A turbine at sea normally stands on the bottom.
Drive piles into the seabed, bolt a tower to them, and every load the machine generates goes straight down into the ground.
Past a certain depth that stops working.
Below roughly 200 feet of water the foundation gets absurd, so the tower has to float and the whole problem becomes a naval one.
Depth is not the only reason.
The strongest and steadiest wind sits well offshore where the water is deep, so a floating platform is the only way to reach it at all.
Floating means the hull is the structure.
These eleven machines sit on concrete hulls that were cast on shore, ballasted, fitted out and then towed out complete.
Why a spar hull stays upright
Stability comes from geometry, not from stiffness.
A spar is a long thin cylinder that floats nearly vertical, with heavy ballast at the bottom and buoyancy spread up the shaft above it.
The weight sits below the lift.
Push the tower sideways and the hull tries to rotate, but the ballast underneath pulls it back, so the whole assembly behaves like a weighted float.
Concrete suits that job unusually well.
It is cheap, it is heavy exactly where the design wants weight, and it can be slip formed in a shipyard basin without the specialist steel fabrication a steel hull demands.
What was actually towed out
Each hull is a tower in its own right.
The concrete cylinder runs about 351 feet tall, roughly 59 feet across at the base and tapering to about 26 feet at the top.
The machine above it is large.
Each rotor spans about 548 feet across, with three blades of roughly 267 feet, sweeping an area of some 235,000 square feet of open air.
Assembly happened in sheltered water.
The tower, nacelle and blades were lifted onto each hull inside a fjord using a land based crane, which is the step that normally needs a very expensive vessel out at sea.
The water underneath is deep.
The site sits in 850 to 985 feet of water, with the units held in place by mooring lines running down to anchors several of them share.
The numbers the project is judged on
Total capacity is 88 megawatts.
Expected output is about 384 gigawatt hours a year, which works out to a capacity factor near 50 percent and is very high by any onshore standard.
The demand it serves is unusual.
The power goes to five offshore oil and gas platforms across two fields, covering roughly 35 percent of what those installations use in a year.
The emissions case follows from that.
Displacing gas turbines on the platforms cuts about 200,000 tons of carbon dioxide and around 1,000 tons of nitrogen oxides annually.
Who actually paid for it
The budget did not hold.
An early estimate of roughly 5 billion krone finished at 7.4 billion, which is close to a 50 percent overrun on a project with eleven identical units.
Then there is the funding split.
A state energy agency put in about 2.3 billion krone and an industry emissions fund added 566 million more, which together is roughly two fifths of the total.
That is the sentence worth pausing on.
Public and levy money covered a large share of a wind farm built to cut the emissions of oil production, and the operator published both figures without ever setting them side by side.
Scale is the other reality check.
Eighty eight megawatts is a demonstration next to 704 megawatts of fixed bottom capacity delivering into a normal grid.
What it proved and what it did not
The cost curve did move.
Cost per megawatt came in about 35 percent below the operator’s earlier floating project once inflation is taken out, which is real progress on a technology this young.
The concrete worked.
Casting hulls on shore and towing finished machines to position removes the offshore heavy lift that makes floating wind expensive, and that method is the transferable part.
The schedule held together.
First power arrived in November of 2022 and the last unit came online the following summer, which is quick for eleven machines of a type nobody had built at this size.
Durability is still open.
Nobody has watched a concrete spar age through decades of North Sea winters, and the industry is still learning what it does not know about one blade at a time.
The reporting is thin too.
Clean annual production figures have not been published in the way a listed utility would publish them, and the record leaves that gap open.
Which leaves the honest reading.
Eleven turbines on concrete hulls cut a real amount of carbon, and they still leave 65 percent of those platforms burning gas.
