Podcast: The Demand Connection Conundrum – part 2 of 4 with Ryan Adams, Innova

00:00:00 – 00:00:09 – 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 Kyle Murchie and Ryan Adams from Innova, so welcome.  

00:00:09 – 00:00:11 Ryan Adams 

Thank you.  

00:00:11 – 00:00:40 – Pete Aston 

We are going to be talking today about demand connections. So, this is actually the first of three sessions we're doing, with a number of industry players on demand connections, which are an increasingly big issue, big, sort of, big part of the network – lots going on in relation to that. So, you've got lots of thoughts, Ryan, on this issue. Firstly, did you just want to sort of introduce yourself and Innova to us just so we can get a sense of what the company is and does?  

00:00:41 – 00:00:47 – Ryan Adams  

Yeah. Thanks Pete. Yeah, so I'm Ryan Adams, as you said, Managing Director at Innova, often referred to as Inverse. 

00:00:47 – 00:00:48 – Pete Aston 

Sorry! 

00:00:49 – 00:02:53 – Ryan Adams  

So no, pronunciation. 

And yeah, we are a developer of solar battery storage and now digital infrastructure as well. Our story is similar to a lot of developers in the industry. So, we started pre, in the subsidy days, then we went into subsidy free development. So, the business actually was founded by a merger between what was a solo developer called Novus, and an EPC company, which is still around, but not part of the group anymore.  

And then I was the first employee of that merger in 2015. And, like any kind of person in a startup, I kind of had every job within the business as we grew. And really when we started to, to grow significantly that also coincided with me taking steps in my career. So, I kind of founded another offshoot of Innova called Novus, again at the time, which is now part of the group again in 2019, and our role within that business was to grow the largest, pipeline of subsidy free developments that we could at the time we thought about solar pretty much exclusively, but obviously BESS came later.  

So that was really successful and that kind of leads to where we're at today. But to put some numbers on it, so we currently have one and a half gigawatts of DNO Solar. That's what we have now, but we've previously built, owned and operated, a gigawatt or so of solar, over the last 10 or so years laterally in a partnership with Schroeder's GreenCo, which is going extremely well. 

We have 5.5 gigawatts of battery storage at transmission level. 4 gigawatts of that is consented and some of that is in hybrid sites with the third category, which is 4.4 gigawatts of data centers. And we have transacted on an additional 400 megawatts of data centers recently. So, one that we've, we've sold. 

So, I think that summarizes broadly where we're at now, and that's why we're interested in looking at demand connections because we've moved partly into that space.  

00:02:54 – 00:03:13 – Pete Aston 

So, from Innova's point of view, what are some of the key issues then around demand connections? So, you've said that you've got quite a lot of transmission data center schemes, some of them are hybrid with BESS. So, what are some of the key issues that you are seeing then as a company, in that sort of demand connection space?  

00:03:14 – 00:06:23 – Ryan Adams  

Yeah, so two primary issues. But let's go back a little step to kind of explain how we got there and then it kind of elucidates why these are issues and why they're important to us. So, we have been the largest customer, of National Grid for about five or six years now, because we foresaw the growth in battery storage in the transmission part of the network and we foresaw, later on and more recently the demand for data centers – so, we are really early there. So, we are the large, we have the largest, best portfolio in the UK. We have the largest data center portfolio in the UK in terms of connections. And so how did we get there? Because we spent a long time thinking about where the market would go, but also a long time engaging with National Grid, NESO and NGET. And that means we have a particularly good understanding of how things work and hopefully quite good at seeing where things will go.  

So, the main issue that we wanted to talk about in the demand space was the way in which demand customers connect into the network and trying to match that with the way that generators connect into network. 

So, if you kind of, if you go back in time, there's a unique set of circumstances which have led to this discrepancy in treatment between the two types of customers. So, it was easy to foresee, and for a long time there's been large generators on the transmission grid, but there hasn't been large demand customers on the transmission network. 

And so, whilst there was a regulatory framework put in place for generators, often, you know, mainly in the, in the form of the generation license regime, there isn't a demand license as such, and there's no clear way for demand customers to connect at high voltage. 

So, at the moment, what that leads to is NGET being forced to build step down infrastructure for customers who wish to connect as demand customers like data centers now, batteries actually are considered generators, as you guys will know. And so, they are able to connect, but demand customers are not. So, what's the problem with that?  

There's two main problems. One at the moment, if you kind of, again, go back and look at where we are as an industry, we have the largest amount of strain in the form of new connections that we've ever seen in the transmission industry, on the transmission network, and that means that that is creating a undersupply of bays to be connected into substations across the UK. So, as you guys know, historically, you would've had an application every now and then big offshore wind farm, large power station, and there would always pretty much be space at the substation available to build a new bay. What we see now is so many applications and so much new development in this space that the substations around the UK are being filled up. So, at the time when they're being filled up, if we say as, as I believe that this is the kind of most scarce resource for the energy transformation that we're going through at the moment, we're wasting a lot of this space. Much more space is required to build step down infrastructure by National Grid for a data center that would be required for, say, a battery. And if you were to change the rules to make them operate in the same way and allow demand customers to connect at high voltage, all of that space would be saved. 

00:06:24 – 00:07:23 – Pete Aston 

Yeah. So, sort of just to step back to summarize the problem generation customers have got a licensed, sort of regime where, and they can build out transmission level infrastructure. So, 400 and 275 kV in England, Wales, 132kV and above in, in Scotland. But demand customers don't have that opportunity; so, they can only build at distribution level assets at the moment. So, in England, Wales, that's 132 kV and below and Scotland, I guess that's 33kV and below. And that's then causing an issue in terms of the ability of large-scale demand users, customers like large, sort of hyperscale data centers to build those assets. 

But there's, you're saying there's also that knock on impact of having multiple bays. So, so instead of a data center and a BESS being able to share, say two bays; the data center needs two bays, and the BESS needs a third bay. And so, you, you're sort of proliferating the number of bays required.  

00:07:24 – 00:08:26 – Ryan Adams 

Yeah, and it's much more than that because if you were just having your two bays, that would be fine because you're still only taking up, let's say, a 100% more space than a, a solar farm or a battery. But actually, you're getting your bays and then you're having National Grid step down with, 400 to 132kV transformers. And then building out, building all that infrastructure out, and then you then connect it 132kV as a customer and theoretically at least you then run those cables to your site and connect. 

There's obviously the separate issue of is it feasible to run multiples of this, of 132kV cables across, you know, the countryside through roads, et cetera, as opposed to two 400 kV cables? Much more difficult. But the, the main issue there is a, because substations are mostly full already if there's 10% of the space left, I could occupy that space as National Grid with say, three or four bays for customers who are in the queue already for clean power. If I use the same amount of space in the current regime for a data center, it may take up all of that space by itself, or there may not be sufficient space.  

00:08:27 – 00:08:29 – Pete Aston  

Because you've effectively got to build a grid supply point.  

00:08:30 – 00:08:31 – Kyle Murchie  

Yeah. 

00:08:31 – 00:08:37 – Pete Aston  

You know, on the side of the existing substation, which is, you know, a huge space and then compared to the just a single or a few 400kV bays.  

00:08:38 – 00:09:26 – Kyle Murchie 

It's worth also saying as well because I'm just thinking that you mentioned there about if it's the, TO building out the assets, so NGET building out the assets. The other issue there is SQSS might say, yep, we need three bays for that particular solution, but their transformer capability, you know, what they've actually got registered on their system is also another limiting fact. You know, if you were talking a gigawatt scale demand, well you might not have nice, straightforward, you know, 250 MVA transformers that can go down from 400 to 132, or 400 into 30kV if that's not available. So, we've, we have seen some offers that have ended up having four or even five bays not driven by SQSS, but then driven by the equipment that's already approved. Yeah, so it becomes quite, you know, that particular point becomes even further extrapolated.  

00:09:27 – 00:10:27 – Ryan Adams 

Yeah, and it is probably worth saying as well, that although the acute issue is obviously trying to connect as many projects as we can, you know, but that from a CP30 perspective, it's Clean Power, it's net zero – it's the goals that we all want. First and foremost, Innova is a renewables company we want to connect our renewables projects, but if we are wasting space, even though the catalyst for the wasting of that space might be a data center that's impacting everybody, it's impacting our renewables projects. 

If you are lucky enough as a data center customer to have just enough space to fit your data center in and get a reasonable connection, what about all the people behind you in the queue at that location? They're now missing out. So, it's, this is just about efficient use of resources. At the moment, the rules are not up to speed because as we said, they don't, they did it, nobody foresaw in the eighties that development would go in this way when they wrote the current legislation, and that's completely reasonable. What we need to do now is to look ahead and think um, okay, how can we best use these resources today? And obviously it's not the way that it works right now. 

00:10:27 – 00:10:26 – Pete Aston 

Yeah, so moving on to some solutions then, we've already talked, before we start the podcast about some of the sort of solutions that you are pushing and that have been discussed sort of as a wider industry. What are some of the key things that you think could unlock this issue for, for developers of like, you know, hyperscale demand? 

00:10:27 – 00:12:28 – Ryan Adams 

Yeah, I think and yeah again, also unlocking projects of all types earlier that would otherwise be delayed if we don't do this; and that's why it's a systemic issue. But so, there are two, well, there's one really easy step that Ofgem could do, which we've talked about a lot, which we've mentioned to DESNZ, and I think there is interest in this which is a lot of this goes back to the interpretation of the Electricity Act itself and what is determined to be a transmission network. And I think, you know, anecdotally, everybody in the industry is happy to say that, okay, of course, connecting in at 400 KV and running a cable to your own site is not being a transmission network – that's just common sense Ofgem could just simply state that in an open letter and say, yes, our interpretation is as follows; this is what we think the implication of that interpretation is, and therefore, no legislative change in the short term is needed. You can go and connect and that would solve lots of, lots of problems. 

Then there is a kind of much bigger, longer-term solution, which is also able to be done quickly, which is to copy effectively the generation license infrastructure or legislative framework and apply it to demand customers. So obviously some things will have to be different. When we've spoken to NGET before Kyle, they often say there are differences between demand customers and generators, and that's right. But those differences are not differences that affect the ability for them to have their own license, there are other types of differences that require rule changes elsewhere. 

So, you know, we don't need to reinvent the wheel. It's been invented before – it's called the generation license, let's just make another one.  

00:12:29 – 00:12:47 – Pete Aston 

So, summarizing those couple of options there. The both of them, both of those sorts of options you just said there effectively allow a customer to build infrastructure out to their site. So, that sort of transmission level infrastructure is still built, owned, operated by, owned by the customer. 

00:12:48 – 00:13:24 – Ryan Adams 

Yeah, and it's a really important point to make because I've had a lot of feedback from some in the industry that will say, well, did you know actually data centers need to step down to 33kV internal voltage, and so you're going to have to step down anyway, and I think that you will know, but all generators also step down to 33 or 11kV, or some other similar voltage. So I think it's important for any listeners who, aren't aware of that, that actually yeah, everybody steps down to, to lower voltage. This is just, a quirk of, of these old regulations that mean that they didn't know that for generators because they've never had to think about it before.  

00:13:25 – 00:13:50 – Pete Aston 

Yeah, so Kyle, I don't know if you could step in here because there is a slight quirk with the way that TOs operate at the moment, isn't there in terms of one of the reasons in England and Wales, why, I guess Ryan and other developers are talking about wanting to build and own their own transmission assets is because National Grid as a TO doesn't build out from their sites to customer sites, but it's slightly different in Scotland.  

00:13:50 – 00:15:33 – Kyle Murchie 

Yeah, exactly. So, from a generation perspective, if you are in England, Wales, then you know, as many of our listeners will know, then you will effectively connecting directly into a bay, either be given a kind of unequipped bay and a kind of skeleton bay and then you will equip it, or if it's gas insulated gear, then you know, the switch game might be there. You've ultimately paid for that as a connection asset, but you connect from there and then build out your circuit to your site. Whereas in Scotland, slightly different arrangement, not just with the 132 kV aspect as well, included in the transmission definition, but also that the TO will traditionally build out to you. So, you know, you'll have, there'll be the transmission substation, they'll then build out the circuit to you then give you effectively a bay and you'll then normally build out the transformer. So similar sort of approach, but you therefore you're getting a connection at a bay, but the bay is basically on your site. 

I suppose then when you apply that kind of demand concept, then effectively the only real difference in Scotland would be the question around the transformer, because they were already building out to your site already and who owns the transformer?  

Compared to in England, Wales, much bigger jump where they're saying, right, we now need to build out because you're saying, Ryan, we're talking at the moment about, you know, if it's a 132kV solution given, they're wanting to build that at their site. But the issue being, well, we already know there's not enough land at that site because they're already talking about another substation in the future – so that becomes a constraint. Whereas if you are saying, well actually build out the 400kV assets, which is in five to my site. We've got some land; you can then build up a step down and build that transformer there.  

00:15:34 – 00:15:39 – Pete Aston  

And does that help for developers? You know, could that solution be another option?  

00:15:40 – 00:17:18 – Ryan Adams  

It could be, and this is what we've been asking DESNEZ for as a, as a well, it gets you 90% of the way there. So, I was going to say as a first step, but it could be the whole step and, you know, it definitely solves the majority of the problems. 

It also solves another problem, which we have, which, is hybrid projects. So, at the moment, whilst we know it's binary for demand customers, you're either going to get 400kV or 132kV and be that as it may in terms of on the ground whether you can actually fit that in. And then secondly, whether you can even run 132 to your site or not, which is a separate issue. If you have a project like we have many, lots of CP30 protected projects which government wants to have built are sat connected with a data center, NESO and co permitted us to go and pair those connections together. Because of this uncertainty, they're now, stuck in regulatory limbo. So, we are not sure, maybe there's a way to connect them; at the moment, we're not sure whether or not we can connect either the data center or the battery without cancelling one or the other, which in of itself is, is a terrible thing because they're protected – we want, everybody wants these kinds of infrastructure to be constructed, but also from an Innova perspective would be penalized financially for cancelling one, even though we don't want to because of this regulatory uncertainty. So, building out to customer sites would solve that problem because it's only this uncertainty of 400kV ownership, which causes the problem. 

An Ofgem clarification of the rules in the form of a letter would solve that problem. Copying a generation license for a demand license type, set of regulatory infrastructure would also solve the problem.  

00:17:19 – 00:18:05 – Kyle Murchie  

Yeah, because it's worth thing that, you know, for our listeners that when you even get to the level of detail of the agreement, you know, the blast flow connection agreement, they are two completely separate versions. 

So, if you've got two, you've got hybrid system where you've therefore got everything sitting on under one BCA. The issue is you have two completely different regimes. Either you're a directly connected generator, which includes storage, as we said, or you are a directly connect, sorry a non-embedded, demand customer or a non-embedded embedded user, with quite different sets of requirements and some kind of heavy similarities when it comes to appendices, but others where quite, quite different. Securities is one that I think we've talked about a lot on recent other podcasts, but, you know, even the security regime is quite different.  

00:18:06 – 00:18:26 – Pete Aston  

And how are, you know, NESO, Ofgem and DESNEZ engaging with this? Is this something that they've really picked up on? I know you've said that you've been talking with them already. Are they engaging with this? Is there sort of a, you know, a sort of momentum building?  

00:18:27 – 00:19:25 – Ryan Adams  

Yeah, I would say so. I think there's momentum building. I think that, if I had to guess, I would say we are on the path to a regulatory change in this direction. 

I think the issue is the uncertainty created at the important time and the speed at which it is, you know, implemented. Because obviously the queue is being changed at the moment. New offers are going to come out soon and some of the problems of inefficiency that are baked into the system, I feel like once offers come out, they will be baked in, it'll be difficult to unravel. So, if there's not enough space, say today for one of our data center connections, but then post the reform we're talking about there is space. How is that going to manifest itself in terms of the offers and the queue if they've already issued Gate 2 offers? And I think that the reaction so far is really positive. We're getting lots of fantastic reaction. Everybody agrees generally that it's a change that needs to be made, that speed really needs to be emphasized and we can't let perfect be the enemy of good enough, which I think is tempting when you have something, this complicated.  

00:19:26 – 00:19:44 – Kyle Murchie 

Yeah, because we were recently on a kind of lot of your discussions have got to the point where there was a session with DESNEZ and, Ofgem quite recently and other colleagues in the industry. I did come away with quite a lot of positivity, but it's then having that momentum isn't keeping that momentum up and actually taking some of those ideas and, and really running through with them and implementing  

00:19:45 – 00:19:55 – Pete Aston  

And does one body actually own the action to resolve this? Or is there a slight sort of black hole in the middle where, where the action sits? 

00:19:56 – 00:21:04 – Ryan Adams 

Yeah, it's a good question. I don't think I know the answer still despite engaging with, with, Ofgem, DESNEZ, DESIT, NGET and NESO. I think I did well to get them all out in one go there. Yeah, for, for quite a while now, I think that that is one of the issues – there needs to be somebody to take ownership of it, and I think DESNEZ have done a good job of doing that recently, but they often act fairly as an opinion gathering entity if you like; so they say we can help facilitate change, but what we need to do is to have everyone to sign off. And, and for me, I think this is a big important issue. It needs bold leadership, and it needs somebody to say, this is clearly a good idea that needs to happen, that's somewhat of a panacea for lots of things. It's not going to solve everything to speed up every connection, but in, you know, in aggregate, it's going to have a positive impact. So, then they go out and say, let's work together to get this done, tell me how we can get this done instead of as is currently the question that's formulated: tell me why this is potentially a problem. It needs to be looked at from a more proactive perspective, in my opinion.  

00:21:04 – 00:21:44 – Pete Aston 

You know we are all aware, that you know recently, Ofgem has published an open letter on the surge in demand applications, and starting to talk about how queues should be formed for big demand projects; there definitely feels like there is some risk around this, you know about starting to look at, dare I say it, Connections Reform for demand, sort of has been talked about. What do you think about this from Innova's point of view, you know how does this sort of impact on stakeholder confidence, in developer confidence, in sort of bringing forward projects?   

00:21:45 Ryan Adams  

Yeah, I think it's a great question something that we wrestle with frequently and I don't think this will be the last time that we have a sort of an additional queue. I think everybody probably thought, okay, we can cut down the BESS queue and then we will be fine and then things can relax, and obviously things haven't turned out that way. And it's understandable that people are applying for demand connections because it's a very very clear government priority to connect demand customers, especially data centres so it's a good thing in some ways that this is happening and we need to work together.  

But I think there is, curation of the queue makes sense, we need to think carefully about how to do it. I don't envy those whose job it is to try and come up with that the right way,  I think we've seen from previous attempts not what to do, and we can talk about some of that. But I think there is a wider issue that needs to be thought of, which is investor confidence in the UK as a market from a large infrastructure perspective, which is being damaged in our view by too much tinkering. And especially too many announcements of things that may happen in the future and aren't certain about how they will look.  

So if you take an Innova perspective today and you look at our data centre connections, we've obviously, we've had to lobby our for about 2 years to be the company that unlocks data centre connections to be able to swap from BESS to data centres, which many investors may not be happy for us doing it; but obviously that was a requirement of the industry at the time.  

And now we don't just have to think is our Gate 2 offer good enough – which hopefully most of ours will be. We also have to think how will AI growth zones impact us, how will the Strategic Spatial Energy Plan impact us, there's talks of a new connections accelerator service, how will that impact us, how will that work? The demand queue curation is coming in, how will that work? The connection is dependent on wider works happening, will they happen on time, are they accurate when they come out? Has the queue been formulated correctly? 

I think there is so many unknowns, the more that people announce, the more elements over the top, the less investors will want to come in and invest in the UK; because they simply cannot see far enough into the future.  

Pete Aston  

And yeah, it almost feels like this, the curation of the demand queue, is almost more complicated than the BESS queue, because it feels as though there is a lot more that goes into sort of sharing the viability of a demand project, than there is an energy project. 

Ryan Adams 

Yeah, again I do think it is a very difficult task for NESO to have an adequate solution, I think some of the things that I've heard through the grapevine may come in. So we just need to be careful, so for example, one of the things is have you got a partnership with an operator, or a hyperscaler? I think at the moment, the problem is there is so much uncertainty on grid, you create a paradox where nobody wants to go first, and you can't get any interest. So you know, if hypothetically if I was to go and talk to a hyperscaler or an operator, are you interested in our data centre project, they would want to know what the connection date is and they'd want to know how and when it can be delivered. But if I can't tell them the answer to that question until they have committed to my project, the project can't move forward and no data centres can be built.  

I think that there is uncertainty and a lack of understanding fairly enough from those outside of the development industry, that assumes that there is sort of homogenous development companies that also finance the construction and actually operate imperpetunatly the assets and therefore they can look at you know, net present return of a project, start today, build it and then operate it forever – it doesn't work like that, there are developers like Innova and like other big names in the industry whose job it is to take risk, find the right types of assets, invest in them and then sell them to people who are frankly better at owning and operating projects for the long term, for decades; and those skills overlap but they are not often held in the same organisations. So, if you were to, for example, make cash a proxy, that would be okay for some large developers, like Innova and others. But lots of the best developers aren't rich with cash, with large balance sheets, a lot of the operators and asset owners are, but they are not the kind of people who are going to go and put millions of pounds down as a bond on a bet that they are going to go and get planning, and developers are. So, you have to recognise the different horses in the race and design around that. 

 Kyle Murchie  

Yeah, and I think it's quite key as well, just thinking about demand and you know, the demand queue, as you say it is different for many reasons; but also because around that point of whose in the, who's the horses in the race because you know, there's a lot of demand customer that ultimately aren't playing in the energy industry – you know, they are buying energy to make a product, or you know, to use in a data centre, but they are not in the same way gaining money from the energy industry like a wind farm, or a solar farm, etc. would do. So, I think again you've got that kind of trade off where demand is there to be utilised and used, and there is lots of players who don't really want to play in the industry. 

 You know, we haven't mentioned actually, the queue, where we are in the queue from a demand point of view, T and D combined is 125 gigawatts, last summer it was 42 gigawatts – so quite a bit of an increase. And what, about 72 gigawatts from a data centre perspective, which I suppose is a large proportion, that is still a very sizeable proportion for everything whether that is hydrogen, demand, and everything else that kind of goes down that stream. So, I think identifying that there are so many different users of demand, and different use cases is going to be really quite key as well.  

Ryan Adams  

Yeah, if I may add as well, I think it would be a very different way to deal with connections, you know from an NGED perspective, but it's maybe something for the government to think about though, the uncertainty seems to be knowing who is going to build what project where. We all know there is going to be lots of projects built, and we all know that they are unambiguously spread across the UK, right? So, you know, if you were starting again, then you would not make NGED play a guessing game where they have to guess who's going to connect where and base reinforcements on that; they need to be basing reinforcements, and I think this has sort of been talked about from Ofgem, on the basis of well we know that there is going to be lots of development that requires similar infrastructure in both generation and demand – so why don't they invest and build infrastructure on the basis that they build it and they will come because we know that they will come, and that will prevent a lot of delays that we are currently seeing. 

00:21:05 – 00:21:20 – Pete Aston 

Yeah. So, I think we're starting to sort of come towards the end of the, the time on this podcast. But maybe if we sort of, you know, wind the clock forward six months, maybe, you know, what do you think, what would you really like to see in, in six months have happened to, to unlock this? 

00:21:21 – 00:22:11 – Ryan Adams 

Yeah, I think one of the, one of the three suggestions that we've talked about, I would be realistic and say, I don't think that in the short term we're going to have a, well, I think it's possible, a formula that's similar to the Generation License for demand customers, I'd like to see that. But I think, a letter from Ofgem clarifying the interpretation of the Electricity Act is absolutely doable. 

And, you know, anecdotally, I think everybody's clear what's correct. So, I don't know what the challenge of, of putting that in writing is what I'm sure it's doable. And then I think NGET saying or agreeing that they will build out to customer sites, which we've only just touched on, but is, is really the kind of simplest solution that can be done the quickest. I would like to see that having been implemented within the next six months, and that is absolutely possible.  

00:22:12 – 00:22:34 – Pete Aston 

Wow, that, that still feels like a, there a big hill to go up here. But that's been a brilliant conversation. I think we probably could be talking about this for, for hours, to be honest, really crucial to get it right. But Ryan, thank you so much for, for coming along, Kyle, thank you, thanks everyone for, for listening and I hope you join us for another podcast soon, goodbye.