Podcast: The Demand Connection Conundrum – part 4 of 4, with David Wildash, Apatura Power

00:01:01 – 00:01:12 – Pete Aston

Hello and welcome to another Roadnight Taylor Connectology® podcast. I’m Pete Aston, one of the engineers here at Roadnight Taylor, and I’m joined by my colleague Philip Bale and David Wildash from Apatura. So welcome. Nice to have you.

00:01:13 – 00:01:14 – David Wildash

Yes, thank you. Thank you for having me.

00:01:15 – 00:01:24 – Pete Aston

So, this is actually the third session that we’ve recorded now on demand connections and all the sort of ins and outs of that, that’s really sort of coming out the woodwork at the moment.

00:01:25 – 00:01:26 – David Wildash

Absolutely.

00:01:26 – 00:01:31 – Pete Aston

But before we get into the detail of that, David, if you could sort of just introduce yourself and Apatura, just give us a bit of a sketch of the company.

00:01:32 – 00:02:39 – David Wildash

Great, certainly. So, thanks very much for having me today. So, David Wildash, Chief Strategy Officer for Apatura. So, who are Apatura? We’re fundamentally a developer company; we’ve had a rich history in developing batteries North of the border, we’ve got 2.2 gigawatts of consented batteries, and we like everybody else waiting to see what happens through dates, through the reform process.

But very much we saw an opportunity 12, 14 months ago, you know, driven by demand for critical, digital infrastructure to look at how we could contribute; looking at our sites originating new land tracks, which was going to play into the data centre market. So, this is very much an area where we are really focused as well.

Our core proposition is to deploy batteries and data centres, and we’re quite unique. We’ve got a large footprint of assets in Scotland, or you know, across the central belt, and we’re looking to originate at this point in time, 2.3 gigawatts of demand specifically for data centers.

00:02:40 – 00:02:42 – Pete Aston

Brilliant. Okay, that’s a nice teaser.

00:02:43 – 00:02:43 – David Wildash

Yeah, absolutely.

00:02:44 – 00:03:00 – Pete Aston

Already I’m going, I want to know more. So, you’re obviously sort of, you, you’re transmission focused, aren’t you? So, what, what do you see are some of the key issues that have been coming up recently? And there’s a lot of them I think, but what’s some of the key issues that have really cropped up in this sort of demand space?

00:03:01 – 00:05:03 – David Wildash

Yeah, so I think if we step back and go, well actually we’ve been in a position where there’s been huge de-industrialization for the last 15, 20 years, and it’s not been an area that’s had to necessarily be thought of in terms of how do you connect major industrial loads at the transmission level. And then lo and behold, we’ve just seen this, you know, this huge rise in need and the desire for compute power, which is very kind of energy intensive – and that’s very much our model of making sure that these are positioned near to where the renewables are, hence Scotland.

But fundamentally, you know, be it data centres, be it large scale industrial processes, the system needs to change to put them on a level playing field with generation to make sure that efficient connection offers can be provided to make sure that we can get these onto the system driving that economic growth for the country that’s needed. So, there’s a real kind of need for change across this and just to put that in perspective, the numbers are huge – I think it was a tech UK report, I believe it was last year, but it was looking at currently 4.5 gross value add, 4.5 billion, I should say, gross value add from the data centre connection, data centre market. You roll that forward to 2035 and that’s suggested to be 45 billion. This is a huge opportunity, and we shouldn’t forget that we are in competition, not just, you know, us as a developer against others who are looking to originate new data centre projects in the UK but also against Europe, globally. These are, there’s a huge amount of capital that needs to be deployed, but they’re going to choose the areas where they can get connections quickly, resiliently, and indeed well as well, so we need to bring all those things together.

But to your point around demand, what needs to change? For me, it’s that equitability between how generations looked at and how demand is looked at.

00:05:03 – 00:05:31 – Kyle Murchie

And one of the things that you may well have found is that obviously there’s been a significant uptake in generation connections at the transmission system because the distribution capacity was used up, which then meant that they’d secured the bays, which then means when you try and go and secure transmission demand projects, if it’s standalone, you are going to obviously be connecting into new connection nodes until the reform happens to then work out which projects are ready and needed and where may there be bays for those projects.

00:05:32 – 00:06:50 – David Wildash

Yeah, and it’s an interesting point around, well actually how many bays do you need for these projects? Certainly with ours, you know, we started these as batteries and I should say, we see a huge amount of benefit in deploying batteries with data centers, wholesaler energy, our charge that the battery can provide potentially offset, but also around power quality, which is going to be a big issue when you bring in major demand onto the system as well. So, there’s lots of benefits from that. So, to start with, our connection offers had bays for the batteries, but when you look at actually the SQSS and what’s needed and what the data centre operators and hyperscalers want, they want resilient power.

SQSS above 300 or 350 megawatts, you need three substation bays. But in reality, when you’re talking to hyperscalers and data centre operators, creating a resilient connection and you know, getting access with a dual bay solution to one of the world’s most resilient networks that we’ve got here in the UK, and it should be sufficient.

So, we spend a lot of time making sure that we can provide that, making sure we can get dual bay resilience, but also looking at pathway to power these major demand sites as well.

00:06:51 – 00:07:12 – Philip Bale

I’m glad you said that around the SQSS compliance in terms of connections. So, your ultimate outcome for the majority of your schemes would be two 400kV bays connected in four data centres. Up to what sort of scale would you like to see GB say these projects are sufficient for having two bays connecting to these systems?

00:07:13 – 00:08:08 – David Wildash

So, it doesn’t necessarily need to be at 400 but yeah, 400, 275 is kind of where we’re looking at. And I do have big umbridge around, you know, the difference between England, Wales, and Scotland and what 132 looks like. But that’s another argument we may come back to. But in reality, I think as we look at the scarcity of substation bays across the UK, we need to make sure that they’re being utilized well. So, you know, putting in a 50-megawatt battery to a 400kV bay where you could get a lot more power out of that, that’s probably not the most efficient solution.

So, we are looking at Jewel Bay 275 or 400 depending on what it was but, you know, we are going up to, up to 550 megawatts of load in a single site. And, you know, talking about that one, that’s Ravenscraig, so that’s probably our kind of jewel in the crown in terms the portfolios that we’re building out currently.

00:08:09 – 00:08:53 – Pete Aston

So, in terms of some of those then key issues with the building out of size, what’s your view as a company on the issue with demand projects not being able to build out transmission assets? Because that seems to be like the key issue that’s being debated at the moment is that demand customers don’t have a license, unlike a generation customer, which has the license and they can build out transmission level assets. So, you’re reliant on the TO building out, transmission level assets for you and providing you with effectively a distribution level connection, whether that’s 132, or 33. So, you know, how are you sort of trying to dig into that particular problem?

00:08:54 – 00:10:58 – David Wildash

Yeah, so it’s, there’s a number of facets and how do we deconstruct this? So, at moment in Scotland, Scottish Power, SSEN will build to site. If you’re a generation, you know, such as a battery storage (we’ve got standalone battery projects), then actually they’ll build up to the isolators. And then as a generator, you can own the transmit, you can own the transformer, which steps down the power to connect your assets – that’s all very fine, that’s well understood.

Now, for demand, you can’t. And there’s a very simple solution in my mind is actually you just need to move where the isolators are and where the transformer is. So fundamentally it comes down to is a transformer classified as a transmission asset under the Electricity Act? And I think that’s a conversation that needs to be had and also just real guidance to get rid of this ambiguity of what is transmission, what’s not. So, I think as an industry we need to work with government and indeed a regulator to make that inherently clear.

So, there’s just a little switch, across a boundary. Then actually you can, you know, transport power at the voltage which is economical for sites and get through the issues that we’ve currently got of demand not being able to own transmission assets.

The other way of looking at, as we are co-locating, generational, BESS, and demand, well, actually you have a generational license for your BESS asset, does that count? So, there are a lot of open questions and ambiguity that needs to be kind of worked through. So certainly, we’re part of the conversation amongst the other guests that you’ve had on today; really think about, well, what are the opportunities, who needs to make the decisions to make sure that as an economy that we do not miss out on these huge opportunities that are ahead of us.

00:10:59 – 00:11:09 – Pete Aston

And what do you think about the possibility of demand projects being able to have a demand license and therefore, you know, and so build out transmission of other assets as well?

00:11:10 – 00:12:11 – David Wildash

So, coming from the, so historically I wear a heavy system operator hat, as long as this is looked at from a perspective of making sure that everything is compliant. I think grid code does need to catch up for the major demand sites such as this, and there’s no detriments to the system, then actually, why shouldn’t you know there be a demand license? Why can’t we replicate what’s already there for a generation license? The challenge which we’ve got is just really creating the capacity within the industry to spend the time to get this written down. But we are run at a breakneck speed, we’re going through Connection Reform, we’ve got a, yeah, once in a generation opportunity to bring a huge amount of capital investment into the UK and if we take too long and we’re too conservative, then we may miss the opportunity as well. So, it’s that kind of balance between risk and you know, UK PLC rewards here. So, I think we’ve all got to go down this route.

00:12:24 – 00:13:34 – Philip Bale

Obviously, I think there’s numerous different benefits that you could be said and pros and cons to any approach, obviously, one of which that if you own the asset, you can build it yourself. You can specify yourself; you may well be able to build it quicker than the TO could build it. I’m guessing one of the things, and one of the advantages that you would look at is that if it is your asset, potentially you could step away from the same SQSS that the TO would do; you mentioned earlier around the 300, 350-megawatt threshold in terms of what’s needed.

Is there also an element of looking at it, of going, well, actually for some of these industrial commercial projects that have the UPSs, that have the generation backup, even if we don’t get to the stage immediately of saying these should be transmission owned systems, there should be that little asterisk in the SQSS, which says, this is not domestic, this is not small scale, these are big industrial commercial customers with the backup generation that provides the extra level of security supply; Some of which would happen at the P2 level conversations where there are examples where generation has been used to avoid putting in extra circuit.

00:13:35 – 00:15:12 – David Wildash

Yeah, I think it’s all part of the conversation. So, you know, there are design variations within the SQSS you need to look through, and I think there is a joined-up conversation absolutely there; if you think about where data centres have been, you’ve got availability zones, which gives a further level of resilience so they can transfer a load between each of, you know, a trilateral arrangement of data centres. Plus, each of those has dual resilient grid supply points, plus they’ve got an uninterruptable power supply as well. So, I think there needs to be a joined-up conversation with the end customer, you know, what is the level of resilience that is acceptable? Is there an interim glide path where the wider infrastructure needed on the grid to support these assets catches up? So, there’s a pragmatic conversation to be had on a site by site basis, but that all starts with the debate, the conversation, being able to engage both with NESO, with the relevant TO, and indeed Ofgem and the government to come together to go, this is the way that this site is going to operate – we’ve done all the checks, we want to be a quality participant in the system. So, we’re at least net neutral, and hopefully net beneficial, particularly when you think about the flexibility that can be provided by major demand when you put it with a battery. This all needs to come through to the conversation, how can data centres and other large demand sites be positive to the system. And I’ve not even started on the curtailment in Scotland, but that’s another conversation.

00:15:13 – 00:15:45 – Pete Aston

Well, I was just sort of going to touch on that very briefly, which is that there’s quite a lot of call for strategic demand as part of the sort of NESO work and it does seem that the place to put a loss of that strategic demand is in Scotland because of the constraints across the sort of the B6 boundary between Scotland and England. Is that something that’s sort of playing a role within, you know, site selection and so on, that you are looking at, or is that you’re just going, that’s a NESO question, you know, let them think about that?

00:15:46 – 00:17:13 – David Wildash

So, it’s something I think about constantly, I see it as part of you know, dare I say, the uniqueness of the sites that we’ve been originated. Again, going back to where my old school, you know, system operator hat, it is a slam dunk of an idea of putting demand where you have excess renewables and soaking up that excess renewables to the benefits of every end consumer within the UK, and we shouldn’t forget that. You know, the cost of curtailment, me, you, everyone has to pay their share of that. So, if there’s a right incentive to put demand where it’s going to be a net positive from a system perspective, reduce curtailment costs, that’s surely got to be a good thing.

So, I think we’ll need to see more, particularly now that zonal has been taken off the table. What does reform national pricing mean and how can we, where are we make sure that system operability is thought about when site selection comes through. The strategic element though, Pete, I think is really interesting because this is where we also need real leadership from the government to dictate and show where the industrial strategy truly meets. Anywhere where rubber hits the road from an industrial strategy perspective, there needs to be that signal of where the best place and optimal place for community is, for economic benefits, for system benefits, open demand.

00:17:14 – 00:17:39 – Philip Bale

And obviously from a technical perspective, very much agree with you – if you put the more demand closer to the generation, one of the ways of helping to decarbonize the energy, and especially if you have the flexibility with the battery storage as well, to try and make best use of energy prices being low in arbitrage, how are you finding the conversation with the data centre operators and dragging them away from their Slough, West London sort of areas in terms of migrating upcountry?

00:17:40 – 00:18:48 – David Wildash

Yeah, so you, you’ve seen a natural kind of move away from, you know, Slough, West London and, you know, what is the flap D markets? All the flap D markets and in the UK and London is curtailed, and what I mean by curtailed in terms of getting a good, viable, timely grid connection. So, there’s that kind of push away from that area, but there’s also a pull from actually where can we find great brownfield sites, where is the land, where are the renewables, where are the potentially early connections as well? So that there’s a pull away to those sites. The other big change is actually, well, what is the compute load being utilized for, you know, latency was, was where the nascent data center industry started in the UK, needed to be close to the city of London for low latency requirements, but now when you’re looking at AI, large language models, et cetera, then there is the opportunity to be a little bit more agnostic in terms of location, so it then comes a cost of power, time to power, resiliency of power question.

00:18:49 – 00:19:00 – Pete Aston

Interesting. We’re coming towards, I think the end of our time sort of slot for the podcast. Philip might have a question that he wants to ask, I don’t know. He has always got a few questions to ask.

00:19:01 – 00:19:02 – David Wildash

Great. So here comes the killer question.

00:19:03 – 00:19:12 – Pete Aston

But so, while Philip’s thinking about that, where do you think you would like the industry to be within sort of six months from now?

00:19:13 – 00:20:03 – David Wildash

Yeah, I think the, I think the easiest thing would be to get absolute clarity on the Electricity Act and what is transmission assets, because given that level of clarity unlocks a number of the other models, and I think it’s pretty simplistic and I say simplistic not being a lawyer and a casey and everything else, but I think that is a shortcut – you don’t need to look at any sort of primary legislation changes, introducing new demand licenses, et cetera. So, I think that’s probably the quickest way that we can look.

If you didn’t do that, then it’s, you know, transmission license exemption. There’s various other tools, but they all feel, they’re probably harder to get to. So, I think that’s probably the quickest, easiest one, which gives you the biggest bang for your buck, just given that clarity.

00:20:04 – 00:20:09 – Pete Aston

Yeah, interesting, thank you. Any final thoughts, questions, Phillip?

00:20:10 – 00:20:58 – Philip Bale

A quick one for me in terms of, obviously we’re looking at transmission connections, I would argue probably because you believe there’s too many barriers to entry in terms of the transmission assets being owned by the TO and you being a distribution or an embedded customer of that TO, and obviously one of the ways around it is to allow large demand customers to become transmission owners, operators, and effectively be transmission customers. Is there also not an argument to how do we fix the big problem, which actually means then that those transmission assets could be owned by the TOs and then potentially an extension of the existing assets, and is the answer to facilitate large demand customers to own those transmission assets or to try and fix the problems, which then means that these assets can then be weaved into the wider network?

00:20:59 – 00:22:07 – David Wildash

Yeah, that’s a good question. Yeah, you did have a doozy though, you really did. So, I think fundamentally, whoever is the ultimate owner of these assets, the grid is interconnected as we know it’s a network, everything has to be delivered to a certain standard, it needs to be reliable, resilient, et cetera. So, you’re going to have to underpin all this with the right set of rules, irrespective of who owns it. Talking about the different ownership models, though, it all comes down to pace, whatever way creates more competition and effectively capacity into the network supply chain can only be a good thing because that is going to mean there are physically more bodies building infrastructure that’s required. The ownership piece, it’ll all be owned by different parties in the end anyway, but they’ll still need to adhere to the same set of rules. So, whatever brings more capacity into the system as a whole. I’m a real big advocate of, I don’t know if that answers your question, but that might be the leaving thought.

00:22:08 – 00:22:24 – Pete Aston

I think so. David has been absolutely brilliant to have you with us, thank you so much for coming in. Thank you, Philip, as well. Thank you everyone as well for listening, there’ll be no doubt more to say on the subject of demand connections over the next six to 12 months, but for now thank you for listening and goodbye.

00:22:25 – 00:22:25 – David Wildash

Yeah, thank you a lot.

00:22:25 — 00:22:26 – Philip Bale

Cheers.