Skip to main content

A High-Wire Act: What It Takes To Bring More Wind And Solar Power To Cities

Tomas Kellner
November 24, 2019
Here’s a bit of old news: The world wants more renewable power. The tricky work of feeding it into our homes, schools and offices doesn't often make the headlines — but figuring it out is key to changing the energy mix.
Dense cities surrounded by expensive land are no place to build a wind farm; they work best in windswept seas and on vast, open plains. Likewise, a giant solar installation works best in the remote desert — not the most habitable place. That’s why energy experts are increasingly interested in building electricity superhighways that can ship renewable electrons efficiently across long distances to customers. It’s a difficult and expensive task that will take years and cost tens of billions of dollars in the U.S. alone.

And distance is just one challenge. Wind and solar power themselves are tough to pin down. Unlike electricity from gas-fired power plants, which can work around the clock, wind and solar aren’t generally available on demand. They depend on environmental factors beyond our control — the fickle breeze, cloud cover and the setting of the sun.

Further complicating matters is the issue of inertia. Although most people haven't thought about inertia since their last physics class, it’s a key factor that helps keep our lights on. That’s because today electricity comes to sockets in our homes in the form of alternating current, or AC. As the name implies, AC moves along a wire in a sine wave, changing direction — or alternating — 120 times per second between plus and minus in the U.S. (In Europe, it switches 100 times per second.) If the frequency gets out of whack a little, lights can start to flicker, oven clocks may slow down and machines may not work properly. Larger frequency disruptions may cause conventional power plants to disconnect from the grid as a way to prevent damage to their expensive turbines and generators — and lead to blackouts.

The frequency is set by spinning generators inside power plants. When huge generators inside conventional power stations supplied all of our electricity, maintaining this constant frequency was relatively easy. That’s because even if you take your foot off the gas, the inertia of the heavy generator will keep it spinning. But maintaining a constant frequency can be problematic when the wind suddenly stops blowing. Although wind turbines also have inertia, they typically grind to a halt much faster. It’s even worse with solar farms, which have virtually no inertia and stop producing power as soon as the sun disappears.

That’s why grid operators, as well as governments, are increasingly looking at another way to transmit electricity: high-voltage direct current (HVDC) transmission. Direct current, or DC, is much simpler than AC: It can flow in either direction at a constant plus or minus voltage. DC has been around since the days of Thomas Edison, but it gave ground to AC because, back then, it was hard to transmit efficiently over long distances.

But today, those technological challenges have been largely solved, and modern HVDC links can transmit three times as much power over the same transmission line corridor as AC. A 2018 study commissioned by the U.S. Energy Information Administration found that HVDC lines “have a number of potential benefits including cost effectiveness, lower electricity losses, and the ability to handle overloads and prevent cascading failures. These attributes mean that HVDC lines could, if properly configured, help mitigate some operational issues associated with renewable generation.”

In May, GE Reports visited Stafford, a town in the British Midlands where GE Renewable Energy’s Grid Solutions unit designs, tests and builds some of the most advanced HVDC systems. We sat down with GE Grid Solutions power guru Colin Davidson to talk about the technology. Here’s an edited version of our electrifying conversation.

 width= The world wants more renewable power, but the tricky work of feeding it into our homes, schools and offices doesn't often make the headlines. Figuring it out, however, is key to changing the energy mix. Top and above images credit: Getty Images.

GE Reports: Thomas Edison was a big fan of DC, but he lost the war of the currents to AC proponents Nikola Tesla and George Westinghouse in the 1890s. I thought the case was settled. Why are we still talking about DC?

Colin Davidson: Well, the truth is that DC never completely went away. Most power plants, whether they burn coal to produce electricity or use wind, generate AC current. For DC transmission, you have to convert AC to DC and then back again. In the beginning, this conversion was very difficult, but people kept trying. GE was actually very early in the game. In the 1930s, they built an experimental HVDC line using mercury arc rectifiers to convert the current and ran it 23 miles from Mechanicville to Schenectady in New York, where GE had its headquarters. But back then, the converters were still very expensive and GE didn’t see the potential of the technology. The first true commercial HVDC line didn’t happen until the 1950s. We started slowly building from there, and even a decade ago, HVDC was still quite a niche industry. It wasn’t really big at all. But in the last 10 years, the explosion in the number of projects around the world has been phenomenal.

GER: What happened?

CD: It’s a number of things. People want more power all the time, but it’s hard to build overhead lines for environmental reasons. One way to build new links is to bury the cables underground, and it’s much cheaper to do that with DC than with AC. Plus, of course, we’ve got renewables, particularly here in Europe, where we have a lot of offshore wind. Wind farms make a huge difference in the way electricity’s generated, but they are often in the wrong place. Germany’s a case in point. They shut down their nuclear power plants, which are mainly in the south of the country, but they have a lot of wind generation up north. They need a way to ship power across the country, and they’re building HVDC corridors to transmit some of that power.

In North America, it’s picking up too. The U.S. has three AC grids: the eastern grid, the western grid and Texas, which wanted its grid to stay out of reach of federal regulators by not crossing state lines. You also have [the Canadian province of] Quebec doing its own thing, too. So there you have four large AC areas in North America potentially operating at different frequencies. But an HVDC link can tie them all together and allow them to exchange electricity, for example.

GER: How does HVDC do this?

CD: HVDC converters can essentially manufacture and match the frequency at which the destination grid operates. Without getting too much into the details, the converter at the end of the line can essentially behave like a traditional generator with a power source that can be turned on or off very quickly. So when something changes, it can quickly compensate.

GER: I think we have to slow down a little.

CD: OK. So the voltage and direction of the alternating current switches, or alternates, back and forth at a set frequency described by sine waves. The HVDC technology we are developing here in Stafford can chop up the sine waves of the alternating current into smooth DC lines in the converter station at the beginning of the HVDC link. From there, the DC travels over a cable to the converter at the end of the link, where another converter rebuilds it into the desired AC sine waves that precisely match the characteristics of the destination AC grid.

GER: How do you chop the AC up?

CD: You use a device called a rectifier, which straightens the AC sine waves. The first such device was the mercury arc rectifier. That was the grandfather technology they used for the Mechanicville line. These rectifiers were based on technology similar to the glowing tubes inside old televisions, but scaled up. As the industry evolved, it embraced semiconductor devices, like thyristors and special transistors. The latest version — power transmission’s equivalent of a 4K TV set, if you will — is a semiconductor device called insulating gate bipolar transistor (IGBT). We use IGBTs to build up a system called a voltage source converter (VSC). These power electronics are much smaller and much more efficient and powerful than anything in the past.

You can think of a converter using VSC technology as a black box that allows you to break down and build up any sine wave you want. Inside the black box there are lots and lots of little individual VSC converters, controlled by a computer, that step up or step down a wave by 2,000-volt increments. It’s kind of like a lot of square pixels on a computer screen forming a circle, one step at a time.

GER: Two thousand volts? That seems like a lot! Wouldn’t the circle be too choppy?

CD: Two thousand volts in our industry is quite small. We have hundreds of these converters — we call them submodules — lined up in a series. The submodules are connected in a series to make “valves,” and six valves — two for each phase of the AC system — go to make up the complete converter. We can quite easily produce a sine wave of about 400,000 volts with lots and lots of little steps in it, so it’s a very good approximation of the sine curve.

 width= John Vodden (left) and Jeremy Snazell led the development of GE’s new HVDC test facility in Stafford, England. “…We built a torture chamber for valves,” Vodden says. “It’s all set up to recreate what life is really like and also to give that little valve a hard time.” Image credit: Tomas Kellner for GE Reports.

GER: Why do you call the converter a valve?

CD: For historical reasons. In the early days of HVDC, the mercury arc rectifier was actually a valve. The electric current would boil off a small pool of mercury at the bottom of the rectifier and the current would travel through the resulting vapor. The mercury condensed on the walls of the valve, which were kept a little cooler than the rest of the valve for that purpose, and was returned to the bottom in a continuous cycle — like water evaporating from the sea, forming clouds and then falling as rain on the land, which then flows back to sea. When semiconductors replaced them in the 1960s and 1970s, people wondered what to call the new things and the name “valve” stuck.

GER: What is the advantage of valves made from semiconductors?

CD: It’s similar to computer chips. They need a lot less maintenance and they don’t suffer from a phenomenon called “arc-back,” where a mercury arc valve would suddenly and undesirably start to conduct electricity in the “wrong” direction, which is a problem since they are supposed to be rectifiers and therefore only conduct in one direction. The rest of the system had to be tolerant of such effects because it proved impossible to completely eliminate them.

GER: When you talk about efficiency, do you mean the amount of AC you can convert to DC and back without losing power?

CD: Exactly. Today, we can build converter stations that are more than 99% efficient. Since you have two of them, one at the beginning and the other at the end of the line, it bumps down to 98%, minus what you lose in the transmission line itself. These losses depend on the length of the line; obviously the longer the line the more you will lose, but it is normally in the range of 1-5%.

GER: You mentioned the explosion in HVDC projects over the last 10 years and the fact that it allows countries to ship electricity over long distances and to synchronize AC grids. Are there any other benefits?

CD: Absolutely. HVDC is really good at integrating wind and solar power into the grid. There is now so much renewable energy in the U.K., we recently had an entire week without coal generation, which was a first. As I already mentioned, HVDC converters can simulate large generators and help stabilize grids supplied by renewables.

GER: Why does it matter?

CD: The trouble with renewable sources is that they don’t have what’s called “inertia.” For most of the last century, since the beginning of electrification, we relied on big, heavy, rotating generators inside thermal power plants weighing hundreds of tons and spinning at a constant rate. These generators had a massive amount of inertia, like a merry-go-round with lots of kids on it. This inertia meant that if somebody turned off a machine somewhere, the frequency at which the generators were spinning didn’t change measurably, kind of like a kid dropping off the carousel.

But as these power plants are gradually getting phased out — think of coal-fired plants — and replaced with much smaller wind turbines that spin according to different wind speeds, the grid’s inertia also gets smaller. This means that when some large device or a factory connects to or disconnects from an AC grid supplied mainly by renewables, the frequency is going to change much more quickly and make the grid much twitchier.

Low inertia also means that when some disturbance happens on the system — say, lightning from a severe storm — it can travel across wider parts of the country faster than before. Fifty years ago, lights would flicker within a radius of 30 to 50 miles. Now the same fault would be felt over double the distance.

Since VSC converters can produce an alternating current of any frequency, they allow you to bring the AC grid back to normal when the frequency stops dropping or rising.

GER: How long have you been working on HVDC?

CD: I’ve been working here for more than 30 years.

GER: That’s a long time! Very few people give a thought to how they get their electricity. What do you tell strangers on a plane when they ask you what you do? What keeps you coming to work?

CD: Oh, it’s the fact that every day you keep learning new things. You never know all there’s to know about HVDC. The subject is far too complicated and far too varied. That’s the reason you get people like me who’ve blundered into this industry by accident and stayed in it.

GER: By accident?

CD: It was my first job after university. I said, “Let’s see what it’s like and I’ll make the next step in one or two years.” I didn’t expect I would still be here.