Co-Location with Data Centers
The burner (MetroVolt) is housed to sit beside computing load, delivering firm clean power on-site — though availability remains an open gate.
Data centers need large, steady, round-the-clock electricity, and their growth is straining grids. One response is to place firm clean generation directly beside the load. The burner in its MetroVolt housing (Aegis / MetroVolt) is conceived for exactly this: a compact deuterium–helium-3 generator co-located with computing demand, delivering firm clean power on-site.
Why co-location helps
On-site generation avoids long transmission runs and the losses and delays of grid interconnection, and it matches a continuous load with continuous supply. A compact firm clean generator lets a data center reduce its reliance on grid capacity and fossil backup. The burner's direct energy conversion and low 5.44% neutron fraction shape a design intended for this dense, load-adjacent role.
The honest gates
This vision faces the burner's real gates, stated plainly. Its availability in simulation is roughly 0.86 to 0.995, versus the 0.99982 uptime hyperscale Tier III data centers require — 30 to 100 times more downtime than a sole Tier III source may allow, so today the burner is a complement to, not a replacement for, Tier III power. Its plug coil is overstressed at the design bore, its operating regime is far beyond any existing device, and its helium-3 demand vastly exceeds terrestrial supply. These are physics facts, not economics.
- MetroVolt houses the burner for co-location with computing load.
- On-site firm clean power avoids transmission and matches continuous demand.
- Availability (~0.86-0.995) is short of Tier III (0.99982) — an honest gate.
- Plug-stress, operating-regime, and helium-3 gates remain open.
Design-and-simulation framing. The Kronos machines are today design and simulation studies: the breeder (Hyperion) and the burner (Aegis / MetroVolt). No hardware net-gain has been demonstrated. Breeder construction is planned to begin Q2 2027, with first-of-a-kind (FOAK) first tritium targeted around 2030. Comparisons on this page are qualitative and use only public, defensible figures; nothing here is a performance guarantee.
Co-location is a compelling design direction, presented with its open engineering gates fully in view.