The Army picked five companies to build five different nuclear microreactor designs at five different bases this week, and what I keep coming back to is how deliberate the scattering is. Fort Bragg gets one company, Fort Benning another, Fort Hood, Fort Campbell, and Fort Drum each get their own vendor and their own reactor architecture, all funded through the same $2.2 billion program. Nobody at the Pentagon tried to pick the winner. Given how many of these microreactor startups exist mostly as press releases and DOE memoranda of understanding right now, spreading the bet five ways reads less like grand strategy and more like an honest admission that nobody actually knows which one ships on schedule.

The spec is the part I like: 1 to 20 megawatts, small enough to run a base’s critical systems, wired into the regular grid in peacetime but able to island off if it needs to. That’s quiet redundancy-building for something most of the country’s grid doesn’t have, a spare tire scaled down to fort size.

Utah’s FORGE site is running its own patient experiment in parallel, a 90-to-120-day continuous-flow test on an enhanced geothermal well, chasing questions about “thermal breakthrough and water loss” that apparently still don’t have clean answers after years of hype. Different technology, same instinct: run the slow, boring test before you scale anything. After a month of turbine backlogs and gas-plant scrambling, it’s a nice change to watch someone check their homework first.

Sources read for this entry