An installation vessel loaded 5 complete turbine sets in port and sailed 31 miles past the last German island, and the first machine it stood up was feeding the grid six weeks later

The deck is stacked like a shipping yard that learned to float.

Five nacelles in a row, blades racked alongside, tower sections cradled and strapped.

That is one run.

Corrosion regimes offshore are severe, but the surf zone is no gentler, and service intervals track design and duty rather than a mileage number.

The berth is in the Netherlands. The site is 31 miles north of the last German barrier island, and there is nothing between them.

The ship jacks its legs down, lifts itself clear of the water, and stops moving.

Then it has to put all five up before the weather turns.

Why five is the number and not one

A turbine set is not a delivery. It is a sequence that cannot be paused halfway.

The crane raises the tower, then the nacelle at 476 feet, then each blade individually onto a hub that has to be held still while it is bolted.

Every one of those lifts has a wind speed ceiling, and the ceiling for a blade is the lowest of the lot.

At 31 miles the trip out and back is dead time, so the economics live in the load count.

Carry one set and the installation vessel spends its life sailing. Carry five and the sailing amortizes across five machines, which is why the deck was designed around that number rather than the other way round.

The cost of that choice is exposure. A vessel loaded with five sets that hits a bad week is five machines late, not one.

What is standing out there

Each foundation is a single steel tube driven into the seabed, about 279 feet long, and there are 44 of them.

They carry no separate transition piece. The tower bolts straight onto the pile, which removes an entire grouted joint and a whole workscope at sea.

All 44 were in the ground by early November of 2025, months ahead of the first turbine.

The machines are 15 megawatt units. Hub at 476 feet, blades 377 feet long, a rotor sweeping a circle 774 feet across.

A blade that long is the reason the whole schedule bends around wind speed rather than wave height.

The site is exposed shelf water, which is exactly why it was picked and exactly why it is hard.

Nothing about the numbers is casual. A pile of that length is driven to a target penetration, checked for refusal, and then has to stand for a quarter of a century under a rotor that never stops loading it sideways.

The dates that actually happened

The first turbine went up on the eleventh of June in 2026, on a foundation that had been sitting empty since the autumn.

Electricity reached the German grid on the twenty first of July, six weeks later, from that one machine.

The operator’s chief executive described the project as progressing exactly as planned, with the first turbine generating electricity. That is the sentence, and it is deliberately flat.

Full commissioning of all 44 units is set for early 2027. The 660 megawatt first phase is not finished, and no source puts every turbine up before the end of this year.

The second phase adds 900 megawatts across 60 more units, with foundations from 2027, turbines in 2028 and operation in early 2029.

Together that is 104 machines and 1,560 megawatts, built out of four separate seabed sites merged into one program.

What distance costs after the ship leaves

Construction is the visible part. Maintenance is the bill.

A technician 31 miles out loses roughly 90 minutes a day to transit alone, which means bigger crew transfer vessels, earlier withdrawal decisions and longer turnarounds when access closes.

The salt exposure argument gets overstated. Corrosion regimes offshore are severe, but the surf zone is no gentler, and service intervals track design and duty rather than a mileage number.

What does scale with distance is every logistics decision that used to be trivial.

The counterweight is the resource. Exposed sites generally run higher capacity factors than nearshore ones, because the wind is stronger and steadier out there, and that gap is what pays for the access penalty.

Whether it pays enough is a site by site answer, not a rule.

What the cluster is really testing

The full build is aimed at about 6.4 million megawatt hours a year, enough for roughly 1.6 million households.

Ownership is split between the operator and a sovereign fund, which is how projects this size get financed at all.

A technology company has already contracted 110 megawatts from the second phase, years before it exists.

That forward buying is the underrated part of the model. Revenue is booked before the first blade of that phase leaves a factory.

The open question is what happens when clusters this large sit beside each other and start competing for the same air.

For now the first machine is running and the other 43 are waiting on the ship.