Flat ground in southern Illinois, graded and staked, with a pile driver working down a row.
Posts go in, rails go on, modules follow. It looks like any other site.
Under it, roughly a hundred feet down, is an abandoned mine that stopped working around thirty years ago.
The interesting problem was never the void.
It was in a filing cabinet.
And it took 147 signatures to clear.
Why the deed is harder than the dirt
In old coal country the land was split in two a long time ago. Somebody sold the coal and kept the farm, or kept the coal and sold the farm.
Those became separate estates with separate owners, and the one underneath is generally the senior of the pair.
A mineral owner holds rights that reach up through the surface. Access to work the seam, and in the older instruments a right to let the ground settle over what has been taken out.
Put a permanent structure on top of that and you are building on land whose owner is not the only party with a claim to it.
So you get a waiver, and the waiver has to come from whoever holds the mineral interest.
Which is where a century of inheritance turns one signature into a genealogy project, because a single seam under a single field can be split among dozens of heirs who have never met.
What went in on the ground
The site covers about 1,050 acres in Randolph County, at Tilden, over a coal mine that was worked underground rather than stripped from the surface.
Rated capacity is 186 megawatts on the direct current side, expected to make roughly 319 gigawatt hours in its first year.
That is enough for something like 35,000 Illinois homes, on a total investment of about 345 million dollars.
Commercial operation started in November of last year, so it has been running the better part of a year already.
The tracker supplier and the developer worked up a foundation and tracker design specifically for the conditions here rather than taking one off the shelf.
Around 300 people worked on construction, and the project is expected to return about 20 million dollars in tax revenue and local investment across its whole life.
What the record actually documents
The 147 waivers come from the developer’s own account of the work, described as surface waivers secured from that many separate entities.
That is the hardest number in the story and it is the one that explains the rest.
The contractor’s own summary of the job is worth reading plainly. Building on a historic subsurface mine took coordination and problem solving you do not see on a normal solar project.
What has not been published is the geotechnical detail. No embedment depths, no refusal rates, no settlement monitoring plan.
The public record covers the deal, the schedule and the supply chain, and stops at the soil.
For a site whose whole novelty is what lies underneath it, that is a real gap rather than a quibble, and it is the part any engineer reading about this would want first.
Where the big acreage number goes wrong
A figure of 840,000 acres of mined Illinois ground has attached itself to this project. It does not match what the state publishes.
Illinois puts the land affected by surface and deep coal mining at over 200,000 acres as of the late 1970s, with more than 22,000 acres classed as problem acreage.
The million acre figure that circulates is a national one from the same era, not a state one.
Mining continued after that, so the current state total is higher than 200,000, but nobody should be quoting 840,000 without a source that says it.
The underlying argument survives the correction. Damaged ground is filling up with arrays precisely because it is not farmland.
Capped landfills are going the same way, with one national count putting the room left at 63 gigawatts.
What a mine site is really selling
Not cheap land. Mined ground carries a title problem, a settlement risk and an engineering premium that flat farmland does not.
What it sells is the absence of an argument. Nobody organizes against a solar farm on a closed mine the way they organize against one on good corn ground.
That is worth real money in permitting time, and permitting time is where projects die more often than they die on engineering.
The catch is that every one of these sites has its own 147, and no two title histories are alike.
Which means the model does not scale by copying, only by repeating the work parcel by parcel.
The mine is still down there, and the array above it is a bet that it will stay put.
