Janet Wood talks to Spencer Thompson, chief executive of Eclipse Power Networks

A typical map of the UK’s low voltage distribution network owners (DNOs) shows a handful of companies that operate monopoly licence areas covering the whole of the country. Zooming in would show a more nuanced picture, with thousands of small areas of network – industrial estates, multi-use areas like airports, office groups, even new housing – served by so-called ‘independent DNOs’ (IDNOs). This group of companies – currently numbering around 20 – provide ‘last mile’ connections to new developments, but also often own and operate small network areas under long term licences issued by the energy regulator Ofgem.
Spencer Thompson is the chief executive of one such company, Eclipse Power Networks, which is ultimately owned by the Octopus group.
I first ask him about how changing homebuyer expectations in the home fabric (such as solar panels, electric vehicle (EV) chargers, power showers, heat pumps and induction hobs) are altering the business for IDNO connections. He acknowledges that change, but says at the moment the major problem for housebuilders, as well as other new developers, is simply trying to get connected to the grid.
Thompson agrees that expectations are changing in new-build housing, but says even without those bigger loads, the electricity connection – which is a small part of the cost to build out a new housing estate – can be a major blocker to development. For example, “We were talking to a customer the other day about 4000 homes. They needed a 20MVA supply from a local DNO. The DNO said the connection would be delayed by six years because of transmission reinforcement upstream. More and more housing developments are being delayed by getting connections to the grid.”

More and more housing developments are being delayed by getting connections to the grid

It has long been a problem for solar and battery projects, he says, and he highlights other new facilities, especially EV point installations which, “in the last six months has increased exponentially. There are so many developers out there trying to develop EV connections,” serving fleets such as delivery companies as well as providing public chargers.

Microgrids needed
He says the answer, for forward thinking developers, is to move on to a smarter approach to the connection. In response to grid constraints, smart developers are looking to create a microgrid behind the meter, with community batteries, solar and other assets that work together to meet the site needs with least possible interaction with the grid. The microgrid “could be community owned assets, or it could be developer owned, it could be a mix of both”.
He admits, “They’re realising that the grid connection, even though it’s a small cost item, is absolutely constraining, but we’re only really seeing those proactive, smarter developers do that”. Nevertheless, “That’s the thing that’s really taking off at the moment for us. And we’re specializing. We are launching a sister company in private networks and microgrids this year for this very demand.”
I ask him for an example, and he says a new development like the 4000 houses mentioned is usually a ‘five-year plus’ development that is built and connected in batches of 100-500. But at the moment “sometimes they can’t even get that first connection, and that’s becoming a more and more common issue. So the more you can do behind the meter the better”. He says that one development initially sought a 20MVA connection, but using a microgrid “you could reduce that down to a 5MVA connection. “If you can reduce the power demand, subject to a few technical issues to solve, you’re going to reduce the cost to connect to the grid, which isn’t necessarily the big driver, but you can reduce the lead time.”
He adds, “The other angle to microgrids is that there is a revenue model here behind the scenes. It can be developer owned, it can be a network company like us, or it can be community owned by the local residents, or it can be a combination of those three.”
But Eclipse currently specialises in industrial and commercial (I&C) networks and here, he says, “you could have even bigger wins. If you imagine industrial parks full of solar, with a battery, you could absolutely reduce the demand [on the grid].”

You are unlikely to go off grid, but could you get to a zero bill scenario?

Eventually a microgrid could be “virtually off-grid” or stay connected and alternately buy energy from the market or supply it back. Thompson says that depends on the ambition of the microgrid and the developer. “You are unlikely to go off grid, but could you get to a zero bill scenario? How can you maximize your solar and your battery?” EV chargers multiply the energy at play, and “if you really want to get ambitious there is the vehicle to grid or vehicle to business angle.”
It is a smart network future that has been in the background for government and industry for a long time. But at the moment, Thompson acknowledges, there are a few barriers to pass. Some are around governance: in the residential sector, “there are some regulatory restrictions around how much microgrid or how much private network you can have, which does need unlocking by the regulator and the various distribution code changes”.

De-clogging the queue
There are technical issues as well that apply to any connection. “There’s a lack of monitoring on the DNO networks as well as the IDNO networks,” he says, which means DNOs and the connecting party have to take a cautious approach (considering peak demand and a 40year+ asset life) to the size of connection needed, which is “clogging up the grid”. He sees “lots of connections that are asking for 2,3,4-8 MW, that when you model it only need a quarter of that”.
DNOs are adding more monitoring but it is very slow, he says – not surprising when DNOs have thousands of miles of network and customers would have to fund the installation. Thompson says, “DNOs won’t do it, necessarily, because they are not currently incentivised by the regulatory model. It needs a joined-up approach, and it needs a steer from Ofgem and central government.”
That would require a directed approach. Thompson says, “you prioritise hotspots, for example in the urban areas and the cities. But I think there needs to be more to prove the concept that this would work to unlock unused capacity, reduce the grid capex and new build which is clogging up the queue.” Innovation projects are under way, but they are small scale and “they’re often quite slow moving, because the main focus for utilities [DNOs] is around their CapEx and OpEx plans, and RIIO [ie five-yearly agreed business plans.”
Thompson is keen to talk about the microgrid option as a new business model. He says an IDNO’s return comes entirely from distribution use of system (DUoS) charges, while DNOs also get income from the return on their regulated asset base and connection charges. “So, we have to be lean and agile. But if you have a microgrid behind the meter with renewables, it does reduce power consumption, which reduces DUoS and drives towards net zero.”

the DSO should have the ability to create local solutions for a small urban area or a small hotspot, and it should have powers for that

Thompson is in favour of the DNOs’ new responsibilities in local markets as distribution system operators (DSOs), saying it will help bring forward these types of solution. He says, “I call it devolution. I support the DSO thinking. You’ve got the NESO that’s going to become a government entity. They’re going to try to be the impartial arbitrator and set the policy for the UK grid, which is a really good move. But you also need DSOs locally. DSOs know their region much better than anyone else. If you drill down back to your hotspots, your heat map, the DSO should have the ability to create local solutions for a small urban area or a small hotspot, and it should have powers for that.” The regime is still emerging “But you’ve got to promote local thinking. If you look at France and Germany, as an example, they have hundreds of distribution companies, often township or mayor led. There’s pros and cons for that, of course, but local solutions are happening more quickly. And I think it is a trigger to promote microgrids, which ultimately would reduce all the grid CapEx that needs to be spent.”
Thompson acknowledges battles with incumbent DNOs but says we have “some really good relationships.” He says, “We’re in the middle between the DNO, and the customer. And also we’re trying to challenge policy, which sometimes can be outdated. We spend a lot of time trying to translate the grid issues to the customer, and vice versa. They have millions of customers. We have tens of thousands. So, we’re able to pick up the phone and talk to them about the situation – given our agility.”
Again, he has an example, of a new build customer expecting to connect in 2026. DNO modelling with the transmission company put that date back to 2029, which was “completely unexpected at the time. And you think about this customer that’s building a solar farm and battery storage. They have a huge amount of time and money invested. Changing it to an IDNO connection we challenged some of their policies, the technical standards, and tried to broker what it is that the customer wants, versus the DNO assumptions that were not appropriate.” He also has examples where “the DNO were proactive with working with us and the customer to get a solution. So, you can do it at a local level. The local knowledge is important because technical standards are not consistent, even between licence areas within a company – an issue that is being very slowly resolved over many years.”

Change the model
Thompson’s general complaint about the connection queue is that “Every single connection was modelled worst case, which means you need to spend vastly more grid CapEx and the delays in the queue are equally extended across the country. They’ve changed the modelling recently, to help that situation. But they are still not modelling on what the customer wants. If they did that, they would improve the situation again.”
Some of the DNOs have done this really well, giving projects a really detailed profile of curtailment that either killed it or gave it a lifeline. Microgrids would come in to optimise such a project.

developers are beginning to work with data centres and even green hydrogen, balanced behind the meter in a microgrid.

In fact, large energy companies have tended to move towards energy parks, rather than single assets, with solar and battery providing several revenue streams, says Thompson. He says “now developers are beginning to extend that to work with data centres and even green hydrogen, balanced behind the meter in a microgrid. That’s going to be one of many keys to unlocking the connection queue over the next five to 10 years.”
He thinks the new energy system operator NESO can help make that a reality. “If we were more joined up regarding where solar, battery, hydrogen and data centres were built around the country, to suit the network as well, then that would help enhance the network in a more efficient way. At the moment, the rules are you can connect anywhere. Which means all you’re doing is creating this hugely inefficient network. But you could control within reasonable rules, but still keep some flexibility of where these technologies connect, where the network gets the benefit.
NESO could create greater markets to do this: they have schemes like Pathfinder, where they’ve got voltage issues, or system inertia issues. They targeted procurement events to connect that technology and they got a very competitive bids to build out those assets.”