On The Brink with Castle Island - Ray Cline and Shaun Connell (Lancium) on Stabilizing the Grid with Bitcoin Mining (EP.254)

Episode Date: October 25, 2021

Ray Cline, Lancium CTO, and Shaun Connell, EVP Power at Lancium, join the show. We cover the changing Texas grid, several models for renewable based mining, and how miners can serve as a controllable load. In this episode: Personal backgrounds and crossing the threshold from energy to Bitcoin Origins of Lancium Features of Texas grid Why Texas is so suitable for renewables How Texas is the 'Saudi Arabia' of renewable generation Why renewables grew so fast in Texas Two reasons why prices reach negative prices Three models for mining Grid connected with new gen miners Renewable colocation with new gen miners Off grid renewable colocation with old gen miners How different ASIC vintages affect the power price you can tolerate for the units and what interruptibility they can deal with ERCOT's surplus renewable power How the Texas grid is splitting in half How Bitcoin is a 'premium customer' for energy assets Why miners depreciate their miners over 4-5 years, and not 1-2 years as Digiconomist claims Are energy producers aware of the importance of Bitcoin mining as an alternative energy buyer? Introduction to Demand Response What a Controllable Load Resource is and how Bitcoin miners uniquely satisfy the requirements History of Controllable Load in ERCOT Why controllable load is so useful to grid operators Why Bitcoin miners can be flexible load, and why other industrial consumers of energy cannot Why the economics of operating an interruptible load as a miner make sense Are other grid operators looking to build programs for BTC miners? Do Bitcoin miners decarbonize the Texas grid in the final analysis? Will Bitcoin mining improve the reliability of the grid? Sponsor notes: This episode supported by Public.com. Start investing with as little as $1 and get a free slice of stock up to $50 when you join Public.com today. Visit public.com/onthebrink to download the app and sign up. This episode is brought to you by Withum, a top 25 accounting firm with a cutting-edge Digital Currency and Blockchain Technology practice. To learn more, visit  withum.com/crypto.

Hello and welcome back to On the Brink with Castle Island. I'm Nick Carter. This episode is supported by with them and public.com. More about them later in the episode. So today we're sitting down with Raymond Klein, who is the CTO of Lancium and Sean Connell, who is the executive VP of Power. Lancium is a Bitcoin mining company that's operational in Texas, and they do some really interesting stuff with demand response and controllable load, which we'll get into in this episode. So this is a really important episode in terms of really digging deep into the energy dynamics of miners in Texas specifically and why the Texas grid is so interesting, sort of undergoing this transformation. If you've followed some of my work on this, I've been covering the concept.

and I give a talk about it in Austin. But this episode, I think, is probably the most detailed account of what's really happening. Why Bitcoin mining paradoxically might actually help make the grid much more sustainable and decarbonize it. And these guys are actually doing it. So it's not just theoretical or navel-gazing. Now, we start off by talking about three different models for mining, how miners can use renewable energy assets. And a lot of people tend to critique bitcorners when they talk about this and say that, oh, the low capacity factor of renewables means that renewable-based mining doesn't work for Bitcoin, but they're wrong. And we actually go into a lot of detail as to how specifically this can work. We then dive into the interesting dynamics in Urquat and how it's getting more unstable as it on boards more renewable energy.

And then lastly, why Bitcoin miners are such a unique load resource on the grid being highly interruptible and why this matters and how this allows miners to participate in things called demand response programs, which help stabilize the grid. This conversation really opened my eyes to what's happening in Texas. So I found it very, very informative, and I hope you'll like it as well. Brought down by bad mortgage investments, Lehman, which has 25,000 employees, will be liquidated. The federal government loans American International Group, AIG, $85 billion. This is a different kind of market, and the Fed is asleep. The federal government is stepping it to stabilize Fannie Mae and Freddie Mac, the two mortgage giants that have been threatened by the housing crisis.

The Bank of England has pumped 75 billion pounds more to Britain's ailing economy with a new round of quantitative easing. You print a couple trillion dollars and all of a sudden people started to worry. So out of this worry, we have something called a Bitcoin. Bitcoin. This is an episode I've been looking forward to you for a long time. I think it's probably the most technically challenging episode

I've done on this show. It's a topic that has seized my attention over the last few months. I've learned a lot from these two gentlemen here. I expect to learn a lot more over the course of this hour. So I'm sitting here. here with Raymond Klein, who is the CTO of Lancium, and then Sean Connell, who is the executive vice president of power at Lancium. And I'm told that you're respectively in Houston and then the Houston of

Canada, if I, a Calgary. Is that right? That's right. I'm in the energy city of Canada, so Houston and Calgary. So two of the two maybe of the centers of Bitcoin mining. I certainly Houston is. I don't know if Alberta has this much mining, but, you know, certainly relevant places to be. So Lancium is a really interesting company that I've sort of come across recently. You guys have pioneered some pretty interesting stuff in terms of grid stability, which is now becoming sort of an issue that's sort of spilled over onto the political scene as of recently. Obviously, Urquod is sort of always in the headlines these days, but more recently Ted Cruz has begun talking about it most recently at the conference in Austin.

So I figured we would dig into, you know, what is you guys do and then how mining could potentially actually be handy in the case of a more renewable grid, which is a very counterintuitive result, but I think a really important one. So before we start running, I would just briefly, both of you tell me a little bit about your own backgrounds and how you came to work in sort of the Bitcoin mining sector? Sure, I guess I'll start. I've got 40 plus years of experience at the intersection of distributed computing, high performance computing and energy, both oil and gas and utilities. And because of my technical background, I followed the development of things like cryptography, digital money, and other topics, which led to blockchain and

and Bitcoin in particular. I started Bitcoin mining, I think it was about 2013, 2014, something like that, with a 5-terahash mining contract from Genesis mining. And then eventually from there, I bought A6 and GPU mining rigs. And I realized that, you know, if you go back to 2017, the last peak, you heard a lot of the same topics that you hear right now and a lot of concern about energy consumption. And I realized that we needed to do something at least about the perception of energy consumption in this space if it was going to grow. I was introduced to Michael McNamara through mutual contacts, became one of the co-founder. of Lancium and I joined the team specifically to help address the energy solution.

Excellent. And what about yourself, Sean? First of all, thanks for having me on the show. I'm a big fan of the content that you put out. So thank you for putting out such great content and having me on the show. My career, I guess I'm kind of one of the few that I found that in the Bitcoin mining space, it really started in the energy space and kind of came over to Bitcoin mining, where a lot of people in the space are in Bitcoin mining are kind of not a point where they need to learn the energy part because it's such a big deal to the mining. For me, I started back in 2002 with a large power generation company here in Alberta. It was right at the start of deregulation and I was starting in an energy marketing role.

And when that title came out, I didn't even know what that meant, right? Because what was happening was deregulation was just starting to happen, which was essentially kind of, you know, changing from the, you know, traditional vertically integrated monopoly of the past and deregulating power markets. And so I was working for a large power generation company that was going to be deregulated and they need to figure out, you know, how are they going to be marketing their power? I spent about, you know, the last 20 years of my life working for this power generation company, essentially building and leading a trade desk that was responsible for optimizing a fleet of natural gas assets in the wholesale power markets in the northeast.

So that's kind of in Ontario, New York, up in that area. And I was introduced to Bitcoin back in 2017 where a friend of mine called me up, and he's actually from Miami. And he gave me a shout and he says, hey, listen, I'm looking at this Bitcoin. What do you think? And I come from the commodity space where my experience is that all prices need to converge. Right. So like you can't have something where it costs you, you know, $5 to make and you sell for 100,

Right. Convergence happens, and we witnessed this in 2008 when gas prices were $10. And so say that, you know, the price of electricity is $120. And there's coal plants that have a cost of $40, you know, working for a company that had a lot of coal generation, you know, feel real good about it because you're making this really big spread between, you know, the $120 and $40. Well, 2008, what happened was is that natural gas prices collapsed. And all of a sudden, you know, the coal generation was $40 and the gas generation was $40. So, you know, prices converged. So it's been my experience that, you know, cheap energy prices always gravitate towards the high energy.

And so when my friend told me about Bitcoin, it was in 2017 when Bitcoin was about $10,000. And, you know, I stepped back and I was like, well, how much it cost to mine it? At the time, it was about $300 or something around there. It was really low. And I was scratching my head and I couldn't make sense about this because I kind of heard these headlines about a million dollar Bitcoin, right? And just like, and for me going back to the commodity space to saying that like, what happens at a million dollars, you know, what is the power requirement? You know, can you really have this big spread exist for this

long? Like how does this need to converge and how can it play out? So my introduction of Bitcoin mining came from the kind of commodities lens and kind of figuring that there's got to be a point where there's going to be a ton of competition in the space and, you know, there's going to be convergence that happens. so that in the long run is that your marginal cost produced Bitcoin should be similar to the cost of buying, right? And that this could be many years out or it can be at times during cycles. So my introduction to Bitcoin came that I was essentially doing the math on this. We were looking at this power generation company considering doing some Bitcoin mining. And then, you know,

COVID happened. It got shelved and, you know, I really wanted to continue on the journey. And so that's That's where I came across Lancium and now I'm in the Bitcoin space full-time and really able to kind of, you know, kind of be really excited about doing this as my full-time full-time gig. So as a bitconer who has begun to learn about the energy space, you know, I found it very, very difficult. It was incredibly complex and arguably more complex in Bitcoin. In your view, is it harder to go from energy to Bitcoin or from Bitcoin energy?

I think it's probably harder to go from Bitcoin to energy because when you go from energy to Bitcoin, you really just look at Bitcoin as a commodity, right? And it's just another one of those things, but it happens to be a really, really big thing that a lot of people don't understand. So it's really, I think it's easier to go to the other way and kind of defending you a bit on the complexity of the energy bills is, you know, my parents are still challenged to understand their electricity bill, right? And I've explained to them, few times and it's really hard to explain to them why the cost of delivering the power costs more than the energy and telling them that's a really good thing because energy is really cheap then, right? But they don't feel good because they're paying so much for these extra charges,

right? So energy is very complex and I can understand the challenge for people coming in the space and trying to figure it all out. Yeah, that's certainly what I've found. So you guys are primarily or entirely based in Texas in Ircott, is that. as they call the grid there. And I want to, you know, focus a lot of the conversation on Texas, which I think is kind of an illustrative grid in many ways. So maybe just sort of by way of introduction, you know, like what is, what's unique and interesting about Texas

or sort of like what are the distinguishing features of the grid there and it's sort of evolution in the last decade or so? Probably the biggest difference that Sean's going to go into through the conversation. you know, the differences between the eastern grid in the United States, the Western grid in the United States, and ERCOT. Ercot is covers most of Texas and most of Texas is ERCOT. So rather than one of the things that's interesting is that rather than having a bunch of state public utility commissions that have to agree on something for the grid operator to execute on it, In Erkot and in Texas, it's, you know, the Texas Public Utility Commission makes decisions. They work with Erkot.

Erkot advises the Public Utility Commission. So, you know, we're not going to Oklahoma or Louisiana or New Mexico to ask for permission to do something. So that's one thing that's very unique. And it's a market-based grid, which means that everything that Erkot does is, done primarily through market incentives. And this has supported the deregulation of the market in Texas and also encourages innovation. Again, you know, if you come up with a new idea,

there are ways for you to be rewarded for it through the markets in ERCOT. And, you know, so I think those are the primary things. It's an entity that has more autonomy. It is market-based, and that market-based really supports deregulation and innovation. Maybe to add on a couple of pieces on, you know, about Urquot and just, you know, it's really a re-touched upon being like a favorable regulatory environment, right? So like it's Bitcoiners welcome, you know, it's got the right type of market design and supply mix. It's also a deregulated wholesale market, which means,

means that, you know, we're allowed to participate in the energy and in solary services in that market. And then another part that, you know, kind of often overlooked is that the billing that you have for your electricity bill is it's a competitive billing, which is like it's called, I believe it's called retail choice in Texas, which means that, you know, anybody can supply energy in Texas in certain locations, like outside of municipalities and co-ups. So really, you know, really makes that favorable place to be an ERCOT. So the question is just kind of like, what are the impacts on the renewables that are coming into Texas?

And so, you know, going back into really kind of a strength of Texas and why it's so great, it's just been a market-based grid in the sense that, you know, there's not a lot of subsidies that occur within or cut where, you know, there's lots of subsidies that occur in other areas. And so really it's been about the actual economics of wind and solar now being at the right price threshold that they can actually start to deploy these research. resources. On top of that is that in 2013, Erkot came out with what's called the Cres lines, which is the competitive renewable energy zone. And those were about 16 major tie lines that connect West Texas to the major portion of Texas down in the Dallas in Houston areas where there's major load centers. And what this enabled was, you know, if you can think about a map with the United States. And, you know, Texas is, you know, the punchline is, it's one of the only locations in the U.S. that has the highest quality of sun and the highest wind speeds, right?

So if you think about the map of the U.S. and you kind of draw this kind of like, you know, kind of a thick middle section of the U.S. from, you know, the border of Canada all the way down the U.S. that's called the U.S. wind belt. And that's where you really have the highest wind speeds across the U.S. And so it's, it's no surprise that that's where a lot of this wind generation has been built over the past 20 years. There's also, you know, thinking about the same map and you go to the southwest portion of the U.S. in the kind of the bottom quadrant. And that is, you know, some of the best sun quality. You kind of overlay these together in West Texas as you really get some of the best sun quality in wind speeds, which has really been, you know, the birth of all this kind of

renewables that's been coming on. But what was required as well was to have those Cres lines that were created in order to move that power across the power lines because without those major tie lines to be able to move this, you know, this, you know, kind of large amount, large sun and wind deposits that pays out in perpetuity, you can't really tap into that unless you have some way to kind of move that power or to have some type of large customer located right at the source of that kind of generation. So adding all this renewables to Texas and, you know, when these lines came out in 2013 is that they thought that it would take many years to fill them up. Turns out they got filled up in just a few years.

So that was much faster because so many, there's so much of wind development, solar development. And over the past, you know, a few years and, you know, it's really showing up this year and now is that those tie lines are full and all this wind has been coming online. You know, Urquot has approximately 35 gigawatts of power at the end of renewable power at the end of 2020. And they're going to be adding another 30 gigawatts in the next, you know, three years. So what we're seeing now is that we're starting to see the split that's occurring in Texas where all of this wind that's being injected in the western part of Texas is now competing for those finite amount of transmission lines that moves power from the west side of Texas down into the main load centers.

And then just as a rough balm park number, it costs about $2 million per circuit mile to build out transmission lines. So there's a cost for building out these new lines. and every grid's always trying to improve itself and to kind of, you know, have cheap power flow to expensive power. So every year, I'm going to try and add lines, add lines, but it's a very slow process and it's expensive. So I guess kind of long answer to the question of, you know, what's been happening is it's been fabulous. You know, there's been a large portion of the energy that's been made up in Texas. You know, that's increasing every year.

But now the challenge that they're having is how do you get this large amount of renewables that's now constrained to those load centers because of some limitations on, tie flows and how do you prevent wind farm and solar farms coming online and just having to curtail that power because it has no place to go. And I guess just to build intuition for for those of us at the 101 level like electricity can't be teleported, right? Like it has to travel and it sort of decays in transit. There's even like a mathematical law that sort of determines this, right? So that's why you need to step up the voltage and have the specific infrastructure to to move it long distances. Yeah, I think also a sort of simple tie to history. When you look at the various forms of electricity generation, hydropower obviously has to be located. The generation has to be

located where you have water to move over the turbines. So the generation there is fixed. When we went to hydrocarbons for generation of electricity, you could transport the hydrocarbons to the place where you needed where the load was at. And if you look at the structure of our grid in the United States, a lot of it was built around Alcoa's aluminum smelters because they consumed a lot of electricity. So you'll look back and you'll see coal fire plants, natural gas plants, and so forth, located near the load and then move the fuel. But we can't move wind and we can't move solar.

So we're now back to a situation where the generation is going to occur in a favorable location, West Texas being one of the best. And then we either have to transport that electricity over transmission lines or we have to move load to those generation sources. And both of those things are happening. essentially Texas is the Saudi Arabia of renewable generation. And they need to find ways to get their production to customers. And so in the past, the customer's been just the grid. So you're constrained on just the grid being the only one that's out there that can actually purchase your entire amount of production.

And so this is the first time ever that there's a new customer on the block that's a premium customer that has the ability to purchase all of your production and has some flexibility. and we can get into that a little later on, but just that's a, it's a first time ever. And just for a quick piece of context on the renewable buildout, was that growth due to subsidies or was that just an organic, spontaneous thing? So there are incentives in place either investment tax credits or production tax credits that have helped spur the growth of renewables. what you see now is that the cost of production of electricity has dropped dramatically for renewables. We've seen photovoltaic cells drop in price.

Wind turbines are less expensive per megawatt than they used to be. So I think it's been an interplay between the incentive. that were put in place through the tax regimes and the growth of the industry to reduce cost. And so renewables are very competitive now. But we still see a lot of tax credits. That's not unusual in any industry that the government's state and federal, incentivize the things they want more of through tax policy. And just on the subsidy part there is it's also kind of worth noting that that one that is the production tax credit, investment tax credit is a federal program.

So it's not just applicable in Texas, but it's applicable across any location in the U.S. So that's kind of been a driver. A common question we'll get is, you know, how can power prices be negative, right? And the response to that is that, you know, some of these developers have a government tax subsidy, like a production tax credit where they're getting paid for a certain dollar amount for every megawatt they generate. And so the example being is that if a generator, wind farm is getting paid $20 per megawatt as this federal production tax credit, that means that they're willing to continue producing as long as the price isn't below negative $20. So they're still, it's incentivized to still produce at negative 19. So that's a common question and just clarify that, you know, the subsidies for the development are federal, the larger ones have been.

Because you wouldn't normally see, in the absence of some external subsidy or credit, you wouldn't really see a power, you know, wind farm operator paying people to take their energy. I mean, if there wasn't demand from the grid, they were just dumb. it into the ground or I suppose not run the turbines. I don't know how it works exactly, but the existence of negatively priced power of them paying people to take it away from them, that implies the existence of the subsidy. Is that right? And like most times, the answer is kind of depends, right?

So most of the time, that could be the case, but there could be also different scenarios where price signal on the grid is the ultimate signal for trying to get a different behavior, right? And so if you need, for example, if there's an issue with some type of congestion, somewhere often a certain part of the grid, and they need to have generation of different location ramp down quickly to relieve this type of congestion, right, is the best thing they can do is to send a really negative price signal, which would help out with the congestion on a different part of the grid. So there'll be times where you can see, for example, like, you know, I've traded in northeast power markets and, you know, you can have prices at, you know, negative 500,

negative 700, just on how their congestion calculations are formulated. And that's a really good price signal because it says, hey, we need you to turn off and, you know, price signal is telling you that if you continue generating, you know, you're going to have to pay this amount of money to continue generating. So kind of back again is that most of the time that is the right in the sense of like there's a subsidy at consistent prices that happen to be in kind of just just below zero. But for those really deep negative price spikes, those are usually from something else. See, I told you I was going to learn something new. And here we are.

And there I was thinking I knew all this stuff already. So I want to get into these models for how mining intersects with these somewhat stranded assets or assets that are producing a surplus on the grid. And you guys walk me through three models in which miners can sort of take advantage of access renewables. And I know, you know, Bitcoin has been talking about this for a long time, but not with a lot of precision. And so that's what I was hoping to get into is more precision around how exactly miners are actually starting to take advantage of renewables. Sure. So maybe I'll kind of start with, you know, what are we seeing for different models for mining that are either well, all of them are helping with the acceleration of renewables and two of the models are helping to maintain grid stability and security.

So I'll just kind of say the three models first and I'll kind of go through kind of what they are. Is that, you know, the first one would be grid connected with new gen miners and a high up time. The second one would be a renewable co-location, so co-locating with the wind farm, with a new gen miner and then firming your load from pulling from the grid. And then the third one would be co-location with a renewable generator using an old gen miner and potentially not connecting to the grid and just using all of the offtake from the wind and kind of oscillating kind of with the output of the wind. So the first one is the grid connected with a new gen miner high up time.

This type of setup is essentially having a large Bitcoin mining facility that's connected at a transmission voltage in the grid. And what you're doing is that you're buying energy from the spot market. So in the case of ERCOT, you're buying from ERCOT. And then you're simultaneously selling ancillary services to the grid, which would be kind of helping with the reliability of the grid.

And we'll get into ancillary services later. Yeah, remember that term. It's important. Yeah. And for now, just say that, you know, the ancillary services is the heartbeat of the grid and we're helping keep the heartbeat strong so that it doesn't have a heart attack. And we'll kind of go through that after. So in the first example of transmission connected, buying energy from the grand is that

you kind of categorize those in the two parts on, you know, you're helping with the renewable price economics and you're also helping with grid stability. So the way you're helping with wind producer economics is that you're strategically located in an area of the grid where there's an excess amount of renewables, a limited amount of load, and limited amount of transmission. So, example, again, is this is the Saudi Arabia of West Texas for wind. And what you want to do is you want to be located in these areas where there's, if your load was not present, you know, during these times when there's a split that occurs on the grid, is that there would be pricing those that are being sent out to lower the prices for that area

because they need to curtail the wind because there's just too much that's being injected into that area. And so what you want to do is you want to find these areas where there's excess, excess wind, limit transmission, limit load. And you're really kind of acting like a synthetic transmission upgrade. You're kind of relieving a bit of the congestion in that grid, and you're kind of helping mop up some of those cheap energy prices. So the benefit for a wind producer is that's having these types of facilities located is if the price floor was previously, you know, negative $20, and now you've got this really large

scale controllable load that's sitting there, you know, that price is now going to be less negative or positive, right? Because you've now taken that kind of left tail of the price distribution and you've moved it more to the middle, right? And so you can just imagine kind of what that'll do to the economics of renewable that have built in models to have these really low prices are negative, and now you're converting those into less negative or slightly positive. And so what that's going to do is that's going to incent more renewables to build out

because you've improved their price economics because you've essentially created this price floor for them that was not there before. On the grid stability and reliability for this model is that the benefit to the grid, on why is this a really good thing for the grid, and why would they be happy with this is that you're providing a backstop to the renewables because as you transition to a grid that say has gone from the very start zero percent renewables

to now 35 percent that means that at times during the day there is wind and solar can make up 50 percent of the energy being supplied to the grid and with wind and solar is that you can't predict with perfect precision on what they're expected output going to be in 20 minutes from now, 30 minutes from now. And what happens when, you know, if your grid is, you know, 50% are more for wind in solar and there's a cloud that rolls in or the wind stops blowing and, you know, Ray had commented that, you know, what really makes Urquat special is that there's really three grids of the U.S. there's the Eastern Interconnect, the Western Interconnect, and Urquat.

What that means is that each of these separate grids are kind of like islands within the U.S. You know, they have DC ties. So all powers produce at AC alternating current. And then there's these separate DC ties that connect the east and the west and then also Urquot. So that way the power that's in those grids is exactly kind of just the generation that's in those grids. So this means that in Erkot, if you've got over 50% renewables and a cloud covers a sun and

you have this really big drop in energy, you can't lean on your neighbors to provide that energy for you because you're a separate island. So you could be really big trouble as you start to put on more renewable. So this is one of the examples of grid stability is that you're backstopping renewables. So if you need to have 10 minutes of three gigawatts of energy because you just had, you know, you've got 40 gigawatts of wind and solar online and you just happen to lose three gigawatts in five minutes, right? You can lean on a control below to backfill that until, you know, the cloud passes or you can

bring another generation, but that's going to be really important. And then the second part is that you're hardening the grid resilience by providing these heartbeat products that really kind of keep that heartbeat strong and we'll touch on those in a bit. But that's kind of the first model is kind of like the grid connected with new gen miners. The reason why it's new gen miner is because currently right now for an S-19, it costs about $3.5 million per megawatt of capacity, which means that if you're amortizing that over a four-year period, that means that you need to recover, on average, $105 per hour around the clock, 8,000, 60 hours in a year,

times four, to recover your CAP-X. So you can't afford to be down very often because you have to recover this CAP-X. So that's kind of why in the first model is you want to be grid-connected with a very high uptime, but with flexibility for support and kind of backstop in the grid. But that's kind of the tie-in on why you run those with new gen miners. So just to double-click on that, because it's a very important concept. The newer, the hardware, effectively, the higher the opportunity costs of not running it. So the more latest-gen hardware, you want to be running that as much as possible.

With older-gen-hardware, you can afford more interruptability because it's a lower opportunity. cost. Yeah, and to put a number around that today at this point in time, if we took an S-19 Pro 110 Terra-Hash model that consumes 3.5 kilowatts, it's going to generate about, you know, round numbers, $500 a megawatt hour. So older generation miners are going to be less than that. And so that has to factor into that's your opportunity cost range. You know, obviously you're not going to run anything at zero. But if you're running very old machines at around $100 a megawatt hour, then, you know, you're going to want to turn off at prices that are lower.

And we'll get into discussions around, you know, why would you turn off at all? Yes, there are reasons why you turn off. We'll get into the details of that. So for older miners, you're more price sensitive. Yes. For newer miners, what really matters is uptime. Yeah. And you can tolerate a higher price, electricity price.

Yeah. The CAPX for an old gem miner is about 100 to 200K per megawatt versus that 3.5. So like that amortization over four years would be something like five bucks, right, versus that 105. So it gives you a lot more flexibility and kind of not needing to be up so much because your your CAPEX was so low on the actual mining equipment. Now, I read a paper from a certain gentleman known as Dig Economist, who works for the Dutch Central Bank, to be clear.

And he claimed that the amortization period for miners is 1.2 years, which I found somewhat incredible and here you are professionals in the mining space telling me that you're amortizing a new unit over four years so I'm guessing that that his view of the depreciation schedule of miners is erroneous yeah I think where that number comes from is if you look historically bit main and other manufacturers would come out with a new sort of you know, flagship model every year and a half. Those times have extended.

And the other thing is that I don't think he's taking into consideration is that didn't mean that the old miners were taken off the shelves. Because generally, a lot of those upgrades every 18 months were more incremental rather than revolutionary. So evolutionary rather than revolutionary. And so we had a mixed fleet, very mixed fleet. So I think it's not correct to look at it and say, you know, miners are turning over every 18 months even historically. Right. Now we're into more of the traditional computing equipment cycles, which are, you know, four or five years. Yeah. And that short estimate was used to buttress a point about Bitcoin, creating a lot of e-waste. So it clearly was a motivated reasoning.

But yeah, I mean, you know, S-9s are still operational. I think I looked at some of the non-s estimates. I don't know if you guys agree with this, but, you know, from some of the data providers, I saw that the network is something like 25, 30 percent S-9s still. They came out in 26. Yeah. And today, an S9, depending on, you know, what you run it at and various factors can have a break-even point of $150 to $190 per megawatt hour. If you have cheap energy inputs, you can still make money with an S-9. Yes. And there are a lot of used ones out there that you can buy cheap. Last year in July, kind of during the bare market, you know, the S-9s were selling for $20 per unit. Right. And so to get one megawatt of capacity is about 700 miners. So that's $14,000 for a megawatt

capacity of miners. And coming again from like the commodity trading lines, it's just, I was looking at that as like that's like buying a synthetic call option on Bitcoin where you own the physical, right? And if you went out and tried to buy a strip of call options on Bitcoin, it would cost you way more than what would have cost just to buy the $14,000 in miners, right? So miners were a cheaper way. of getting derivatives exposure to Bitcoin itself than actually buying the derivatives at an exchange. And the caveat is you need to have a megawattic capacity to plug these things in and capacity scares. But if you have that capacity, much cheaper.

And then it kind of ties into this example that we're going to go on Model 2, which is, Model 2 is the renewable co-location with an actually this one's going to be with a new gen miner. and then it'll come up for the second one for the use miner. So Model 2 is, you know, co-locate with something like a wind farm with a new gen miner. And what you're doing is you are going to do grid firming. So you're going to, you know, take the offtake from the wind farm. You're going to firm from the, firm your power from the grid. Grid firming means, you know, when the wind farm isn't generating access,

you are supplementing your power with just regular grid energy. Correct. Yeah. So we'll use this one here an example of saying like it's a 100 megawatt wind farm and you're going to put beside a 100 megawatt Bitcoin mining facility with new gen miners. And if the wind farm, a term that's in the energy space is a capacity factor. And capacity factor means that what percentage of the hours in a year will be, you know, essentially producing power. And so capacity factor, you know, ranges across the US from as low as I think 25% to as high as somewhere around 45% for onshore wind. And then offshore wind can be much higher. So like out in the New York area, I don't know the numbers, but I would say north of 80, but I'm just guessing on that. But it's a much higher capacity factor. So for this example, we can say that the capacity factor on this wind farm is 40%. So that means that on average, 100 megawatt wind farm will produce 40 megawatts. And then grid firming

means that on average you're going to be pulling, be a customer of the grid, pulling in that power from the grid and kind of firming up your load so that you've got a full 100 megawatts that's towards your to your farm, to your mining facility, control below. So this one requires a high uptime as well because you're going to be using new gen miners. And so you need to be able to make sure that you're able to recover your CAPEX. And so the question would be like, you know, why does this make sense? Right? So why would, you know, say a wind farm want to do this? And the answer is, is that, you know, a wind farm only has one customer who's capable of buying all of their production, right? And so that customer is the grid, right? And

and that wind farm is grateful for the grid. Like, we're all grateful for the grid. We don't have electricity in our house, super grateful, right? Like the world will be much different without the grid. And, you know, kind of take that step further is without the grid, right, they're actually just a wind farm that's just spinning a windmill with no connection to the grid and not monetizing anything, right? So having access to that customer is great. However, in areas where there's really high wind speeds, such as in ERCOT and the region, the power pole region above ERCOT is called SPP. And so ERCOT and SPP, um, and so ERCOT and SPP, um, they receive about a 50% haircut on their energy versus what the value of a fixed block of power would be.

So a fixed block, for example, is what was the average electricity price over all the hours during the year? And so say that was, you know, easy math, you know, 30 bucks. And so when the wind's blowing, right, the wind's blowing for everybody. And so it's not just, you know, so there's a lot of generation coming on the grid. And so that usually means that power is cheap. because power is abundant. And so that price of $30 is the average cost all hours, but the hours that wind is running,

they're only getting paid $15, right? Because the wind's blowing for everybody and it's cheap for everybody. And so they're getting half as much as what that fixed block would have been. So you can think about how in this model is it's Bitcoin becoming the premium customer to win farms around the world. So as an example, you know, as a Bitcoin as a premium customer is like, pretend you are an American business owner who owns an operate to wind farm. So you are the entrepreneur, right? And so this is your wind farm. And so currently you've got one customer, which is the grid. You're grateful for

the grid. It's able to monetize the wind that you have available at your location. and the grid is able to buy all of your hourly energy production, but it's at a variable price, and you have no say in what that price is, but at least they're giving you something for it. And so they're going to buy everything you have. It's a variable price, and you have no say in the price, but you're grateful for them being available as a customer.

So enter in customer, customer two, which is Bitcoin, right? And so Bitcoin, as the premium customer says, this customer has the ability to buy all of your energy production at a fixed price, mutually agreed upon between yourself and the person that wants a facility. And they're going to buy everything that you have as the wind blows. So there's not going to be any risk to you that if you're selling something that you can't produce, so as your turbines are spinning, they're going to take everything. And they're going to give flexibility to you, the entrepreneur,

to sell your energy to your grid, customer if they need it more than you do. Right. So if they feel that, you know, it's important that they need that power. It's more essential to them than, you know, Bitcoin customer B is going to give that up. On top of that, customer B, Bitcoin is going to help you, the entrepreneur, maintain multi-customer optionality by offering protection and security to customer A, the grid. Right.

So they're going to ensure that they are safe at all times and able to continue. to do their day-to-day business, right? So that sounds like a pretty amazing customer, right? And so, you know, kind of bring that back is, you know, like, why would, you know, the wind facility want to have a co-location of this flex below it is because it's going to improve the price economics, right? They're still going to be able to serve the grid. And then that's going to incent them that they can actually do this more and they can

start to overbuild the capacity of the grid with this premium customer that's going to be their support, but always given that multi-customer optionality. and flexibility to say if customer A needs it more than I do then you know I don't need it at that time so that's kind of uh model two anything to add there ray yeah i think from the bitcoin mining side when we talk about model one it's the traditional model that people are used to you know you build out your facility you put your miners in you have your power contract and you go to town you start mining So any flexibility that you need there in terms of mining load are really dependent on other things we'll talk about later. As we look at Model 2, it's predominantly a model that most Bitcoin miners would be used to except for the last part that Sean talked about.

And that is that if you're going to defer when the wind farm, when the grid needs the, electricity that the wind farm is generating, then of course, you have to be able to ramp down your load. So you need some flexibility there. So from a Bitcoin mining perspective, Model one is the traditional model that miners are used to. Model two introduces some flexibility in terms of ramping down the load and so forth. And then as we get into Model 3, that's where the greatest amount of flexibility is going to be required. And so that's, you know, I just wanted to add the sort of mining side of the view of these things. And just before we're getting a three, so Model 2 is something, you know, I think many

bit corners, non-mining bitcoiners, wouldn't have been aware of just yet. I mean, when Bitcoiners think about mining with renewable assets, they think about a sort of like fully off-grid model, which would be Model 3, which we're going to talk about. And they're like, well, how does this work? The capacities factor is too low, you know. you need that continual uptime for the mining to make sense. So is there a growing realization among Bitcoin miners and then also energy asset owners that this model too is actually a viable thing and is something that can work?

I think the answer to that is there's probably much more awareness from the energy asset owner side than there is from the Bitcoin mining side. You know, we're still seeing a transition within the mining community from the, give me the best power price, I want to stay on 100% mentality. And the Model 2 and the Model 3 that we'll talk about are definitely departures from that. But Model 2 in particular, you know, if you own wind farm or you own a solar farm, or or other generation asset, for that matter. A lot of the people in that space are looking at this and saying,

hey, this gives me a bit more autonomy than just dealing with the grid. Or if you're in an area where the grid is not competitive or deregulated, it gives you know a complete option that is separate from the grid. So I think there's more awareness on the energy producer side. There's growing awareness on the mining side, but it's still, I think, slanted more toward the energy producers being aware of it. And I'm sure we're going to hear some more in the coming months from these producers. I mean, I certainly haven't seen much there in terms of case studies and anecdotes. But my guess is that'll change.

Let's take a quick break to talk about our sponsors. Now, on this show, we only really like to work with sponsors that we actually know and can recommend. And both sponsors today are great examples of that. So our first sponsor for this episode is with them. They're a top 25 accounting firm with a cutting edge digital technology and blockchain practice. So wherever your company is, from pre-seed to IPO, they have tailored solutions just for you. They've helped some of the largest companies in the crypto industry with audit, tax, and advisory needs. And they've also helped some of our portfolio companies, too.

To learn more about their advisory, audit, and tax services, go to on the brink. Dot link slash with them. That's on the brink. Dot link slash with them. Next up, we have our newest partner, public.com. Getting started investing, especially in public equity, can be really difficult and confusing. It seems like there's a lot of barriers to getting started, especially with cryptocurrency. Public.com is a platform that allows you to participate with fractions of shares instead of having to buy whole shares.

It's also very community driven, so there's a large community of investors discussing stockpicks and trades. And now they've also added support for cryptocurrencies. Now, I've really enjoyed my time on public so far. One of the things I really like are the audio discussions, Public Live, where you have public market CEOs and equity analysts sharing their views on particular stocks and themes. It's been a really interesting experience so far, and I like that there is public equity alongside cryptocurrency. Now, one of the interesting things about public.com is that they take the privacy of the community really seriously. They don't participate in payment for order flow, which is a common practice

in which the brokerages actually sell your trades to third parties. Instead, they route your trades to the exchanges directly, so there is no middleman. You can start investing with as little as $1 and get a free slice of stock up to $50 when you join public.com today. Visit public.com slash on the brink to download the app and sign up. That's public.com slash on the brink. Now, this is valid for U.S. residents age 18 and over. it's subject to account approval.

See public.com slash disclosures. This is not investment advice. Moving towards the third model. Now, this is the way I used to think about Bitcoin mining with win in solar assets. And I always kind of struggled with it because I didn't quite understand how you could mine with the low capacity factor profitably. But what you guys are telling me is that. that a no firming model co-located with a renewable asset can work. So going back to model one is that an important kind of concept to kind of grasp is that

a generation asset has a wholesale meter. So that wholesale meter is where they're getting paid. So they're injecting power into the grid and they have a wholesale meter for payment. As a customer from the grid is that you have a retail meter. And that retail meter is what you're getting billed on. Right? And so one is getting paid for. One is what you pay. And so in Model 1, that was grid connected and that's just a retail meter. Right. So like you're having something you're pulling from the grid and you're making payment. There's the other side with being a CLR is that you're kind of a bit of a platypus and that you've got some payment coming from the sensory services. But for the simplicity of this example is that you just have a retail meter. In model two is that the wind farm has a wholesale meter. Right. And to firm the load is you're going to have a retail meter. So you have two meters, one for injecting power to grid when that customer needs a bet more. And then second is that retail meter for firming the power from the grid.

So there's two meters on that one. Going into number three is that number three is the full swing of first one retail, second one retail and wholesale. And this one is just wholesale. So the wind asset is grid connected or not. And I'll get back into that. But in this example is that this is a wind asset that is existing. building and it's connected to the grid. And so there's going to be a wholesale meter for, you know, connecting to the grid.

And in this setup, we'll call it a 50 megawatt wind farm and a 50 megawatt mining facility is that in this example is that whenever the wind is blowing, you know, that power is going to be going to the mining facility, right, so that flexible load. However, it still applies the same as from the second model is that. that whenever the grid customer wants that, what needs that power more than, you know, the mining customer B, is that power is still going to be sent to the grid, right? You still have this wholesale meter. But in that example, again, of, you know, the capacity factor of being 40%, so that means on a 50 megawatt wind facility, on average, is producing 20 megawatts, right? And so you have this 50 megawatt mining facility that's taking, you know, it's oscillating, right, between zero and 50 megawatts throughout the year with an average.

average of 20 megawatts that's actually being put to that mining facility, except for in the cases where, again, where the grid customer needs that power more than when customer B, and it pushes the grid. And the economics work on this because in this example, you're using an old gen miner, right? So if we went back in time and this was done last year, right, you paid $14,000 for the equipment, right? So the CAPX was literally zero, right, on a cost of amortization. which means that, you know, if you're able to monetize your power at a higher value than, you know, what you could get for the grid, this is a free upgrade, right? So it's kind of like, you know, on Bitcoin mining is that up until now, the grid has always

been constrained by power lines, right? So to actually move power around and to monetize it, it was through power lines and direct use of that electricity. In the world of Bitcoin mining is, you know, Bitcoin mining is like a refinery that refines that electricity into hashes and then monetizes that electricity through Ethernet cable, right? And so now all of a sudden, you know, you're not constrained anymore by power lines because now you have Ethernet and kind of moving towards satellites and now there's a really unconstrained world of power where power can't be congested.

And so like this, in this example number three, is that, and this is required with the smart response technology that Lancium has, is to make it so that it oscillates the output exactly with what the wind farm is, and it's kind of ramping up and ramping down and, you know, doing more hashing, less hashing, more hashing, less hashing. And the economics makes sense because the cap X per megawatt is so low and that you're getting a higher return, you're getting, you know, a simple way to put it is that, you know, if you run a business and you have one customer and that customer is a monopoly on you, wouldn't you love to have a second customer, right?

And so having two customers is better than one, and that gives the optionality between, you know, if A is better than B, I'm going to sell the A and vice versa. So you're improving the economics of that wind asset because it's lowering, you know, no longer is that wind farm turning offline when prices are negative. They keep on turning and they push that power to like the Bitcoin mining facility, which improves their economics, which says, hey, we're going to do this again, right? We're going to build up more wind farms. Continue this example down the line is that they may even choose to go to a location that doesn't

have a transmission wire, right, that's connected. So they might not be grid connected. and they might say, hey, I think that there's going to be, you know, transmission lines that have been built out here in five, ten years from now. What I'm going to do is I'm going to bootstrap this and I'm going to have some type of mining facility right at the heart of this location that has, you know, kind of a large deposit of wind that has a really high capacity factor. And I'm going to do the same model and I'm not going to require a transmission line.

So there's no wholesale meter, there's no retail meter, just your connection with your Ethernet cable for Bitcoin mining. And then in the future, you're like, well, you know, I hope that the grid customer comes to me and they'll be fabulous. I'd really like that. But I may not need them, but I hope they do come. So that's kind of an example three. So the existence of Bitcoin could actually induce the sort of anticipatory building of renewable energy assets that aren't even connected to transmission lines, but potentially would be in the future. Yeah. And I think I will correct Sean a little bit because he just gave you the explanation from a power trader perspective.

And that is the wholesale meter and the retail meter in Model 2 and model 3, there's another meter. And that's the meter between the generation asset and the Bitcoin mine because the Bitcoin mine is purchasing the power from the wind farm. So in addition to the grid connections, there is a meter between the generation asset and the load, the local load, which is the Bitcoin mine. But yeah, you could have a situation where you have no connection to the grid at all. And the Bitcoin mine is basically following the generation of the wind farm, which is actually where Lansium started. That was the problem that we were trying to address initially, which was how do we follow

the output of a wind farm? when we talk about capacity factors and so forth, those are all averages. Truth is that most wind farms are producing some amount of power 95 plus percent of the time. Now, they're obviously not producing 100% of their nameplate, but they'll be producing a certain amount of power. and so you want to match up that that consumption with with the power that they're generating

which could be standalone so we've we've I've actually had you both on for about an hour and and we haven't broached that you know the most important topic here which is I think the controllable load resource miners is controlable load and so you know when i discovered this i had a pretty like simplistic view as i guess you know sort of a common theme here um of you know mine is just tripping on and off uh in response to you know high prices on on the grid or scarcity of energy and you know allowing more critical uh you know load centers to have access to that energy during a time of scarcity and then you know i learned that there's a even more sophisticated version of this, which is helpful in sort of additional ways that are even

more complicated. So maybe let's just introduce demand response generally and then get into controllable load. Yeah. Demand response covers a broad range of responses that loads can have to demands of the grid. And so that's generally the grid operator giving instructions to loads to reduce the amount of load that they're pulling from the grid. And all the grid operators have demand response programs, but most of them are programs that require sort of 10 to 30 minutes of response time from the load. You'll get a signal from the grid operator. It says, we want you to be turned off 30 minutes from now. right um and urcot being we've talked about this the being essentially an islanded grid

they have to utilize more responsive demand response products um with requirements to respond in some cases sub second in other cases in seconds in additional in addition to the more traditional 10 to 30 minute response times um and just to give you a sort of um background point on ERCOT, their primary design goal is to make sure that if the two large nuclear power plants in Texas, which generate about 2.7 gigawatts of power, were to go off simultaneously that the entire suite of demand response products and generation response products that they have would be able to stabilize the grid frequency if that happened.

If they can't stabilize the grid frequency, you end up in what's called a black start condition because generators start turning off to protect themselves. And you get to the point where basically you have to, lack of a better term, reboot the grid. And that's a very difficult thing to do and takes a lot of time. So that general demand response, there are loads that are going to be told by the grid operator to either turn off or turn down the amount of demand that they have on the grid. And within ERCOT, then, as we get to the faster response types of resources, ERCOT built out definitions for what are called controllable. load resources. There are also controllable generation resources, but they tend to be a little more automated. I think the rules for controllable load resources, Urquhart started talking about them in about 2004.

They came up with a set of requirements based on, well, let's just think about what generators do, and loads are kind of the inverse of that. So we'll write a bunch of rules that say, if you want to be a controllable load resource, these are the things you need to do. And really, there are two primary things you have to do as a controllable load resource. The first one is called base point following. And what that really means is the grid operator wants to be able to tell that particular resource how much load they should drop. in a particular period of time. So they don't want you to just go crazy and say, okay, we're going to flip a switch and turn everything off.

They want you to ramp down according to their instructions, possibly hold for a long period of time, and then ramp back up. That's called base point following. And that's all you're doing is, I'm sitting at 50 megawatts and Erkot's telling me I need to be at 48. I need to come down to 48. Then they tell me I need to come down to 45.

and you just need to follow their instructions. So that's one characteristic. The other characteristic is what's called primary frequency response. Now, you know, we all think of the grid at operating at 60 hertz here in the United States. It actually fluctuates around that number, and there's an acceptable tolerance band around that number where no particular action is required by anybody to change the frequency. But if it goes outside of that tolerance ban, particularly to the downside, and when do you see this when like a major generator goes offline,

the frequency will drop dramatically on the grid. And so what they need is they need resources that are going to respond to that and correct the frequency. And what a controllable load resource does is we sit there and we monitor the local frequency of the grid. And when it goes outside that tolerance band, we unilaterally drop an amount of load proportional to how far it's gone outside of that band. And that helps along with generation resources and so forth to stabilize the grid. Bring that frequency back into the acceptable tolerance. span. And those are really the two things for a controllable load resource. Basepoint following,

following the instructions of Ercot of how much power to consume, and primary frequency response, independently responding to frequency changes beyond acceptable tolerance. And it's the second one that really provides more stability to the grid, along with everybody else reacting at the same time. So there's a lot there. I want to make sure I don't get my questions wrong because this is, you know, trying to learn about the insulary services is like an unbelievable rabbit hole. Although I will say that the Urquot has some good educational resources, which I don't know who they made them for, but I feel like I'm back in college or something. They made some good stuff. So yeah, you can take ancillary services 101 right from,

I've been trying. I've been trying. So I think some people are actually familiar with demand response because you have these kind of residential or sort of household demand response programs you can opt into through your sort of utility company. And so some people will already be familiar with this idea. But that's as far as I understand. That's sort of the more limited version, right? That would be responsive reserve. I'd be right about that. Yeah, that's a portion of responsive reserve. And, you know, there is as much nomenclature as you would like around these things. So typically those kinds of things are referred to as load as a resource rather than controllable load resource. It sounds like you're just flipping the words around, but no, they're

actually different things. So if you signed up for through your retail electric provider, uh, you have a smart thermostat and you give them permission to adjust the temperature on your smart thermostat so that your air conditioner doesn't run in the middle of August. That's an example of a demand response program, uh, that people can opt into and they understand that. It's like, okay, I, I get, uh, I get an incentive on my, energy price from my retail electric provider by letting them control my thermostat. And there are limits on that, you know, for two hours in the afternoon on an August weekday, you know, something like that. So yeah, there are people who are familiar with those things.

And the initial ideas for controllable load response were going to be those, the thoughts were, distributed energy resources, distributed loads, kind of those things on steroids, right? So, you know, aggregating a bunch of different thermostats, aggregating motors and pumps and things like that in industrial situations. And what they found was, one, it couldn't respond fast enough to do the primary frequency response. And particularly when you got to things like motors and pumps and so on and so forth, you know, if I change your thermostat so your air conditioner doesn't come on, that's one thing. It doesn't hurt your air conditioner. But if your air conditioner is running and I flip a switch to

turn it off, that can damage the motor, right? So there were a couple of aggregation type models that tried to qualify for controllable load response. And they weren't fast enough and they caused damage to the elements that they were trying to control. So when we came along, it was, you know, we worked with a number of folks and we got a response that we were kind of, I use the term, the rainbow unicorn that they were looking for. It's like, we never thought this would happen, you know, but I'm. Obviously, you can do it.

And they were pretty excited about it. But it was a different model. We're talking about a data center, which is not distributed. It's centralized in one location. And we're controlling the elements of that data center to control the power consumption. Now, we actually control multiple data centers. So that's where the distributed element can come back in.

But that wasn't part of the original thought process. So people can relate to demand response. The controllable load response is a little bit different. And it was the solution that we came up with was actually completely different than what most of the people in the space had thought about previously. So I guess what happened here, the chronology is that Urquat laid out theoretically what a controllable load resource would look like. no one qualified for that for years and years.

And then eventually you guys came along and were like, hang on, Bitcoin miners can actually do this because they can be dialed up and down. They can be turned on and off without any real issues because each hash is independent of the next hash. Just as a sidebar, I love how the statistical properties of proof of work actually allow this to happen with the hash as being IID.

But proof of work is going to be the end. of the world. Right. So because of because of the way the Bitcoin miners work, you can ramp down your consumption on a short-term basis and not really suffer any adverse consequences aside from, of course, like the marginal loss of revenue there. But that's just not how other industrial processes work. Like a smelting plant, you couldn't just turn it off. I mean, I guess you could. You guys were telling me. There were other industrial sources of load that had some similar related properties like paper mills, is that right? Sean's a little more familiar with paper mills because they had an example of that up in Canada. Yeah, and it's just a certain type of industrial load

in Alberta that if you need to have, you know, like as a power generator, you can either be ramping up power. So like, you know, an easy way to understand, understand demand response. is the purpose of power generation is two things, is one is just to provide the generation that's going to exactly match what the load is. So that's kind of like load balancing. And the second thing they do is they provide insurance as a buffer to make sure that the heartbeat of the grid stays at the same range. So they do two things, right? Just exactly load follow what the demand is and provide this buffer. And so think about demand response is just exactly doing those things, but having load try and participate in those pieces.

And so like there's different programs, but they're doing either load following or kind of these different and sorry services. But essentially now, you know, they're calling on load to be a bigger part of this energy transition where they can't just be a dumb load, but they need to say, hey, how do we get creative in addressing these issues

as we're transitioning to, you know, a decarbonized grid? In Canada, you know, there's, my experience has been that, you know, there's, as a generator, you can either ramp up power or you can find a way to ramp down load. And so there's some instances where, you know, some large industrial customers are willing to turn down their power. An example would be like a paper mill. With a paper mill is that they use a large amount of electricity.

You'd have to call them up and you need to give them a lead time. But really they can, you know, stop their process and not be in too much harm. If the prices make sense, if it's economic to do so. So the challenge with that is that, again, it's like a lead time that they need to have be notified. They can drop their load. And that's kind of a slow kind of ramping back up again to get back to the normal kind of output.

And so to your point there, Nick, it's just like never been for, has there been this, you know, perfect commodity where, you know, all the work that you did up right up until that exact moment you're being compensated for. And then as soon as you kind of turn back at the production, you're being compensated again. So there are some, you know, some analogs and kind of analog punch. lines that you can start to think about kind of, you know, as Bitcoin providing this, you know, service that was never available in the analog version versus the digital version, which is Bitcoin mining. And the other thing that reminds me of is, you know, aluminum smelting is the forerunner

to Bitcoin mining because, as you guys mentioned, a lot of, you know, aluminum was a sort of like way to monetize, like excess energy. And so you ended up having these smelters in places like Iceland or Messina in New York with all the hydro there. And then, of course, a lot of those are now Bitcoin mines, including, I think, in Rockdale, Texas, or a lot of Bitcoin mining occurs as a former Alcoa plant. So I just love the analogies where you had a physical process that, you know, did a specific function. Now Bitcoin does that in a more efficient way. I know we only have about 20 minutes left. I want to get into the actual economics of this because there's a journalist, I think, an arts technologist.

that wrote an article recently saying that the economics didn't work and why would Bitcoin miners ever turn off their equipment and the price of energy would have to be really really high to get miners to turn it off but and I think that's actually a bit of a stumbling block when I talk about demand response people like well why would miners ever turn off their equipment why would they you know do this altruistic function for the grid but in fact it's not altruistic it's sort of just plain on economics so tell us a little bit about how the economics work in terms of qualifying as a CLR and agreeing to sort of trip off your equipment,

should it be necessary? So one of the things to keep in mind just round numbers for today, the break-even price for S-19 Pro, I think I said earlier, is $500 a megawatt hour. That's how much it's generating. And so therefore, when your cost of operation, gets to that point you're going to want to shut off, right? The other thing is that today, if we look at the responsive reserve service, which is one of four ancillary services,

and is the one that we typically offer into, on average today, that responsive reserve service is paying $30 a megawatt hour. and that is in essence a call option on the power that as a controllable load resource, the power that you're consuming. So with any call option, you know, one of the questions is, well, how often does it actually get called and under what circumstances? So when we look at the two characteristics, first of all, the base point following.

Well, when is that going to happen? That's going to happen when there's a grid emergency and we don't have enough capacity and Erkot wants to reduce the amount of load that's on the system. So an example would be the winter storm we had here in Houston this year. Those things don't happen all that often. But when they happen,

Erkot has to have the ability to respond. So in that case, you'd say, well, it's not going to be good for the Bitcoin miner because they're going to get shut down for an extended period of time. Let's say three to five days. And typically, you know, hopefully no more than three days. Okay. So when you factor in three days out of an entire year, that's less than one percent for the entire year of the time. So you use the $500 a megawatt hour, say, okay, I'm not going to be generating $500 a megawatt hour during that period of time. So it's going to reduce the amount that I make by, let's just call it 1%. So out of $500 per megawatt hour, you're only going to make $495 per megawatt hour.

It's not going to kill you. Right. But the important thing on a major event like that is you've also purchased a block. of energy. And if you're not using that, that actually gets sold back into the marketplace. Well, if I'm buying that block of energy and the number that Sean put out earlier, say, around $30 a megawatt hour, and now of a sudden there's an emergency and Aircott's trying to provide the right incentives for people to shut down their loads, now of a sudden the price of power goes up to $2,000 to $9,000 a megawatt hour.

So would I be okay shutting down my $500 a megawatt hour Bitcoin generator to get $2,000 to $9,000 a megawatt hour for just the power? I think those numbers are pretty simple, right? Right. And you're also getting paid to subscribe to offer up your load in that time of, of, you know, grid stress. So you're getting paid generally to participate in, you know, well, you're selling a string of call options to the grid. Right. So I'm selling a string of call options, most of which will never be called. But when they do get called,

I can sell my energy for a price that's much higher than the opportunity cost on, on that call option. So then the other part of the controllable load resources, the primary frequency response. And so you look at that and you go, okay, well, how often does that happen? It happens a few times a week, but it's, you know, a couple of minutes here, a couple of minutes there, so on and so forth. I did some mental math this morning and said, well, what worst case, let's say 15 minutes of every day we had to shut down for primary frequency response. funds. Again, that turns out to be 1% of the day, right? And we don't turn down 15 minutes every day. It might be 15 minutes in a month, right? And on partial load, right, right?

And it's partial load. And it, you know, so, so the economics, when you get into the details of it, we're just going at high level economics, but when you get into the details of it, you know, they're buying insurance, Urquat, that is. We're getting a revenue stream because they're purchasing that insurance from us. And the economics of the particular events are not very negative for us. Maybe I'll just kind of add about, you know, like the Interior Services back to kind of generation provides low balancing in this, you know, kind of insurance buffer is that, you know, the reason why it's, super incredible for controllable load resources is that what we're doing is we're competing against generations such as natural gas and coal plants to provide this service, right?

And they have a true opportunity cost, right, of providing that service. So for example, if 100 megawatt natural gas plant, you know, wants to operate and sell 80 megawatts of energy, well, first of all, they can only sell 20 megawatts of the sensory services. It's a longer story just on droop factors, but just for this, examples that they can only do 20 megawatts. So they need to decide, you know, as a power generate, do I sell energy or do I sell the insurance, the insulary service? But when you sell the insurance, you're not on most of the time because you have, you're qualifying yourself to give the option to have the grid to ask you to turn on strategically,

but mostly you're not on. So it's kind of expensive to sell the insurance for you. For sure. So they have to have a starting point of saying if they sold 20 megawatts of insurance, they need to be sitting at 80 megawatts so that they can ramp up to the 100 megawatts. Whereas a controllable load resource has an unfair advantage, where the generator has to choose energy or the insulary services, where the controllable load resource is an and statement, where they get to buy the energy and sell the insurance.

So essentially they're consuming energy all the way through and they're selling insurance. And to raise point that then they're being called upon for the, you know, the handful of hours in the year, which makes total economic sense to do so. So TLDR on all of this, it does make economic sense to, you know, participate in these programs. In terms of, I'm just curious, like, ERCOT, are they sort of aware that it's Bitcoin mining that qualified for CLR? Do they care? Or do they just think, oh, it's like remarkably flexible data signers. No, they're aware of the fact that it's Bitcoin mining. We had a number of conversations with them before we qualified and explained that our ability to do what we do is based on the inherent

characteristics of Bitcoin mining and the ASIC machines. And so, you know, as far as ERCOT is concerned, They don't really care one way or the other, whether it's a, you know, Bitcoin mining load that provides these services or, you know, a bunch of motors somewhere. What they do care about is the stability of our ability to respond. And, you know, we've been tested through the winter storm. we've qualified a number of times where, you know, they monitor, like any other controllable load resource, they monitor our performance. And so, you know, that's what they're going to judge us on, not whether it's Bitcoin mining or something else, is, are you providing the performance we need on a consistent basis? And we've been doing this for 18 months, several different locations, different size loads, and growing. And we'll be building out our own campuses much larger load format.

But yeah, they're aware that it's Bitcoin mining. And is there a process of learning for other? grids because you know I guess they have to some of them have to design these programs that may not exist is it just a function of other grid operators learning from the arcad experience and saying oh wow this is interesting now new you know load source that you know so we should build our own CLR program something like that yeah it's it's interesting because a number of other grid operators similar to where Irkott was, have a category and some rules in place that could be used, but most of the time they just, you know, they defined that, they put it on the shelf.

They said, you know, when something comes along, we'll think about it. And so, you know, it is a process of education of saying, you know, those rules that you wrote and put on the shelves in some cases, we can actually do those things. So let's figure out how we move that forward. In other cases, they don't have the rules. And that would be a longer process of working with them so they understand the characteristics, the benefits, and so forth, and then working through whatever their process is to define rules and then services from those rules. So just to summarize, you guys had a press release recently with an interesting paper talking about how flexible data centers, flexible load, what you do supports the greening of the grid. And Sean, I guess that was your point about how generally speaking natural gas turbines are the things that spin up and down to sort of match the demand.

Now, you know, the ability to do CLR can sort of supplement that and maybe even replace some of that, you know, fossil fuel. base generation, just give us a bit of a summary in terms of the sort of net effect of your presence on the grid, both in terms of monetizing some of these renewable assets with a new independent buyer and then, you know, being this great resource to the grid. Sure. So in the example that was done on this white paper was as it was adding five gigawatts of data centers to West Texas and having more flexibility, which I believe was defined. is about 13% of flexibility.

And the net impact was in building out these 5 gigawatts out to 2030, is that you incentivized more renewables to be built out so that at the end, there were more megawatts being produced from renewables than you were consuming for a data center. And that net effect was that by 2030, you'd reduce CO2 emissions by 4 million tons per year. So that's kind of the impact on the grid, which is incenting more, which ties into that kind of incentivizing, incentivizing kind of renewable economics while also decarbonizing the grid.

And then also in that, by adding these 5 gigawatts of data centers, you're also providing those grid reliability products around the primary frequency response, backup power, being able to provide for this increasing amount of renewables that's coming online for backstopping renewables in case there's a sharp drop in generation on their end. But overall, just hardening the resiliency of the grid. while adding these data centers and decarbonizing the grid.

So Ted Cruz's comments in Austin, you sort of aligned with those. Think he was basically on the right track. I thought he was very on point and had some great speaking points. I was really surprised how well informed he was and thought it came across really well. So I guess looking forward, I mean, you know, we're still in kind of a land grab here with miners. You know, there's still that big arb you talk about, you know, between the cost of mining, Bitcoin, the price of Bitcoin, I mean, you know, let's, I don't know exactly how many gigawatts of Bitcoin mining are active in Texas, probably, I don't know, 500 megawatts may be as much as one gigawatt. Just guessing, if it grows to four or five gigawatts, I mean, which I think is possible in the next few years, will the grid be able to accommodate that? I mean, will that improve the reliability of the grid on a kind of net basis? So I say that depends on where it's located as long as you're not locating in an area that's constrained for existing transmission.

So almost anywhere in the grid is good, but there are some specific spots that wouldn't be grid. Good. Important also know that Texas is also adding, Urquod's adding a significant amount of renewables over the next 24 months. So in tandem of saying adding four gigawatts mining over 24 months, you know, essentially that there's going to be another 32 gigawatts of wind and solar over the next three years. So I think at the end of 2020, there is 35 gigawatts, which was 31 wind for solar. And by the end of 2023, they're expected to have 67, which is 37 wind, 30 solar. So taking that back is saying that, you know, if you're adding 32 gigawatts of wind and solar

and say the capacity factor, easy math on this 50%, then that's 16 gigawatts of a generation that's coming online, we're already seeing these price splits happening in the grid. So four gigawatts of Bitcoin mining put into the right locations would be a happy customer for those megawatts. And I think that there's a good chance that we start to see more splits over the next couple of years just because this renewable spilt out has been on track for 20 years. And projects coming online now have been thought about several years ago. And so Bitcoin mining facilities aren't at scale yet.

And so it'll take a while for those projects to get to scale, to get to the point where they can keep up with the build out of renewables. So I also think that, you know, kind of adding 4 gigawatts of Bitcoin mining in West Texas would, again, go back to, you know, at least securing that price floor a little more so that it's not so negative. So it's also helping with that price economics of those wind farms that are getting that close to 50% discount versus kind of that flat block of power. I think it would be a net benefit to the grid for providing that resiliency products, improving

economics for renewables. And I don't think it would actually be keeping pace with the amount of renewables that are coming online. But I think that likely we'll probably overshoot on the other side for Bitcoin mining afterwards. And it all balances out in the long run. Last word from you, Ray? Yeah, I think the thing that's exciting about what Sean just described. And I think this is part of the, you know, Urquat benefit Texas attitude is we're talking specifically about Bitcoin mining and, you know, the benefits both ways to the increased renewable generation. You know, there are other industries and so forth that are going to be looking to take advantage of that as well.

And so from, you know, as a Texan, the economic outlook for Texas of having all of this power available is very attractive to companies. And we've seen a number of companies relocate to Texas for that reason and others. But, you know, we're also looking at other applications in the industrial space that could possibly benefit from the power and the other. ability to provide, you know, controllable load resource type benefits. So it's pretty exciting. I mean, it's, you know, we're going to have a tremendous amount of renewable energy, and that's going to spawn, I think, a lot of innovation on the business side as well, not just in Bitcoin mining.

So that's pretty exciting. Well, once you get through the jargon, I agree. exciting. You've got to cut through it first, but I think you both have done a masterful job today. This is one of our longest podcasts, but could have been much longer. So I want to thank you for your time and best of luck, you know, being the guardians of the grid down there in Texas. We'll do our best. Thank you. Thanks for having us. Yeah, thanks for having us the show.