Exploring ERCOT with Brandt Vermillion (ERCOT Market Lead @ Modo Energy)

You know, in in Texas, we have lots of natural gas wells. Right? So in West Texas, there are these big well heads that actually have these, like, pools of natural gas literally.

And if it's, you know, well below freezing, those pools can actually start to freeze up. You basically just can't get natural gas into the pipelines. So you have Basically, this whole system where there are just failure points all over the place. Right?

And so you have power plants that aren't necessarily equipped to keep running. So you have some that are beginning to have to come offline. You have other power plants that aren't receiving natural gas to run and they're running out of fuel as we're getting into these really cold temperature demand is going the roof. You also have and in this case, you know, the temperature's gonna be coldest overnight, right?

So, your heating's gonna be through the That's when your demand's gonna go really high. So you're not gonna have any solar. And then to layer then that last component, you have wind turbines that aren't necessarily weatherized either. And so you're gonna have very low wind output and and not many batteries yet back there.

And in twenty twenty one, pretty low total the, storage and stall capacity. Yeah. Hello, everybody, and welcome back for another installment of transmission.

In today's episode, Quentin sits down with Moto Energy ERCot market lead, Brent vermillion.

Over the conversation, Brent touches upon his time in the ERCot control room, what goes on behind the scenes, and how things are changing for batteries in Ercot. As always, if you are enjoying the conversation, please don't forget to hit like and subscribe. It really helps us to reach a wider audience. And with that, let's jump in.

Hello, Brian. Hey, Q. How's it going? Very well. Nice to see you again. We actually kinda sit next to each other a lot now.

Yeah. Nice. Yeah. That's been a good time. Good transition to the office. Considering three months ago, we didn't know each other.

It's pretty mad how much time is been together. Very excited to have you on a podcast, actually. For our listeners who are interested in Ercot, you actually bring a very unique perspective, and I'm gonna ask you lots and lots of questions today. But before we get started, can you just explain what an RTO is and an ISO is?

And for once and for all, explain whether her car is an RTO or an I ISO. Sure. Yeah. So the whole idea of an RTO or an ISO is, essentially just a non profit entity that is essentially overseeing the operations of the electric grid in a given area.

So in North America, I wanna say there are seven or eight different ISOs or RTOs, give or take. And generally, they're yeah. So you have, like, PJM or MISO or Kaiso, which Some of those are multistate entities. Right?

They're overseeing, the flow of power and, good number of states. So PGM, I think, is as many as six, seven states.

Misos even more. But for Ercot is actually only overseeing the flow of power within the state of Texas. And so that's generally, they're gonna be the distinction between an RTO and an ISO. So those acronyms just being regional transmission operator, and I believe independent system operator for ISO So Ercot is, in fact, just an ISO given that it's only operating within the state of Texas.

And so in British parlance, a bit like National Good ESO or in Europe, an an Illia or something like that just to put everyone on the same page. Yeah. Exactly. And so, we've done this before.

We'll go again. Court stands for the electric reliability council, I believe, of Texas. Yeah. So it's essentially just like overseeing both the operations and the market side of, the transmission grid in the state.

So that's kind of organizing and centrally dispatching generation, as well as working with utilities to look at power flow and essentially how that is you know, moving from those generation centers to demand centers and ensuring that's, at least at the transmission level that, you know, all demand is being served, a grid staying balanced sixty hertz, and that the market is clearing efficiently. What I'm learning from the last twelve months of looking at the US market is you really have to understand maps get your head around it. Everything's a map. Right?

And yeah. So Ercot cover, Ercot is, is the ISO that covers Texas. Almost all of Texas, but not all of it. There's little bits around the edge.

It doesn't quite cover. Right? Yeah. Exactly. So, like, El Paso, for example, is probably the the biggest, like, demand center that isn't covered by Urcott.

I believe they're actually part of, like, WAC, I think. So and I might actually technically not be an RTO. Oh, here we go. No.

Yeah.

We can get into the the blank spots in the in the map of, you know, your ISO's and RTOs. But Yeah. So, Ercot, I believe, serves something like ninety percent of demand in the state of Texas and the actual, like, energy infrastructure that, is within the Ercot network covers actually a little bit less than that in terms of, overall planned usage. Yeah.

There's some some pretty, large land areas, maybe in, like, the Panhandle of Texas that are technically not a a part of Urcotte as well. So where there's maybe not quite as much demand. But yeah. And we're gonna talk about Urcott Lowe's today.

But before we get there, I wanna talk about you, Brent. So, yeah, we've been working together for about three months.

And I've just been blown away by how much I've learned by just sitting next to you and and and by osmosis, figuring out the stuff you know. And so What's your background? You've you've been you've been at Ercot for a while. You've you've, you've come at this from an engineering perspective. Do you want to just talk about what you studied what you did at Ercot, and then we're gonna deep dive on some bits.

Yeah. So I graduated, from College University in, twenty nineteen. And then I moved down to Texas to go to work for Ercot, and I studied electrical engineering at school. So, yeah.

Basically, I went into, like, a rotational program, with or cup itself, kind of getting a sense of what the different engineering teams do, what I like doing, and what maybe I was better at than what I and what I wasn't as good at. So that was a really great experience, about a year long program of just kind of figuring out where I fit best and what I would wanna kinda go into, I guess, I mean, I was already full time, but, like, truly full time. And so from there, I went into working in the actual Workock control room. It was basically just a support engineer.

I think the title is, like, shift engineer. So a lot of that was just actually working, like, you know, days and night shifts, you know, twelve hours at a time, overseeing grid operations alongside, like, seven operators in the controller. And we've seen the pictures. This is the big room with the walls with all the lights on it and lots of screens and buttons to press.

And the fairly high stress job with people keeping the lights on. Yeah. Yeah. No. It's, it's pretty much yeah.

It's it's pretty amazing when you when you first walk in there, you're kind of blown away. You you're almost starstruck, I guess. So, yeah, the the desks in there have about eight monitors and, you know, the giant screens at the front that it's probably a forty foot tall. We call it the wall board.

The wall board. Yeah. It's just like that. You know, data everywhere. Right? So it's just kinda hard to wrap your head around, but, like, once you're in there, it's it's pretty amazing, like, overall, like, just fire hose of information, and it's it's pretty invigorating.

Yeah. So it's there's a lot to do. I mean, there's definitely some slower times, but when it when it really picks up, it it picks up for sure. And it's it's it's almost like you're you're working in a hospital and you're like triaging, like, what's most important, right, in terms of solving multiple issues at once, and it's it's a it's a really fun challenge.

Where is it? Where is it in Texas? Is it in Austin or is it in Dallas or his Yeah. That's so Ercot is they have a couple different offices, but the, like, primary engineering is actually up in Taylor, Texas, which is about twenty, twenty five miles northeast of Austin.

And then, like, the stakeholder meetings and things like that are actually held at a building in Austin proper, so kinda close to the airport. But, yeah. So the the control room's in Taylor as well. Yeah.

Yeah. The control room isn't Taylor. And what does the con this may sound like a stupid question, but what does the control room have to do? What's the what's within the control of the control room?

Right. Yeah. So a lot of what the control room is doing. So there's there's red. So I mentioned the seven other operators that each kinda have their own individual responsibilities.

And a lot of Seven of the upgrades. Sorry. Seven of the human beings in the room with you. Yeah.

Exactly. Yeah. So there's, yep, eight people in there. Seven of them are operators as the title.

And usually, they have, you know, a wealth industry experience in some level, whether they were working at a power plant earlier in their career or actually maybe we're linemen or something like that and eventually graduated to working in maybe utility control room and then eventually working at Ercot. And then there would be somebody like myself working as, like, an engineer that's maybe a little bit more familiar with the applications that are kinda running, all the time. And so that kinda goes into a question where there's an aspect of it that is essentially just saying, hey, we have this process that's running every five minutes, which is, hey, we're basically recalculating power flow for so understanding, you know, we're receiving data from generators and utilities that saying, hey, we're we're outputting this much power, where we're capable of outputting this much power as well.

This is kinda what utilities are seeing, like, in terms of their different substation and switching stations in terms of, you know, how much, powers being carried on transmission lines and things of that nature. And then, therefore, actually being distributed to different loads going into the distribution system. The workouts kind of receiving all that data. They have their own energy management system as a term, basically, like, It's kind of a a bespoke software, that manages the grid essentially or at least allows you to manage the grid.

And so going back to that five minute process, right, so you start by a calculating power flow, then it goes into what's called contingency analysis, a real time contingency analysis, and that might be something that might not be as familiar for people coming from the UK. Right? Cause you guys don't have n minus one security? No.

We don't. Before we jump down there, though, so Yeah. Yeah.

Let's just get our heads in the same place here. So we're talking about a control room with eight people in it with loads of screens, loads of computers. And then you're talking about this five minute process. Yeah.

Right? Which is that? Is that an automated thing that happens or is it a piece of paper running around? And what do you mean by five minutes?

As in, it happens every five minutes or it takes five minutes? Yeah. So it kicks off every five minutes. I think it's on the twos and sevens in terms of the minutes, the clock minutes, right, is typically what happens.

Right? So that every every five minutes, two minutes and sevens, you have to do this thing to keep the lights on. Yeah. So it does run automatically.

The energy management system is essentially set up. So that, you know, every three hundred and sixty seconds or whatever, five minutes, it kicks off. Yeah. So that goes into power flow, which happens takes about maybe five seconds from then contingency analysis, which is maybe more like a minute, minute and a half, usually.

And then from there, there's an additional part of it that runs every ten minutes. That's kind of just for, like, overall, like, larger voltage stability, and that's where if we wanted to talk about GTCs or generic transmission can sit Sorry, generic transmission constraints intercaught. That's where those kinda come into play, in terms of establishing limits for, like, these larger interfaces of lines. So it could be a group of you know, three to ten lines, something like that, where it's like aggregating all the amount of power flowing over those lines, and there's basically new limits that are recalculated for that as part of the process.

But that's kind of a digression going back towards the five minute process. Right? So the kind of conclusion of that is the actual central dispatch of generation, within, Urcotte. So that's done via security constrained economic dispatch or scanned.

So security constrained economic dispatch. What does that mean? Right. So the security constrained aspect of it.

It kinda goes back to that n minus one security thing. So n minus one just being that if you lose any given element within the bulk system. So that's, you know, transmission lines, transformers, even generators to some extent.

Basically, what what does power flow look like? So it it you have your base case flat power flow, right, which is this is what's actually happening in real time. And then from there, it's the there are contingencies that are within the Ercot model, right, the network operations model. And those contingencies are saying, alright. Well, this this transmission line, this element of the bulk electric system will be removed. And then let's recalculate power flow to see if now any other elements become overloaded as the result of that lost element. And so, essentially, if the contingency analysis portion of that five minute process if one of the recalculated power flows essentially shows that I a transmission line is at risk of becoming overloaded.

As a result of the loss of another one, then Ercot is controlling to that, essentially, and that's part of that optimization, which is that economic dispatch portion.

Of the of the acronym. Right? So, basically, that's an additional component beyond, like, you're so if you think about economic dispatch, right? So you're you're kind of baseline level of that is you're trying to serve all the demand, as as cheaply and efficiently as possible.

Right? So that's based on the bids and offer curves which is pretty much the same system. I mean, liberalized energy markets everywhere is the same thing where you put all the bids and offers together or or the or the generating units that can that can provide power or consume power. And then you figure out the most economic way, the lowest cost way of getting an enough power on the grid.

Right? Yeah. Exactly. Right. So you have, like, marginal marginal dispatch is kinda, you know, I think one of the generalized terms for that.

Right? So you have All this demand, you have all these generators. They all offer and to dispatch or to generate power at various prices. Right?

And so all things being equal. If there are no other constraints on the system, you're just gonna get the cheapest group of generators to, you know, fulfill that demand. Then there are additional constraints that kinda coming on pop on top of that with it with the generators themselves. Right?

So if for, like, CC GTs, for example, or thermal generation, often they'll have There's gas units, combined cycle gas units. Yeah. Exactly. Right.

So you'll have, like, a minimum that they that they wanna output. So that's, like, that's one type of constraint that would go into that model. So it it's saying, alright. Well, if I'm going to dispatch this unit to generate at all, they need to supply at least thirty megawatts of power rather than just saying, oh, I'll only get one megawatt from them.

That type of thing. So that's a baseline constraint. And then from there, you're kinda layering and that, contingency analysis and making sure that all elements within the system are protected to some extent. And and yeah.

So I think I think that's kind of a good summary of of what that economic dispatch looks like. So Yeah. So and and that's how you see some some large, price variation if we were to get into locational marginal pricing. Right?

Oh, with we're totally gonna go there. But I'm Yeah. Before we get there. So what's, cot's con there's two questions here that so I thought, caught was a fifteen minute market.

So why are you talking about a five minute process? Yeah. So this is kind of a, an interesting nuance thing to get into, I guess. So You have, every five minutes, we have that process that happens.

Right? But the settlement periods in Ercot are fifteen minutes. So, essentially, what that means is those fifteen minute settlement intervals are averaging out three, five minute Right? So the the whole idea behind that is essentially it can kind of smooth out some of that price variability and price variation that you might have within a given five minute, period of time.

So you there are definitely plenty of instances within Enercare where you have, all of a sudden, you get a high system level price or system lambda price of, you know, multi hundreds of dollars per megawatt or potentially even in the thousands, that then drops all the way back down to a more standard level that you might expect. So somewhere in the dozens of dollars per megawatt hour, that type of thing.

And so that's kind of not necessarily an extended price spike. And so that might mean that the average for that settlement interval is, you know, closer to that more expected or more typical baseline. And I think that's that's part of the idea of the the settlement interval being longer than the dispatch interval, I guess. And also things happen faster than every fifteen minutes.

Right? You have to manage the system than fifteen minutes. Yeah. What's interesting about what you've talked about there is the you said there's a five minute process.

So there's a process that runs every five minutes. That you guys are doing in the control room. But, actually, the I thought it was interesting. You talked about how long each step took.

Right? Because one bit of it the economic dispatch that putting all the bids and offers together and working out what the least cost amount of power is. You said, I think if I got it right, you said it took about five seconds. And then all the constraint bits, IE, all modeling all the power flows and seeing which lines are gonna be overloaded and all the, I guess, it's factorial equations to figure out the different ways you can cut the mustard, cut the mustard, cut the cake, cut the mustard as well.

That, I think you said, took about over a minute, a hundred seconds or something. Yeah. Yeah. Yeah.

The contingency analysis is is absolutely the m of a lengthiest portion of that process. The economic dispatch is kind of almost a mar modern marvel really, in fact, that optimization is so efficient. I mean, the reason why is it not to always bring it back to Great Britain, which we need to stop doing in this podcast because we have a lot more US stuff going on. But a lot of our listeners are from Great Britain.

And there's an issue in with National Grid at the moment, what's the the the this particular initiative that's been around for a long time about this thing called skip rate where some assets You've you see it right. We talked about it. Some battery assets, some of the assets get skipped, even though they're cheaper because national grid the constraint analysis bit. Natural grid needs to focus on the constraint analysis.

Yeah. But if if you had the same computer and one calculation took five seconds, to do the economic dispatch bit, which is the bit we're all we all understand in Great Britain. And then the constraint analysis takes over a hundred seconds. So it's twenty times more computation.

That's a bit the national grid in the UK says it's the hard bit that they need to manage, and that's why things get skipped. It's just it's it's very interesting to me to hear it from Ecot, the kind of the the difference in computation there. There's just a lot more to calculate.

Yeah. I mean, I think so Right. So if we start go go back to that five minute process, it starts with, a base case power flow, which is just looking at What does that the system in actuality look like right now in real time? And if you layer in the fact that, you know, there are quite a few elements in the bulk collector system. There's you know, thousands of transmission lines and transformers. Right? And theoretically, have all of these contingencies.

So all of essentially at the loss of any of those individual almonds. All those are modeled by Ercot. If theoretically, that contingency analysis could actually take even longer if you were going to process each and every one of those lost each the loss of each and every one of those elements. Right?

But there's actually screening process for the contingencies in Ercot, which actually helps reduce the amount of time it takes. Okay. But if you think about even just with that reduced number of contingencies that are being processed, it still takes, you know, like you said twenty times as long as just a single power flow and also that economic dispatch portion. So part of that is because, you know, if you think about the number of contingencies that are still being processed and the fact that our original power flow still took, you know, four or five seconds to compute on its own, you know, if even if you're doing a hundred contingencies, I think the power flow actually ends up taking less time typically because it's also more, like, I believe the way it works, we could really get into the weeds on text state estimation and things like that.

We don't need to do that. But is that it's essentially the powerful that gets recalculated is localized. I wanna say to the area in which the contingency were to hypothetically occur. So I think that helps again reduce computational time.

It's I mean, it's pretty ingenious design in terms of how it all comes together to keep things happening on a five minute granularity. It can't help the thing. We just need more compute. Right?

I I assume that it runs in AWS or as you're somewhere in the cloud, or is it running locally?

It gets run on local servers. Oh, okay. Because then you don't have to rely on Amazon not going down or things of that nature. Yeah.

So We most have as I think. Let's do it. Yeah. I think we're got to scout some pretty substantial server for sure, but, you know, you could always use more.

I'm already aware now we're having this conversation that we're not gonna cover all the things that we said we're gonna cover. I'm just gonna jump around a little bit. Let's talk about you're in a control room. You got the team with you, and things start things start to get a bit hairy.

Let's talk about some storms that have happened or some of the challenges because they're the hot topics, I mean, everyone in Texas knows Eckot now after those storms and the blackouts. Yeah. May well, not.

So maybe with some, some lack of contacts as well surrounding that too. Right? So, yeah. So I guess one big thing to point out is that, you know, Ercot doesn't own any physical assets.

Right? They're just kind of overseeing the operation of everything. So, you know, there's and they're a non profit entity. So they stood they stand nothing to gain by know, power prices being incredibly high or anything like that, which I think was, at least, maybe some of the general public, kind of a common misconception.

Oh, the boycott was profiteering. Yeah. Yeah.

Yeah. There was some amount of Verkata's responsible for this in the sense of, you know, they're trying to make money off of this, essentially, which yeah. I mean, I think that's probably, hopefully, commonly understood now that that's not the case. But beyond that, yeah, so the control room can definitely get, to be a pretty wild place, you know, when when things are hitting the fan no doubt where you have different people that are trying to kind of perform their own job.

So, you know, you have somebody who's just trying to manage frequency. You have, people that are trying to, mitigate congestion and severe transmission issues on the system that basically can't be handled by the economic dispatch. So the five minute things running, and then there's bits that Even that even a computer can't handle and so that people have to come and do. Yeah.

So, basically, yeah, to explain what that would be is you can have an individual constraint that is essentially just very hard to resolve. And the way that comes about typically is, your example. Sorry. Because we're we're very we're talking feels very meta this conversation.

So what would be a typical constraint in a cult. Usually, the typical ones are actually feasible to resolve, I guess. So a typical constraint, we we could just call like the west west to east transfer, essentially, or west axis, the name of the GTC, or if, you know, I know we like our acronyms in America. So I'll try to spell them out, but generic transmission constraint or IROL or interconnection reliability operating limit, I think, is what it is.

Which is actually a a nurk term for another one for you. So the North American Electric Reliability Council. But West X. West X.

Getting power from the west to the Yeah. So, you know, it's a group of lines that are trying to take, you know, all this wind and solar generation that's out in West Texas and get it to the demand centers. If you have quite a bit of wind it's a very windy day, you know, combined with a pretty sunny day, where there's as of now, you know, we're talking teens of gigawatts, So fifteen thousand megawatts maybe of of solar at the same time. It's quite a lot of power to get from maybe some sparse population centers in in West Texas.

To your Dallas, Austin, San Antonio Houston population centers. Yeah. If you if you're looking on a map, if you haven't looked at a map of Texas, which is fine. You know, you don't no one's expected to know where know everything.

But generally, demand centers on the right. So you've got Dallas Fort Worth, Houston, Austin, Austin, San Antonio on the kind of east side. And on the west, Texas a vast vast vast space. And on the west side, you have this huge open land kind of desert, the kind of area, and there's tons and tons of generation there, like wind and and solar.

So you gotta get that from the west to the east, and that's what we're talking about now. Exactly. And so high level, that's one of the most common constraints in Ercot is that you see as being at risk of overload, essentially, where yeah. So you only have so much transmission capacity to move all this power.

And Ercot has to essentially be aware of how much powers are flowing over those transmission lines. And if that group of lines exceeds a certain limit that is imposed by Recod essentially that has been studied, the, like, stability analysis studies.

Then they're gonna have to curtail generation on the other side of that constraint. So they're gonna curtail Turn down. Turn down generation. Right.

So they're gonna have lines we're talking about overhead lines. This is overhead lines connected by pi to to pylons above ground, right, lots, lots of them. So your high voltage transmission lines like, you know, the giant towers, you'll see if you're driving on the highway, for instance. Yeah.

So we're talking about, like, three hundred forty five kilovolt lines. And Yeah. So if those are essentially, you know, getting closer to that limit that's studied by Ercot, and that's when we see, generation begin to be reduced, right, in West Texas. And often, your your wind and solar generation is going to be lower on that marginal cost stack just based on where they tend to offer in.

And so they will typically be almost always will just be dispatched to meet pretty much the base load, even, like, your minimum demand in Texas, it basically will assuming no constraints, you'll see all the wind and solar online.

But once you introduce the transmission constraints element, that's when you start to see potentially on the order of thousands of megawatts of wind and solar generation curtailed.

Right. So, yeah, so that's a common constraint in our cop, but if we go back to the idea of, irresolvable constraints, typically what that would be is just kind of a a tricky transmission pocket, essentially. So there's maybe, often what that might look like is there's not very much generation near it. And so why I say that is because, you know, the closer, electrically, in particular, so the closer topologically, if we think about a map, a map rather than geographically, you know, just in terms of, you know, you have a node here and a node here, it might be, you know, within a couple of notes essentially of of this potentially, overloaded element.

Right? And the closer that generation is the more likely a higher percentage of that marginal megawatt that can be produced by that generation is going to flow over this potentially overloaded element. So if we're very far from a high amount of generation, in part often this will happen on on lower voltage lines in particular because they tend to be electrically further away. What's lower voltage and numbers?

So the three main transmission level voltages in Ercot are sixty nine kilo volts, a hundred and thirty eight kilo volts, and three hundred forty five kilo So, yeah, those are generally so low voltage in this case would generally be sixty nine. Yeah. So often that'll that'll be a case that'll be the type of case where you'll see these irresolvable constraints. And essentially, right?

So you have this generation that's a little bit further away, and therefore the marginal megawatt produced by this generation has pretty minimal impact on the flow on this potentially overloaded line. And similarly The marginal maker, what? Let's let's just put it on. So let's say a gas speaking plant They can generate a bit more to the marginal megawatts.

So gas peaking plant is running, and they can do another megawatts.

And what we're saying that additional megawatt is what we care about, the difference, the delta.

That is yeah. So essentially, this This goes back if we go all the way back to the economic dispatch, right, the and we think about how the price of power is derived essentially.

So, you know, you have that system level of price, a system Lambda. That's just generally the cost of the the idea behind that is it's the marginal cost of of power on the system. And if we break that down further into the locational prices, so you have all these different nodes across the state, and each of those receives a locational marginal price. So, essentially, that's what if you had a generator at that node, that's the price that they would receive at the given point in time.

And so if you have separation between that system level price or system Lambda and the node that your generator is cited at, So your LMP that you're receiving, your marginal price hit that location, that is generally gonna be caused by congestion or some overload element across the system or potentially overloaded element if we go back to that and minus one concept. Yeah. Because you're not it's not about it's not necessarily about overloaded elements. It's about what might be overloaded Yeah.

In this concept of n minus one, which is crucial Yeah. To think that how we how we think about this. Yeah. Because you're actually solving problems that don't necessarily exist yet in some cases.

It's, yeah, it's about, you know, getting ahead of problems before they happen, right, and ensuring that you know, a pro a problem that ends up if it happens, it ends up being maybe more small or minor or easily resolved isn't actually, you know, a very large problem. That can be even potentially begin to cascade. You're just getting ahead of things. Right?

And what does that actually mean then? So I know we jump around here, but so n minus one But I think I know what that means, but I'm I'm only eighty percent sure. Does that mean that any line can go down at any point and the system stays on. Is that how it works?

Or is that Everything is only at fifty percent load all the time, or is it something else? Yeah. So really the idea is is kind of like you had just said is that you can lose any one element on the system, and the system is just still fine. It's still able to operate There's no, you know, crisis occurring.

There's no potential cause for emergency. Ideally, the controller is still in a quality state. In terms of there's there's no, you know, there's no sweat dropping off foreheads or anything like that. If lose any one element of the system.

Right? You're you're prepared for all outcomes or any not maybe not all, but any within the realm of one element being lost. Right? So Yeah.

That's that's functionally the idea behind it, is is really just to ensure preparedness, in the in the case of losing any one helmet. It, I mean, it totally makes sense, of course, but it's one hell of an overhead. Right? Yeah.

Yeah. I mean, it's, yeah, it's quite the quite the remit, right, in terms of, being able to control for that. I mean, there's there's a lot of possibilities within, you know, thousands of elements that could be lost. And then what does the system look like as a result?

Yeah. So can can you tell us what happened in storm Yuri? Can you say, well, just a bit of background on what Yuri was. Yeah.

And why why it matters so much, and, a bit of a blow by blow account of what happened and why it was significant.

Right. Well, that could be a whole podcast series, but Well, maybe we've come back to it another time. But, yeah, in a in a nutshell. In a nutshell. So with lots of detail. Yeah. Because part of it is because I don't, you know, I don't wanna just say, oh, well, it was all this sector of the industry that brought you know, the system, you know, onto the verge, basically.

But, I mean, if you wanna begin, let's just say Texas is not used to it getting very cold. Right? And so, you know, heading into, I think it was, like, the right around Valentine's Day in, February of twenty twenty one. So heading into that week, I think Sunday night was the first night where temperatures really began to drop off.

And we're talking as low as around zero degrees Fahrenheit, so well under the negatives in Celsius, which Texas very rarely sees below freezing temperatures really it's off not often the case that it gets much below forty degrees Fahrenheit or just, you know, single digit Celsius. Right? So that's I mean, when you have when you're seeing, you know, five degrees Fahrenheit on the forecast in Texas, that's that's already alarm bells to begin with. Right?

And so part of that is because, you know, demand is obviously going to go very high in that case, as it would in most places that even that would typically see that level of temperature.

But beyond that, Texas is not really prepared to keep houses warm. Right? It's it's houses here are built more to retain cool air in the summer and keep houses cool rather than to retain heat in the winter. So you have that.

You have a lot the fact that a lot of homes have electric heating rather than gas heating, which, there's a this is very engineering, heavy, but you get into some resistive heating, which, essentially, to keep beyond a certain level of loading for that those heating elements, they become much less efficient. Mhmm. And so they take, more power essentially to function. So that can drive demand further.

There's some really good academic studies on that. But so that's a component.

You have some some nonlinear components here as well. So there's that. So, essentially, right, you have really high demand. You're looking at that.

What does that mean by High demand. What is that? That's at ten gigawatts, hundred gigawatts, what is it? Yeah.

So for Texas, that would just be we're talking, I think, at the time, that was probably more in the seventy something gigawatt range. We were at maybe looking at over eighty for that week. So almost might end up getting there, but Almost twice the peak demand of Great Britain on a very cold night. Yeah.

Yeah. And, yeah, this past summer in twenty twenty three in Texas, I think, Pete demand got, probably a couple gigawatts already eighty megawatts So that would have been the new the new record demand or or whatever you wanna call it. Yeah. So, yeah, just for that context.

Right? So then the the next element, right, is you have. So, I guess it's just for an additional contextual point. Generally, the it would the fuel mix of Texas.

Natural gas. I would say, this is kind of a ballpark number, but I'll say it's maybe around forty percent of demand typically throughout the entire operating year. That's just based on the last couple of years. Natural gas as a natural gas power station.

Yes. Yeah. In terms of your if you're gonna break down your fuel mix by natural gas wind solar, you know, that type of thing. Natural Gas, I I would say is probably in the the forty to fifty percent range.

Okay. So if you think about, you know, you have if you have, you know, your eighty gigawatt peak demand that you're expecting, then by nature, you're you're hoping to have actually, in in this case, it's probably gonna be even higher amount of natural gas generation because typically in in really cold conditions, if it's cloudy or outside and wind turbines aren't necessarily, equipped to handle, like zero degree weather as much, especially here in in Texas where they haven't necessarily been weather proof as much as they might be for for an offshore wind farm. Yeah. Yeah.

Yeah. Off the off the coast of GB or whatever.

But yeah. So you're gonna have a high proportion of natural gas generation. Those natural gas generators are, again, built with summer in mind rather than winter. And so you have a lot of instrumentation and cooling elements that are actually housed outside, right, to help keep them cooler in summer rather than warmer in the winter where they might be built in a big brick building somewhere.

And if you were talking about it all, like, in the northeast of the US and New England, that that's all your natural gas generators are built in a completely different way. So some of the gas plant wasn't wasn't plat wasn't planning to run well, isn't really designed to be run at such cold temperatures as well. Yeah. In Texas, so that's you have that, elements of it.

There was maybe a lack of weatherization, for cold weather. As well. Like, so lack of preparedness in that sense for some of the natural gas generators, that's a component. That's definitely not every natural gas generator, but it definitely happened.

And then going further, into the natural gas side of things, you have the actual gas industry itself and the transmission of fuel to those power plants. So that's both in the pipelines and as well all the way back to, you know, in in Texas. We have lots of natural gas wells. Right?

So in in West Texas, there are these big well heads that actually have these, like, pools of natural gas literally. And if it's, you know, well below freezing, those pools can actually start to freeze up, and you basically just can't get natural gas into the pipeline. So you have basically, this whole system where there are just failure points all over the place. Right?

And so you have power plants that aren't necessarily equipped to keep running. So you have some that are beginning to have to come offline. You have other power plants that aren't receiving natural gas to run and they're running out of fuel as we're getting into these really cold temperature demand is going through the roof.

And then you also have squeezed on all sides. You also have and in this case, you know, the temperature's gonna be coldest overnight. Right? So your heating's gonna be through the roof.

That's when your demand's gonna go really high. So you're not gonna have any solar. And then to layer then that last component, you have wind turbines that aren't necessary weatherized either. And so you're gonna have very low wind output and And not many batteries yet to get back there.

And in twenty twenty one, pretty low total so, storage install capacity. Yeah. So maybe on the order of maybe five hundred megawatts, a ballpark, it's somewhere in that range, definitely less than a gig. Not in the multi gig that we're in right now.

And Right. And growing too. Yeah. So, yeah, in that context, kind of, just, perfect, perfectly imperfect storm.

Right? So you have all of these different elements kinda coming together to, working in concert to really, throw the the best punch mother nature has to offer at the power grid. So it's coming up to Valentine's Day. The cold weather's coming in.

You have all these different scenarios, which are squeezing the system as a whole. At what point, you know, what what's what's happening at urquhart this time?

Yeah. It's very much an all hands on deck situation in terms of so for me, for me, I I was, I think I was going to my second full time shift as a shift engineer. So it was a real trial by fire situation for sure. So, yeah, I think I'd I'd gotten NERC certified.

You have to pass a test. Be able to go into the control room, actually do things. And I started training in, like, October of twenty twenty. And so, yeah, I think it was, I think it was my second full time shift rotation.

Going into that week. So you're the scapegoat.

Yeah. It's me. I I I pressed the dig, red, but Yeah. Yeah. Yeah. Yeah. At Attercot, you have, you know, you kind of are getting reserved crews to kinda help out and run additional power flow studies.

I think it was that to help, troubleshoot all this, all these, like, additional, like, congestion issues are you know, trying to resolve constraints essentially that are irresolvable by the economic dispatch. You have a lot of that going on. And so it's maybe more than a typical crew would be able to handle effectively and efficiency efficiently in the control room. So you kinda have additional support being called in in terms of you know, people kinda in in the back room running running, support studies.

Right? So I kinda help help say, hey, hey, we have this issue. Can you run the studies for, and basically us a solution that we just kinda have in our back pocket because a lot of these are m minus one situations. Right?

So, if you're controlling for something that hasn't happened yet, you basically just A lot of a lot of that is just having a plan to resolve the issue if it does occur. And so that's what all these studies are being conducted for is you just need to have something on record to say, well, we know what we're gonna do if this happens, and then this we have this overload that we have to deal with quickly. And it's just all about preparedness. So you have a lot of that going on.

You also just have yeah. I mean, so we have if if we go back to, like, the GTCs and so you're The constraints your, yeah, your larger interfaces of constraints that are across multiple transmission lines. So those, you know, typically a transmission line has a limit that's just supplied by the utility. Right?

So if it's, hundred and thirty eight k v line, maybe that limits it it can handle four hundred MVA or essentially four hundred megawatts, but with the reactive powers is MVA. Right? So It can handle that much flow across that line, but if you group them together, that's when you have an interface is what is what they call it. And that's where you that's, like, how that Westex can train.

Constraint is derived as you have a group of, like, ten different transmission lines, but you can't necessarily just add up all of their, transmission, like, thermal transmission limits together because that's actually gonna be a larger number than the limit that is ends up being supplied by ERCot because when you have this really large transfer of power across a large geographical area, you can run into, like, voltage stability and transient stability issues.

Which is, you know, that's that's super engineering heavy, but in Doctor, right, was from from west to east. Yeah. Yeah. Yeah. You have huge voltage sag essentially, which for maybe non engineering types is essentially just that's potentially very bad. It's different, slightly different, but If you we have it in transmission lines in Great Britain from the kind of super grid out to the West of Wales or the West of Cambria. We have the same thing, but I mean, we're talking about, like, one twentieth of the distance here and certainly not the same power.

But it's everywhere. It's everywhere. Kinda lost where we're at. We're at where we're coming back to.

Right. Yeah. Good point. We're so we're talking about Yuri. So storm Yuri comes in. It's getting really, really cold.

It's about two valentine's day. It's a nice kind of anchor in the conversation. Yeah. And then the control room notices it's all it's all about to kick off.

And then what actually happened? Because there were blackouts. Right? There were blackouts. What what what what caused well, we know what caused that, but what happened to make that happen?

Yeah. So we walked through kind of the different, maybe contributing factors to the overall difficult situation. Right? So you know, as we get into the night of, I think it's Sunday, February fourteenth is maybe the the day that I lived in in for me.

Romantic? Yes. Right. So that's quite the present. My my girlfriend, of course, stranded at home in our apartment.

I'm holed up in the hotel waiting to go into work the next day. Quite the Valentine's day. But you we're you're getting into Sunday night. Temperatures are beginning to drop even further.

So we're going from, you know, right around freezing to well below freezing talking about single digits Fahrenheit, negative Celsius. Right? So demand's going through the roof and getting towards around maybe eleven PM, I'll say.

Things are, you know, getting tighter for sure conditions are, like, looking like it's going to be, you know, a tight operating night essentially as demands really high and Urcotte's trying to to fulfill that. Kind of an all hands on deck situation. Lots of transmission constraints being dealt with. Trying to make sure all all the generations online essentially as much as possible and that it's, you know, functioning okay, you know, frequency, of course, staying around sixty hertz. But that's that situation is evolving. It's getting, you know, tighter and tight grid conditions are getting tighter and tighter, in terms of you have less and less reserves available in terms of the amount generation you have to meet additional demand.

You're really kind of reaching down the back of the sofa now. Yep. Grabbing everything you possibly can. If you can generate your online on the system right now. Yeah. Shaking out the piggy bank for sure.

Yeah. And So then, kind of, all of a sudden, you pretty, pretty much it almost happened simultaneously.

You started to see generators have to come offline, whether that was kinda going back to the different factors that we we talked about, whether that was, failure at the plant itself, where, you know, some sort of mechanical component wasn't functioning anymore because of cold weather or something of something of that nature. So you had actual power plant trips, you had just run backs in terms of, you know, something's not running right. So we need to reduce power and eventually come offline.

And then you also started to see generators say we don't have any fuel because we're having gas companies declare force majeure, act of god in terms of this crazy weather event So these contracts that we have are actually under undeliverable. It's not funny. Right? Because lots of bad things happen.

I mean, people without power for days. Yeah. This includes you know, critical things like hospitals and some people sadly die. So it it was a it was a tragedy.

Yeah. But that's the the The extreme the number of extreme events that happened at once is just remarkable, and the phone was ringing and people were saying, our control room, we've gotta come offline. And and then what? Do you have to choose was it planned?

Blackouts? Do you have to okay, we're gonna have to shed some load here. Who's coming off first? Yeah.

So the the way that essentially works is, you know, once you reach a point of and so I guess to kind of finish off where we were before. So you have all these plans coming off, pretty close proximity to each other. We're talking with, like, within minutes and sometimes with within seconds of each other, we're talking about some relatively large combined cycle gas plants where you have, you know, five hundred megawatts coming offline pretty rapidly or If it's multiple, then you have maybe potentially into the gigawatts of power. There's supply coming offline pretty quickly.

Above all else, the one thing that Urcott is trying to do is keep grid frequency at sixty hertz. So matching supply and demand, and all of the complexities around it really go into solving that problem. The heartbeat. The heartbeat.

Right? So if you have all the supply quickly falling off, typically what you would need to do is replace that to make sure that you know, we're still supplying demand at sixty hertz. But if it's coming off very quickly and it's in these large magnitudes, that's not necessarily easy to do. And so what ended up occurring is frequency started to sag off of sixty hertz.

And usually, like, a a pretty large frequency drop is maybe only, like, like zero point one hertz or, you know, one tenth or a hundred millihertz, something like that. And those are all the same just for different units. Following those three with the same number with different numbers. Yeah.

Yeah. Yeah. Yeah. Just express different ways sometimes within the industry that happens. I just get an idea about is this happening?

Is this taking hours? Is it taking minutes? This is they're talking about, like, this is within minutes now. So I wanna say this is getting towards one AM.

Okay. Don't remember exactly, but, yeah, we'll we'll call it that. So, yeah, we're talking about within minutes. You have frequency really starting to sag off of sixty.

But when I say severe sags, kind of for that reference point I had, that zero point one hurts. Right? That would be relatively significant. You'd start to see some reserves beginning to get deployed to restore frequency.

But at the actual absolute nadir, frequency was pretty close to, like, fifty nine point three hertz. So for contacts, that's, like, point seven hertz relative to that. Already would would have been relatively significant zero point one. So that's that's a huge deal.

We're talking about, you know, seven times more than what's already a relatively significant frequency excursion.

So when that happens, if you if you really wanted to dig through the ERCP protocols, you could see the timing of, you know, as frequency gets into really severe scenarios, basically, how long you have until utility loads or utility demand starts essentially just removing itself, start tripping off. So you have under circuit break is opening. You've got timed switches that's saying I'm gonna hang on. Hang on.

Hang on. Hang on. Hang on. If this frequency doesn't fix itself, I'm gonna have to drop off.

I have to for the sake of protecting equipment across the system. And essentially, yeah, those the thing the actual physical things that are gonna be doing that are called under frequency load re relays. So They have timing settings to just say, you know, if frequency is below this point, you're going to basically hold off, like you were saying, for this long, And then at that point, you will open the circuit breaker, disconnect this this load. And so that kind of the the amount of time decreases as frequency gets further away from sixty.

Right? So at, like, fifty nine point seven, that might be I don't know exactly, but, you know, dozens of minutes probably. But as you get towards, you know, fifty nine point four, fifty nine point three. We're talking about, like, only a couple of minutes.

And I think below fifty nine point three, it's on the order of, like, seconds. Cascading. It's good. There's a diagram with this cascading effect.

Yeah. Well, you just get it. I think it's more extreme. You've got less and less time.

You'd have one of your favorite things on it, logarithm scales. Yeah. I do love a logarithmic scale.

Yeah. So essentially, again, going back to our scene, we're talking about frequency being, I wanna say, I wanna say it got at a slow point to believe between, like, fifty nine point three and fifty nine point four. And I think the number was something, like, seven or eight minutes, essentially, until basically all transmission demand across the system was going to be shed. And we had literally a true blackout and would have to you know, restart the grid from Like an entire system blackout. Yeah. That is the worst case scenario, isn't it? The worst case scenario.

You're gonna have the National Guard deployed to Texas, and people are gonna be getting water on street corners and bat just the all time worst case scenario essentially. Right? So you're minutes away from that, basically. You'll try to establish frequency from zero.

The cool thing about the electricity system is there's always a always a frequency that everyone can that can what's the word I'm talking about here? Synchronized to. Yeah. But but when everything goes to zero, someone's gonna start it up again.

We have things called BlackStar and lots of other contracts like that. But real, real bad. Yeah. Yeah.

So let's not let's not let that happen. So you're trying to avoid that. And so, basically, that's Ericock syndrome, of course, is already well into taking action before this. They're trying to, you know, replace the generation that's been lost.

But once you reach this point where you're like, alright. Well, we're minutes away from, like, complete darkness.

So we need to get rid of demand to rebalance frequency. Right? So I think ultimately it ended up being something like twenty gigawatts at peak outage that was that was cut. And a lot of that happened. Pretty pretty quickly. Oh my.

Pretty quickly on the on that night of September. I'm sorry. February fourteenth into February fifteenth. But it didn't just stop there.

Right? So twenty gigs came offline ish. Yeah. But then some of these places were out of power, but they didn't have power for days.

Right. So if we go back to all of our generation that was lost and all of our frozen natural gas wellheads, well, they stayed frozen because it was below freezing for four or five days or whatever.

So you still have all these gas contracts that are going on delivered. Still have power plants that had mechanical failures at some point. They just have to get it fixed before they can come back online. And they're probably running out of fuel as well.

You still have wind generating very amounts of power. You have solar generating very low very low amounts of power. Oh, and power it's funny worth just noting power prices here. So power prices were as you'd expect in a scarcity event through the roof.

So you'd think so the whole idea is that when power prices go so high that the system the you know, economics fixes everything. Right? Power price goes really high, it encourages everyone to generate, and you kind of solve the problem. Yeah.

Or when there's nothing left in the tank, You can have these really high power prices and nothing to resolve it. Yeah. Yeah. So it's the market, of course, was still running, but it was almost like market was suspended because it was basically just like, if you are able to be online, you will be running essentially as a generator.

Are you getting paid, ham hand somebody to do so? Getting paid, like, I think at the time, like, nine thousand dollars for megawatt generated.

Yeah. That's on an hourly basis, but pretty much just, like, never never before seen, probably never will be seen again type power price of us. Yeah. Yeah. For real.

That's that's what being touched. Might might need to legitimately do that. But Yeah. It's it was kind of a a true, like, You gonna say it say it?

Black squad? Yeah. Yeah. We'll call it that. We'll call it a black swan. I saw a chart on, on Reddit.

Not that I spent a lot of time on Reddit, but I saw a chart on Reddit, which is like, you know, you could do Google. You can search Google terms, the amount of the amount the usage of black swan as a phrase. It's like ironic. Yeah.

Yeah.

Five years with COVID and Yeah. Wars and all that stuff. But that's read it for you. Alright. We talked about Yuri. There's a couple more topics I wanna talk about briefly.

Firstly, let's talk about batteries in our cart. What's the what's the state of the nation for batteries in our cart? And then I wanna ask you your contrarian view which, I mean, I do already know it, and I think everyone should hear it. It's very interesting. But firstly, batteries and Aircot, what's the situ?

Yeah. So, like, we, I guess, kind of, to go back to your reefer one more second. We said we had probably only a few hundred megawatts of of storage capability at that point.

As of now, in twenty twenty three, it's end of the year. I think we're we're over three thousand megawatts of of the build out being both commercially, well, just both operationally and commercially online.

So, yeah, and I think the as of now, based on Hercon numbers are a little bit higher and part of that is because there's a period of time when a resource is online versus when they're actually operating commercially. There's lots of testing that they have to do, essentially, to get approved for essentially full interconnection, full participation in the market. So their, capacity number is gonna be a little bit higher than, like, the EIA or the energy infrastructure administration agency, something like that. They provide a data set, basically, just kinda, cataloguing all the generation across the, I think, entire country, if not continent.

And so within Ercot, their numbers a little bit lower. Just to summarize it. That's the kudos to that that data set is amazing. Yeah. I mean, just who whoever's responsible for putting that together to serve as a big high fiber. We can thank the US government for that. Thank the US government.

So, yeah, that's a great data set. Great resource. But, essentially, their capacity rejection for the end of next year, I wanna say is, like, eleven gigawatts. Mhmm.

Storage. We've got eight or nine in our numbers, haven't we? Something like that. Yeah. So, yeah.

Yuri, half a gig online. In the next year, somewhere between eight and twelve So, yeah, to frame that, that's, you know, about a little less than than four years to get to that point. So that's pretty insane build out for sure.

There are lots of batteries connected to Urcotte. That's about And even more soon. So what's that? What what what was next?

I think next, the the big thing that people have been talking about and will be talking about is ancillary service saturation. So sure a lot of your UK listeners are are familiar with that one, but the idea of just, typically, for storage resources, their entry point into the power market is often to start by just receiving ancillary service contracts. Right? So that's providing, you know, your your frequency response and your reserves to to integrate essentially and getting compensated to do so.

Part of the reason for that is one because especially for the reserve contracts, you're not going to be deployed, all the time. So you can have you can essentially sit idle without actually having any throughput and still get paid for that. So that's valuable to a storage resource where they have, a a life span where, you know, the number of times they cycle, they get less efficient and eventually don't really function quite nearly as well as they as they did at beginning of operation. So, yeah, summarize.

That's part of why ancillary services are desirable. And on often when they're not saturated, storage resources can not quite set the price, but really help drive the the price of the service and get pretty well compensated for it. So well, just a note on that though. Yeah.

In Ercotts, what's quite unusual from a European lens is that it's not just batteries that provide in these services. There's a lot of different participants. Yeah. For example, you have RRS as one of the reserve services within Ercot, and there's a huge component of that that is typically just fulfilled by load resources.

And part of that is because there are kind of subtypes of that ancillary service where, specifically, one of the subtypes is just at a kind of going back to what we were talking about before, the under frequency load relays. If you reach, I think it's fifty nine point seven hertz or below, the all those load resources are going to be deployed. And what that means, essentially, you know, if you have if you're a big industrial customer and you have a non frequency load relay, you can just say, alright. Well, frequency gets that low, which has happened basically one time, in quite a long period of time.

It's just a it's a blunt instrument. That's it like dynamic dynamic. But, like, demand side response deals. Yeah.

Yeah. Where you trip stuff off and you get paid for it. Yep. Exactly. And it's very It basically never happens.

So they're pretty much willing to carry it for close to for free because it's just, you know, like, why not? We have this relatively large, amount of power demand so we can get compensated.

For buy, and, you know, an additional revenue stream. It's not gonna carry on though. Right? Because if you if you can have dynamic services and a battery can respond and It's you don't have the opportunity cost in a battery like you do of switching off an industrial load.

Do I quite even want that service? Long term or when when batteries can do it better? I would say yes because there's room for both of them, and it's kind of just an additional tool in the back pocket and it's not and that specifically that type of ancillary service is really not costing, but it's been very much money. Yeah.

It's incredibly cheap. So I would say that one in particular, I'd say is pretty safe. By the way, if that trip happens, that's gonna be the time of scarcity when those units won't wanna be consuming anyway because price will be so hard you'd think. You think.

So, you know, if you're, manufacturing plant and prices go to nine grand, well, Doesn't matter whether you tripped or not, you don't wanna be running. Yeah. Almost like a That goes back to, you know, the if we wanted to get into the buy lateral type four. So they're probably gonna have some kind of off-site contract where they're hedged against that.

But, yeah, so so that there's that component. There's also, as you to other generation resources, particularly in, answering services like ECR S and non spin. So ECR S being the UrC contingency reserve service. Which is the new exciting one.

Right? That is the new one. It started a new shiting. June of this year. Yeah.

So additional answer service, pastly being awarded. Right? So that's fun. More more It sounded like you didn't think that was funny.

I know for a fact you really think that's fun. Yeah. Well, it is in a design. Right?

I mean, it's the the not fun part is it's, costs a grid more money. So there's that, but maybe it's more reserves. That's good. Going back to that, you have thermal generation that can pay in those last two answer services because the deployment time that's required by Ercot, if those services are deployed is is a larger period of time.

Have thermal resources that, you know, they're not gonna be able to deploy the way a storage resource would be within seconds, obviously, for storage or nearly instantaneously, really So the fastest thermal resources are only gonna be able to get online in, like, ten minutes and a lot of them, would be even more like thirty minutes. So you have thermal generation that can participate in those services, essentially. And that makes them more competitive than your regulation and IRS services.

Which, are, at this point, pretty well dominated by storage. They're not maybe dominated as a a strong word as of now because participation has been more, like, sixty, seventy percent rather than ninety, a hundred percent, in terms of the con the average amount of contracts being awarded and maybe a given operating month. Essentially, you know, as going back to our build out numbers as more and more storage comes online, that percentage of contracts being awarded to storage resources should continue to rise theoretically.

Because they want to be awarded ancillary services. And, ultimately, there reaches a certain point where there's essentially too much storage capacity to all receive insider service rewards. The market gets more competitive drives the prices down. A musical chair situation where somebody's gonna be left standing up.

Right? There's not enough contrast to go around. But there's plenty of other market. There's plenty of other ways to make money in the market.

There is a huge locational element to that as well, though. Is he the producer is currently giving us signs that we've gone on too long? And I wanna ask you the last question, which is, what is your contraption? You don't need to plug anything because we're on the same side here, but we plug stuff it off as it is.

What's the what's the contrarian view? What do you believe that not everybody else does? Yeah. So one thing we didn't quite get to mention yet was MPR eleven eighty six.

NPRR for their our last acronym of the day. NPRR is nodal protocol revision request. So work out has notal protocols, which are essentially just the rules by which the grid functions and a revision request, obviously, just being a request to make a change to the stakeholder process that essentially, derives legislation for the grid. That, revision request, NPR eleven eighty six, was recently approved by the your account board of directors.

Essentially, what it means is storage resources will have to reserve, enough state of charge So essentially enough energy to, fulfill any ancillary service obligation that they have received.

For that given operating hour. So if, for example, a surge versus has five megawatts of RS at the beginning of the operating hour in which they hold that responsibility to fill that obligation.

They need to have five megawatt hours of energy stored within their resource. And that then will decrease throughout the operating hour. The amount the obligation that they need to or the amount of energy that they need to have stored to fulfill their obligation. So what they're saying is to put this into to just to paraphrase a little bit if I've got this right. What aircraft is saying is, look, if you want a frequency response contract, that's fine.

But we need to make sure you've got enough gas in the tank or enough energy in your battery to really support frequency the duration we think it's that may be necessary if you're gonna get paid. The problem is that all sounds very reasonable. Yes. The problem is if you're gonna just if If if that's the requirement, then you don't have as much state of charge to do other stuff at the same time. Yeah.

Yeah. So, I mean, essentially, it all comes back to, you know, theoretically in the worst case scenario, you might have the Sandler service deployed for the entirety of your operating hour, and Ercot is just trying to make sure that they're actually going to receive the power that they contracted for a full deployment. When you say deployed, you mean, like, the frequency is low for the whole hour. Yeah.

And you get paid for the hour, and so you have to discharge for the throughout the whole hour. You're you're gonna have to actually provide that throughput. That's essentially the main component of the realism now. There were previously, prior to the legislation being approved.

Some additional components or maybe a little bit more restrictive in terms of the amount of ancillary services that storage could qualify for and things of that nature. That's actually not in the legislation anymore. It's been or get out through the stakeholder process essentially. So at this point, that's Good news.

Yes. Yeah. Because this is all this like, let's cut to the chase. This is bad for batteries.

Yeah. In the sense that it it's an additional it's an additional thing they have to do, and it's something else they have to manage. And it's it's very conservative. Right?

Because, I mean, ninety nine point nine percent of the time, an ancillary service type is not going to be deployed for an entire operating hour. But that's kind of the battle that's that's waged in a lot of those stakeholder processes is that ERCot is always trying to protect against that zero point one percent as much as they possibly can because it's not zero percent. It is technically feasible. It can happen.

And so it's kinda clashes with the idea of, you know, the probability wise, it's probably not gonna happen. Storage resources are just saying, well, hey, what if we only had ninety percent of our state of charge that we absolutely needed for an entire operating hour? And then if if it was actually this crazy event where, like, we needed every bit of ancillary services deployed, maybe we can move it around our operating fleet. If you're, an operator that operates multiple units, we can kind of move that responsibility around.

But the problem is is it's just kind of a slippery slope where it's just, like, some, participants might be able to part to operate in certain ways that, like, could make that work, and some just might not have enough resources. There's just additional complexities that Ercot can't really guard against. Right? So I definitely I've I've I know I'm coming from Ercot, so I have sympathy to that perspective as well.

Essentially, I going all the way back to my concern. Yeah. What's the contrarian bit? You've just probed something that's really bad for batteries.

Right? Yeah. But I It's here here with us. Is is that it's not actually that big of a deal.

Oh.

You heard it here first.

Yeah. So essentially, one, this is something they were probably already more or less doing is essentially having enough energy within a given storage resource to actually meet their ancillary service obligation. Because if you have a deployment of an ancillary service and you don't fulfill your obligation, theoretically, that could result in a compliance penalty. So you don't want that to happen.

You don't wanna have to pay for a mistake like that. You're already doing this math anyway. Pretty much. Yeah.

I mean, and maybe there are some participants out there that are a little bit more aggressive with their state of charge management, and that's kind of what this rule is mostly in place to kind of guard against? Much no names. No names? Yeah.

But, the other component to it is we were just talking about the ancillary service market saturating because we're gonna have all this additional storage capacity. Right? So, you know, a year from now, you're gonna have so much of the storage fleet that has no ancillary service obligation. And even if you are a resource that has an ancillary service obligation, it's probably only gonna be a percentage of your overall capacity.

And then beyond that, one of the arguments in the stakeholder process from maybe the storage side of things was that, you know, in these extreme scarcity events, you know, it's over the course of this past summer. We had, you know, some pretty tight days operating on the grid. And you saw storage really, contribute to kinda getting through those type periods. So, you know, often that's driven by solar ramping down in our cot.

And net load. So your demand less wind and solar going pretty high in a short period of time, you kinda have to fulfill the gap that's that's left by the solar generation with more flexible generation. So, you know, thermal or storage. So you saw Storage with the the knight in shining armor many times that Yeah.

It did a great job. Absolutely. And it's it's an incredibly valuable service to the grid and how flexible it is and how fast it's able to act on those, conditions appearing. But one of the arguments was that you're gonna be was that Ercot essentially was gonna be limiting that in cases like this where you have those scarcity conditions and you have all of this energy that's tied up in your ancillary service obligations.

But I get my perspective on that is that if it's really that, tied up an operating condition and grid conditions are are that critical, ancillary services are gonna be deployed anyway. So you have you're gonna have UCS and non spin and RS. I guess ranks higher anyway, doesn't it? Because I mean, we're coming back to this this heartbeat thing.

Yeah. It's it's way above the the shedding load. Yeah. It's also, you know, twenty eight steps in between there theoretically as well.

So Yeah. As those if it's if it's tight conditions, ancillary services are going to be deployed. And, like, so so with, like, ECRS and non spin, for example, the deployment of those involves essentially just a resource releasing that reserved capacity that they had to the economic dispatch. And so they might still not even have to provide the throughput.

It's only if they get centrally dispatched by Ercot. So I, in my opinion, if it's a true critical situation, the capacity's still gonna be there. So in my opinion, that's maybe not quite as significant of an argument. I'm certainly sympathetic to the storage cause, especially in the fact that this is kind of a guarding against the the point one percent type thing.

But I I'm also coming from maybe more of the operations background where I kind of see the value in that. What matters is what sorry. Lots of things matter. I was about to, describe this as one binary thing, which is it isn't.

Yeah. Yeah. But one of the things that really matters is the dollar impact. Right? Yeah.

Yeah. And from what we said earlier, I mean, look, if you're if you're building an asset now, the majority of your business case probably isn't ancillary services anywhere, Yeah. And maybe maybe it's a very short term, it still is. But if you're taking a medium to long term view, it's it's one ancillary services are only going to be a smaller component of that.

And given the suppressed prices in them that are coming in the near future, they're not going to be a significant contributor anymore. You're gonna be much more energy arbitrage, focused. And beyond that, the from, like, the reliability argument, the capacity is still gonna be there. So Yeah.

Going to sum it all up. I think NPRR eleven eighty six is, both somewhat necessary and also go it's gonna be alright. For storage resources. I don't think it's gonna have that large of an impact to their medium to long term bottom line.

Absolutely framed. And with that, we've reached the end of the podcast. We didn't touch location, marginal pricing much. We didn't touch modeling and forecasting, We hardly talked about other types of generation classes.

There's so much we will talk about. But maybe next time, I want to say a massive thank you for joining us on a podcast and joining us as part of the the Moto team. Yeah. If you're listening to this and you haven't yet, please do hit like, subscribe.

You get a lot more of this kind of content in your ears once a week. Alright. Thanks everybody. See you soon.