Author: Derek Michalski, Chief Editor.
Offshore wind’s move to High Voltage Direct Current transmission is usually explained as a simple physics fix: AC struggles over distance, DC doesn’t, end of story. That’s broadly right, but the industry rule of thumb is less precise than it’s often made to sound, and the harder problem sitting behind it — getting equipment from different manufacturers to operate safely on the same DC network — gets far less attention than it deserves.
Why AC Runs Out of Road
Subsea AC cables generate reactive power because of the capacitance between the conductor and the surrounding seawater and insulation, and that reactive component grows with cable length. Industry studies generally put the point where compensation costs start to erode AC’s advantage somewhere in the 50–100 kilometre range, depending on voltage and cable rating — one published case study of a 1.2GW AC link at 80 kilometres needed three large shunt reactors plus a 375 MVAR STATCOM just to keep the line usable. HVDC removes the problem rather than managing it: because current doesn’t alternate, there’s no reactive component to compensate for, and transmission losses stay low across distances of several hundred kilometres. That’s what has made deeper, farther-offshore lease sites in the North Sea, the Celtic Sea and off the US Atlantic coast commercially viable in the first place.
What the Hardware Can Actually Do Today
Modern HVDC links are built around Voltage Source Converters, which use IGBT switching to control active and reactive power independently — a real advance over older line-commutated converter technology. VSCs are also theoretically capable of “black-start” — energising a collapsed onshore grid from offshore wind rather than waiting for conventional generation to come back online — but this is not yet a proven, commercially deployed capability. It’s currently being tested through live trials such as Britain’s Ofgem-funded SIF BLADE project, involving National Grid, SP Energy Networks, SSEN Transmission and developers including Ørsted, RWE, Equinor and bp, which is in its Beta phase through September 2027.
On cables, Prysmian’s newest 525kV XLPE submarine cable system is rated up to 2.5GW per bipole — nearly double the capacity of the 320kV systems currently in service, and roughly comparable to the output of two large nuclear reactors. Combined with “hub and spoke” platforms, where one offshore converter station gathers power from several wind farms and distributes it across multiple onward links rather than running a dedicated cable from every farm, this is what makes a meshed offshore grid — multiple wind farms and multiple countries sharing an interconnected web of HVDC links rather than isolated point-to-point cables — a realistic planning goal rather than a diagram. Britain’s National Energy System Operator has already built its Holistic Network Design around something close to this, coordinating infrastructure for roughly 23GW of the offshore wind capacity needed to hit the government’s 50GW-by-2030 target, specifically to avoid redundant single-farm connections.
The Part That Doesn’t Show Up in the Diagrams
Here’s what a meshed grid actually requires that a point-to-point cable doesn’t: converter stations built by competing manufacturers coordinating, in real time, on voltage control and fault response. Historically, HVDC projects have used one vendor’s converters at both ends running proprietary control software never designed to negotiate with anyone else’s system. That’s manageable on a single link. It becomes a genuine risk on a shared DC hub, because DC faults propagate almost instantly — without the natural current “zero crossings” that make AC faults comparatively easy to interrupt — and converters that don’t respond in a coordinated way can turn a single cable fault into a wider outage across the hub.
This is being worked on, not just theorised about. InterOPERA, a Horizon Europe programme running from January 2023 to April 2027 with roughly €70 million in funding (about €50.7 million from the EU), brings together Siemens Energy, Hitachi Energy, GE’s grid business, transmission operators including TenneT, RTE, Energinet and Statnett, and developers such as Ørsted, Vattenfall and Equinor to build standardised functional specifications so converters from different vendors can be certified to work together. Britain’s National HVDC Centre has run a parallel, more hands-on version through its Aquila Interoperability Package, centred on the planned Netherton Hub (formerly Peterhead Hub) in Scotland: rather than requiring vendors to disclose proprietary designs, manufacturers submit “black-boxed” real-time models of their equipment for joint stability testing. That work produced what the centre describes as the first successful multi-vendor, multi-terminal HVDC interoperability demonstration in December 2024, followed by a public demonstration in March 2025.
The closest thing to a working precedent is Caithness-Moray-Shetland off the north of Scotland, delivered by Hitachi Energy as Europe’s first regional multi-terminal DC grid, linking three ±320kV converter stations. It’s a single-vendor system — which is a large part of why it worked — and it isn’t yet proof that a multi-vendor hub of the kind the Holistic Network Design anticipates can be built the same way.
A Tighter Bottleneck Than It Looks
The supply chain constraints on the hardware side are real and well-documented: three companies — NKT, Prysmian and Nexans — reportedly account for around 75% of the HVDC cable market, and reported backlogs run into the late 2020s, with at least one major facility said to be booked through 2028. But cable capacity can be expanded with capital and new factories. Getting a Siemens Energy converter station to safely hand off a fault to a Hitachi Energy or GE station on the same grid requires years of joint standard-setting and testing first — and based on where InterOPERA and Aquila currently stand, that work is running on a longer and less certain timeline than the manufacturing expansion it depends on.
Where the Bet Actually Gets Tested
This isn’t hypothetical for some future grid — there’s already a specific site where it has to work. The National HVDC Centre names its own primary real-world use case for the multi-vendor work: Netherton Hub, near Peterhead, where SSEN Transmission is building a 400kV substation, a 132kV substation and HVDC converter stations to bring together three separate links — Spittal to Peterhead, Eastern Green Link 3 and Eastern Green Link 5. Main construction started in June 2026; SSEN has not published a completion date. Eastern Green Link 3 alone is a 2GW, 525kV bipole between Peterhead and Lincolnshire, and SSEN says it exists because system studies for the ESO’s “Pathway to 2030” Holistic Network Design confirmed the need for it — meaning Netherton Hub isn’t a demonstration site bolted onto the meshed-grid plan. It’s inside it.
The coordination work behind it is real, and still small. Aquila — now running as “Aquila Lite” — began as a £2 million project from 2022 to 2024 and has since been extended by a further £450,000 through 2026. What that funds is monthly meetings between engineers from GE Vernova, Mitsubishi Electric, Hitachi Energy and Siemens Energy, working through black-boxed models of each other’s converters to agree how they’ll respond to the same fault. It has produced the first successful multi-vendor, multi-terminal HVDC demonstration, in December 2024, and a public repeat in March 2025 — genuine firsts, and still demonstrations rather than certified commercial specifications a hub can actually be built to.
InterOPERA, the larger EU-funded effort running alongside it, is more explicit about the distance left to cover: its own roadmap describes standardised, competitively-procurable multi-vendor HVDC systems as “real-life applications anticipated by 2030.” That’s the same year the UK’s 50GW offshore wind target — the target Netherton Hub’s own links were sized to help meet — comes due. The cable shortage resolves on a schedule investors can underwrite: new factory capacity, opening through the late 2020s, at a cost NKT, Prysmian and Nexans can already name. The interoperability problem resolves on a schedule its own architects still call “anticipated,” attached to a hub that’s already being poured in concrete before the standards it will eventually need are finished. That gap is the harder, quieter story — the one that doesn’t show up in a capacity chart, and won’t be visible at all until the day Netherton Hub tries to bring three different vendors’ converters onto the same fault.
