Bridge power first, then the ground underneath it

There are two ways to put a large load on the map. One is to find power and build beside it. The other is to build, and bring the power with you.

The first is cheaper and it is why most campuses are where they are. It also means joining a queue you do not control, in a place chosen by a transmission planner rather than by the workload.

Xalant is planned the second way, and the whole power design follows from that one choice.

The ladder, and why it is a ladder

A site needs power on the day the hall is commissioned, and it needs cheaper, cleaner power for decades after that. Those are not the same machine.

Bridge generation is the machine for the first requirement. It is specified as an engine package with fuel on site, sized with redundancy, able to run a hall on its own with no connection to anybody. It is unremarkable equipment, which is the point: it is the part of the stack that has to be boring.

The ground is the machine for the second. Where the geothermal resource is there, an enhanced system is the plan; where it is not, a fluoride-salt reactor class is the plan for later. Both are slower than the hall, both are further from being permitted than the hall, and neither is allowed to be on the critical path for opening one.

Replaceable in place

The sequencing only works if the later generation can arrive without the earlier generation being torn out in a hurry. So the pad is laid out for both: the interconnection point, the switchgear and the cooling loop are specified once, and the generation behind them is what changes.

That is the design claim, and it is a design claim rather than a finished drawing reviewed by a regulator. Nothing described here is permitted, and nothing described here is generating.

Why the queue is the argument

The reason any of this is worth the extra cost is in somebody else's data. The Lawrence Berkeley National Laboratory publishes the interconnection queues for the United States every year, and the picture it shows is the reason an off-grid build can reach power sooner than a grid-tied one even though it pays more per unit to do it.

We do not restate their figures here, because the figures on this site come from one registry with a source and a fetch date attached to each one, and that work is not finished for this series. The link is in the sources below and it is worth reading in full.

Where this sits today

There is no Xalant generation anywhere. Xalant is still being formed, no site is built, no fuel contract exists and no reactor vendor has any agreement with us.

What exists is a sequence, a set of interfaces between the three machines, and four markets in the pipeline where that sequence is being costed. The technology pages describe each rung in more detail, and each one says which stage it is at.