Electricity Across Oceans: Is HVDC the Future?

How can we connect power grids across long distances or across seas and oceans? The answer is high voltage direct current, or HVDC, interconnectors.

In this video, I’m going to explain how these differ to regular AC, alternating current transmission, the recent technology advances that are causing HVDC interconnectors to start popping up all over the place, and what challenges remain to be solved if we are going to connect places as far flung as Europe and North America.

00:00 Intro
01:08 Why do we want to connect different grids?
01:35 The classic question of AC vs DC
02:29 Types of Transmission Line Losses - Resistive, Inductive and Capacitive
04:19 The Different Layers of an HVDC Cable
05:20 HVDC Projects around the globe
05:25 ElecLink
06:23 North Sea Link
07:13 Basslink Interconnector and Marinus Link
09:15 Sun Cable
09:38 Xlinks
10:20 Technological challenges for these projects
12:15 The other, bigger challenge - Politics
13:12 Outro

Or for a one-off contribution you can support by buying a coffee ☕️ here —

The Engineering with Rosie team is:
Rosemary Barnes: presenter, producer, writer
Kevin Irman: research, calculations, assistant editor
Cowritten with John Bone

[Electroblud]

Hey, high voltage test systems engineer here.
This is an awesome explanation video! I love the representation of HVDC you are doing here. Economic reasons and technological feasibility have a great balance here.
You said the cables have a high capacitance due to their proximity to the ground or water. 4:10 While this is technically not wrong, the reason for the high capacitance is their outer, grounded sheath! These cables are under ridiculously extreme electrical stresses compared to open air transmission lines. Their insulation is quite high tech, but the sheath is a key feature. It homogenises the electric field inside the insulation, greatly reducing stress. Also, it keeps the electric field inside the cable, so the outside is (nearly) field-free. Very important as to not disrupt all kinds of stuff going on outside the cable.

[jimmyfahringer]

It's so nice to have an Australian lady to teach us about AC and DC.

[treecrusher]

We also have an HVDC across the ditch in NZ connecting hydro in the South Island to the population centres in the North Island.

[hmw-mstx]

Where I live (Manitoba, Canada) we have three HVDC transmission lines in operation. Most of our power is from hydro and the vast majority of it is produced in the northern part of our province on the Nelson River. Most of that power gets consumed in the southern part of the province and back in the early 1970's two HVDC lines were run north/south. These were known as Bipole 1 and Bipole 2. These two line were run side by side in the same right of way and in the 1990's a tornado knocked both of them down. In the early 2000's it became necessary to run a third line and for security reasons (both weather and terrorist) it was decided this third line should be run down a different route. This is were the story veers from engineering to politics. The best technical route (i.e. shortest) was down the east side of the province. However, the government of the day decreed that it would run down the west side of the province. No amount of reason would sway them and the engineers in the project remained quiet for fear of losing their jobs. The retired engineers from the original Bipole projects had nothing to worry about and there was a constant barrage of reason and logic aimed at the government in the opinion section of the newspapers of the day. They christened it the "Western Manitoba Heater" and pointed out that the extra 200 MW of line losses over the east side route would consume all of the power of the most recently completed dam. None of it worked and Bipole 3 was built on the west side for considerably more money than the east side route and will now lose an extra 200 MW in perpetuity. It's no wonder you see so few engineers in politics.

[kjlovescoffee]

Wouldn't it make more sense to connect the UK to Faroe, Faroe to Iceland, Iceland to Greenland, and Greenland to Canada? That avoids many of the problems associated with super long cables, as well as avoiding some of the deeper seas. For example, From St. Ives in the UK to St. Johns in Canada (the shortest possible stretch), the cable would be over 3, 000km long and have to be over 4.7km deep in some areas - most of it 3km or more.

By comparison, from Thurso in Scotland to Kirkjubomururin in Faroe (415km, depths mostly in low 100s, with the middle dropping to 1km), Eioi in Faroe to Faskruosfrorour in Iceland (450km, depths mostly below 500m), Isafjorour in Iceland to Kulusuk in Greenland (640km, depths mostly below 600m), and Narasarmijit in Greenland to West Bay in Canada (1000km, of which around 800km drops to 3.5km but avoids the mountain range down the middle of the Atlantic. Add around another 800km between Kulusuk and Narasarmijit around the coast because it doesn't look like Greenland have an electricity grid, and I don't foresee anyone building HVDC across the landmass.

This seems a lot more achievable, not to mention cheaper, than trying to lay a cable straight from the UK to Canada, and adds the benefit of Faroe's hydro and Iceland's geothermal along the way. Both, as well as Greenland, could probably turn themselves in to major energy players, with landscapes highly suited for hydro and pumped hydro. Well, in the summer, at least!

[manyinterests]

You have a great ability to explain highly technical concepts without using math, and at the same not trivializing the issues.

[bigutubefan]

The map showing your Trans Atlantic 'sub sea' cable route, omits the Republic of Ireland, through which your cable must go, if it doesn't take a huge detour.

[jylfarm]

Since 1992, Hydro-Quebec already operate a 1500 km of DC line. They are looking to get many more of them online soon. The current line is transiting about $1M in electricity, every day.

Regarding politic, it already a problem for Hydro-Quebec (Canada). The state of Vermont (USA) is dependent on HQ. Furthermore, The state of Massachusetts (USA) often consider that it will be imprudent to get more electricity from Quebec. New-York city is also VERY Reliant on HQ. HQ already supply some electricity to Ontario (Canada). All in all, if HQ were to shut down the delivery of electricity to New England, the NERC NPCC regions will likely crash.

That the reason HQ looks at Europe. HQ is 94% Hydro. It does want to develop the WIND resource too. So, with Europe, what HQ will like to do is become the"BATTERY" where it does purchase electricity cheaply on the European market... and resell some when the sun is not shining or the wind not winding enough. HQ has an impressive amount of dispatchable electricity with Hydro.

[NickCohn]

Always wondered why DC was being used for these long links while most long distant lines are AC. Thanks for the explanation!❤😊

[bigboy]

600 million for eleclink? That’s how much the feasibility study would cost in Australia

[jonathanlanglois]

10:45 You mention Montreal. I think that I need to correct one very important detail about Québec. We have very little solar power. Almost all of our power needs are provided for by giant dams located a thousand kilometers north of the population centers. There's also a number of wind farms. In eastern Canada, the potential for wind energy is greeter than solar. It probably would be the right place to connect an undersea cable given that the power grid is rather well developed in Québec. Eastern Canada has gotten a few upgrades with the lower Churchill project, and there is room for another dam to expand capacity.

[nasapps]

For years now, Norway and The Netherlands are connected as well ... North Sea. The Dutch also have an energy platform island in the North Sea, this allows connections with sea based wind farms (windmills connect to the platform, operational already!). The platform provides an opportunity to connect with the UK grid. Another Danish - Dutch platform will be built for more Dutch and Danish wind farms and to connect Danish and Dutch grids. Sea based connection platforms can be the interconnection nodes in cross sea networks.

[jimparrUtube]

Yes, geopolitical insecurities are likely to become more and more important issues to factor into undersea HVDC. While battery storage (from pumped hydro to various cell chemistries) is seemingly more expensive, I believe the cost/kWHr storage will abruptly fall year by year in the next half decade as newer and more cost-effective stationary storage plants come on line around the globe. Furthermore, it is MUCH easier to use such storage to smooth peak-demand on grids locally.
And last but not least, the massive and enthusiastic trend for home owners to go solar and also contribute to virtual power plants has been proven in the UK to remove massive stresses from ageing and (currently - no pun intended) inadequate electrical grid structures.
B.T.W. Seriously clear explanations in this one Rosie. Thank you.

[bobloblaw]

The graphic at 1:00 appears to show a connection between the Isle of Man and Marystown, Newfoundland, not between Liverpool and Montreal. And most of Ireland is missing.

[johnredford]

Great video! You mentioned connecting Montreal to Liverpool. That means connecting to Hydro-Quebec, a vast and not fully utilized resource. It generated about 225 TWh in 2022, an average of 25 GW. Quebec itself uses about 80% of that. HQ has been trying for decades to get links down into the US Northeast, but they get blocked by local opposition. E.g. there was to be a line across Maine down into Massachusetts that was strongly supported by the state of Maine, but delayed by a referendum paid for by a Florida natural gas company. It's not quite dead, but in a coma. The nominal reason was the effect of the power line right-of-way on Maine's forests. Running an undersea cable down to Boston and New York might actually be easier!

I also keep hearing about HVDC from Iceland to Europe to distribute geothermal power, but so far it seems more economic to use the power to make aluminum. HQ does that too.

[flotsamike]

In the winter you can get a few extra hours of solar power in your system going north to south. It's always a 12-hour day at the equator. If you're lucky enough to get some place on the other side of the equator you can get Summer sunlight all year round

[jaredspencer]

This is so cool. A quick note for 5:55, that €420m is income (revenue), not *net* income (profit). And the profit is what makes the system "pay for itself", not the revenue. If ASDA makes £1B in revenue, but it cost them £1.2B to make those sales (from the cost of buying the products, paying the employees, renting the buildings, etc.), then they've lost money, even though revenue is high. In the Getlink case, it still looks like their net income (profit) is very high, so the effect on investors in the industry is likely exactly as you say. But it would still take 3-4 years to pay for itself, even though that's a really quick turnaround relative to other industries. Cheers!

[Soothsayer]

thx. for the informational video. You said the DC power losses over long distance is ~3%/ 1000 km, what is the comparative loses to AC power for the same distance?

[LeftCoastStephen]

Thanks for the explanation, as a civil engineer my knowledge of electrical things is very limited.
As a Canadian, I’d like to point out that Montreal is no where near where you labeled it! It would be better to end that transmission line in Labrador as that is where the bulk of the hydro generation is.
There is currently a proposed transmission line from Eastern Quebec to NE US that is being block for running through wilderness areas. I wonder if an undersea from Labrador to Massachusetts would be feasible 🤷‍♂️

Thanks again

[Mladjasmilic]

Electrical engineer here:
Only benefit of AC is cheap and efficient change of voltage inside transformer.
Now, with modern DC-DC converters, they are almost as efficient.
Benefits are:
Grid does not have to be in synch any more, as all is DC, so whole world can be connected.
Power routing is easy, as you need to have a little adjustment to direct flow of power, by boosting voltage a little on power line.
Efficient home solar instalation, as you would not need to convert it from DC to AC for appliance use and just use 240V solar battery (or central tap 120-0-120 in split phase system).

Most device in home can run on DC power supplies:
Any swtich mode power supply (chargers, PC power supplies)
Stoves and other resistive heaters (like boilers)
Most LED and all incadesent lights

Devices that can be adapted:
Any device with universal motor (vacuum cleaner, blender), but you would need to give up speed control or change the circuit
Inverter air conditioners and fridged

Devices that would not work on DC and need am inverter to work:
Anything with old iron core transformer
Anything with an induction motor (fans, pumps)
Anything with a dimmer circuit (ligjts, vacuum cleaners)
Clocks dependent on grid frequency