                         # trigger reorder before floor
    return horizon

L7 keeps a reserve floor per unit and triggers reorder before inventory approaches it, so a delivery slip never forces an unplanned burn stop. Because deuterium serves both machine types, a fleet view aggregates demand across the breeder foundry and burner fleet and smooths ordering.

Deuterium supply is the least constrained node in the material graph; the binding constraints are lithium-and-TBR for the breeder and helium-3 for the burner. L7 nonetheless tracks it with the same rigor so that fuel is never the reason a unit under-produces.

Consumption figures today are twin-simulated at design-stage burn rates. The tracking interface is exercised now so that, from construction start Q2 2027 through FOAK, real deliveries and burns flow through the same accounting.

Content reviewed August 2026 · design-and-simulation stage