Assuming you guys are all familiar with the webinar, we stay safe. If any of you would like a copy of the presentation, you can see Kim at the back of the room, and we're happy to hand that out. I'd like to start out by saying that at Capstone , we've weathered the worst, and now we're built to win. The financial storms of 2023 didn't break us, but rather they made us stronger. Now we have the foundation that readies us for the next step change of future growth. In 2024, one of the first things that we did was launch a major strategy that we call the three pillars of strength: financial health, sustainable excellence, and revitalization of culture and talent. Those three pillars are what has helped us deliver over 30% revenue growth over a trailing 12 months, margin expansion across every revenue stream.
As a result of that, we've delivered six straight quarters of positive adjusted EBITDA. Because of that work, we've now unfurled our sales, and we're ready to take in the winds that are upon us, the tailwinds of distributed generation. Those tailwinds are primarily made up of four main things. The aging infrastructure of utilities is causing grid fragility and resiliency issues. The technical maturation of both efficiency and control software allows onsite generation to be a very viable solution to help utilities. Another interesting one is the decarbonization without disruption. As the movement to be green, the realization that you couldn't get capacity factors above 15% or 20% caused a real problem in sustainable low carbon footprint energy generation. That's something that Capstone can do very nicely.
Lastly, the electrification of load growth, both in the EV space and in the utility space, where substations, especially due to weather storms, fires, and things like that, have put a real strain on the substations and transformers that the utilities have today. These tailwinds are real. What these tailwinds are doing is giving us basically three acts to play on the stage. Our first act is something that is our base that we've been doing for over 30 years, and that is our distribution model. We have the ability to have local expertise in every single market across the globe. The second act is just recently, you've probably heard about our announcement where we acquired one of our major distributors in the West Territory called Cal Microturbine. That acquisition allows us to go to what we call territory direct sales, which creates margin expansion at very low go-to-market costs.
In addition to that, we've also launched a Global National Account, or GNA, program that not only helps our distributors, but also helps customers that are looking to work directly with the OEM. We won't stop there. We'll also bring our local expertise of that distribution model to those GNA accounts. Lastly, this is what brings us to the big sky. That GNA program will help us deal with some of the major plays that are out there that are driving these tailwinds of distributed generation. As we looked at all of the tailwinds that were out there and looked at total addressable markets, we've decided to just focus on four. These are huge. Obviously, data centers are a big one, and it's the talk. We're going to get into that in a little bit.
I mentioned earlier about electrification, and in particular ports and terminals and terminal operators who made the change to electrification, but those terminal owners didn't have the power infrastructure to support that electrification. Power on the go and temporary power is a very real solution for those folks. We believe we have something that can address that need. I mentioned substations and transformers and microgrids. That's the onsite generation. More and more customers are starting to look at real either island mode power generation or connecting with the grid. The technology today makes that very viable. In fact, for us, we create something we call grid enhancing for the utilities. We can not only be island mode for that customer, but when we do need to tie into the grid, we can help the grid enhance its stability.
Lastly, this is a sleeper for most, but for us, we see it as a very exciting one. That's called station power. As much as the utilities need to move quickly because of all of these demands from these tailwinds, it's hard for them to add big blocks of power very quickly. We've taken a look at combined cycle gas turbine power plants, and we have a solution that can help those plants increase peaking power by 10% to 25%. That's a really exciting place for us to play. As we looked at this and we look at our company today, it's a very nice business, and we've done a lot of great things to make this a solid foundation of a business.
When we drive those three pillars of strength: financial health, putting discipline into our business and everything we do commercially, financially, operationally, and business-wise, creating sustainable excellence and driving repeatable business models that create margin expansion, and really focusing on our culture and talent. At the end of the day, it's about the people. If we have a filter that drives the right people to come work for us, we can do amazing things. By driving that three pillar of strength, we believe we can capture just enough of that tailwind and just that small sliver. When you saw earlier in the slide $304 billion of total addressable market, if we just get a small sliver of that, we become a very, very exciting company. That's what we're working on. Four areas with focus. A little bit about Capstone .
We are not a startup company. Capstone has been around for over 37 years, proven technology. We have units out in the field today that have been running 20 to 25 years, not one or two units, many units. That's something that's really important because it's built a service base for us with over 10,000 units deployed worldwide. Because of that distribution model, our local expertise is what keeps those units running and keeps that base strong. As we look out into the future, this is something that is very real for us because we've been here and we've done it. If you look back in the history of when Capstone started in the late 1980s, there were a number of Fortune 100, Fortune 500 companies that were trying to do microturbine applications. All the big players were on the scene.
Over those 30 years, they've all either given up or walked away from it. That's something we like to say at Capstone is, we exist because we made the impossible possible. We did that by really believing and striving to be the first choice when energy matters. We didn't give up and we didn't give in. We kept iterating on success. That is how we delivered these four key pieces of IP that make us different from many other power generation technologies that are out there today. Our proprietary recuperators not only improve our efficiency of the gas turbine cycle, but they heat the compressed air, which then allows us to combust at much lower emissions levels. We are a very, very clean technology. Now our air bearing, which allows us to take one single moving part spinning at 60,000 RPMs with no oil, no lubricants, again, very clean.
It makes our emissions the lowest out there. Our combustion technology, same thing. The way we introduce air and mix it, we are able to deliver the lowest in NOx at 9 PPM and CO and CO2. All the technologies that are out there, gas turbines, recip engines, even what we're learning about fuel cells, even though they seem to be clean, we're finding fuel cells actually have an issue with CO2 and then how they clean that up and what you do with that after treatment. This combustion technology has been proven and it's been refined. The last one is a very interesting one, and we're going to get into this a little bit, but this is our power electronics. It's been around. The architecture has been around for over 30 years. It works. It's tried. It's true. It enhances the grid, which many technologies don't do.
In addition, we've got some other gold nuggets that I'm going to share with you in a little bit about this power electronics technology. Let me shift now to data centers because I know a lot of people like to talk about data centers. The question is, how does Capstone play in the data center space? You don't have a big 50 megawatt gas turbine or a 12 megawatt recip engine. The biggest engine we make is one megawatt. One of the things that we realize data centers need to do is two things. Number one, they're chasing five nines of redundancy and resiliency. Number two, they apply a concept of lean manufacturing to their data centers. That is, every kilowatt that they use, they want to put it where it brings value.
That value to their customer is at the server rack, not at generating chilled water for cooling the server racks. That's something they call PUE, and they chase their PUE to one. When we brought industry experts in and we decided to come up with our reference design for data centers, which by the way has a very high power density of 45 megawatts per acre, so much better than gas turbines, recips, we compete with fuel cells, much better than linear generators. With that concept, we've put three one megawatt C1000 engines derated at 110 degrees Fahrenheit because that's the way data centers operate to provide a power block that they typically use of 2.4 megawatts. While we're doing that, we're recovering the waste heat of those engines and converting that into useful chilled water for cooling.
We do that at one tenth the energy consumption of an electric chiller. What happens here is while we reduce the electric load of those electric chillers, we're generating power, thus creating an energy surplus. That goes to the server racks, which add value to the data centers. That is our concept that we call Energy Surplus Program or ESP. One of the other important pieces for data centers is they like to deal with one entity, one throat to choke, if you will. That's why we're offering an engineered equipment package. We're not just selling the microturbines. We are packaging this and designing it and creating a selection that matches that absorption chiller with our microturbines as well as the heat rejection. We're also offering battery energy storage. As one engineered equipment package, we're able to deliver a total solution to that data center.
We got the reference design and the technical concept down. Does it make financial sense? It was the next question we had to ask ourselves. We developed this quick, rough, crude, we call it a revenue-based pro forma. If you had a data center that was not going to get power from a utility for three to four years, how did they get up and running in less than a year? This is where our solution could come into play. What we did here is we modeled a 12 megawatt data hall. With that, even in the revenue stream for that data center, we went with a very conservative number of $100 per kW per month. Lately, that number is $150 to $200 in terms of what the data centers can make in terms of revenue.
When you look at all of the operating costs and you distill it down, what this solution delivers is something a little over three years payback with an exciting 32% 10-year IRR. You're in market within two years. Financially, this definitely proves out. I'd like to shift to something that's really, really exciting. Some of you may or may not have heard the news this morning, but we've announced this. There's a major shift in the data center world, and especially when it comes to AI or artificial intelligence. Last May, NVIDIA announced their whole new infrastructure is going to be based around operating on 800-volt DC power, direct current. There are a few reasons why they want to do that. They need that for scalability and future proofing.
They also see that by going with DC power, they're able to reduce the amount of copper that's needed in these AI factories, as they call them. Lastly, they're able to create more efficiency because there's less conversion losses. Basically, what's happening is you take power from the grid and they convert it multiple times, step it up, step it down in order to deal with the different types of equipment in the data center and then deliver it. There's no simple solution that says I'm going to take AC power and deliver 800 volts DC. That's been the big challenge at hand. As a matter of fact, a lot of big players out there, Eaton, Vertiv, Schneider Electric, are all trying to figure out how they are going to solve that problem for NVIDIA. In 2027, NVIDIA is delivering all their compute racks on 800 volts DC.
In the end, if you really think about it, Edison was right and Tesla was wrong. DC is the way to go. Now, I'm not talking about Elon because usually he's always right, but Nikola Tesla in this case does not win in this situation. This is where the AI factories are going. Now, here's what we've done. What's really neat is that power electronics piece that I was telling you about. One of the little gold nuggets that was in there that we didn't even understand what it did was we are now able to deliver that 800 volt DC power.
The reason why we can do it so easily is because when we are spinning at 60,000 RPMs generating AC power, we had to convert that to DC power to clean it up and get the frequency down so we can convert it back to AC power so that we can deliver 415 volts AC at 60 hertz. That's how our product today works. With a simple solution, because we were delivering 760 to 780 volts DC today out of our product, we call it a generator control module. All we did was a simple software change and we are now able to deliver 800 volts DC. This is big. Nobody else can do it. Some linear generators possibly can do it. Fuel cells possibly. Gas turbines, reset engines. They're all low speed AC power.
They have to go through a much stringent, rigorous conversion process to basically get that to 800 volts DC. We do this with a simple software change. That's why this is exciting. That's why we're talking to a lot of the major players and they're excited for what we've been able to prove. This is where our future is today. It's not only a great distributed generation company that's been proven for almost 40 years, but we've got technology that existed all this time, and we're now able to deliver to the future of AI factories. At this point, I'm going to open it up to Q&A.
On your slides can show growing revenue, if I understood correctly, $100 million to $1 billion. How much capital will you need to get it to point A to point B? How will you raise that?
Raise that capital. The question was that one of the slides showed going from $100 million roughly to $1 billion. How do we grow and how do we raise the capital to support that? What's very interesting about our business model is that in our factory in Van Nuys, California, it is mainly assembly. We've pushed all the special manufacturing transformation processes to our suppliers because that's typically what you do in the aerospace industry. A lot of our components are from the aerospace industry. Their scalability, we're only scratching the surface with them at 20% of their capacity. For us to build that kind of size, our factory today, after we finish our lean manufacturing floor layout, we will be able to do seven megawatts a week with one shift. Call it 50 weeks a year since we're going to give some people two weeks vacation.
That's 350 megawatts a year for one shift. Go to two shifts, you're 700. Go to three shifts, you're a gigawatt just in that one factory. The other really interesting thing that we're able to do, and we've been talking to a couple of data center customers about this, is mobile manufacturing. OpenAI has been pushing hard for some of their key suppliers to be able to manufacture on their campus because it's a long build-out. Because of the way we do things, we don't need a lot of special equipment to build our product. Therefore, we can do mobile manufacturing and stand up a factory in no time. In that particular case, if we just want to do a gigawatt out of our Van Nuys factory, there's not a lot of capital that's needed. It's more about people than it is about equipment. We don't have a lot of specialty equipment we need. The CapEx there is very low. John, I don't know if you want to add to that in terms of how we've modeled it.
Yeah, we look at it and we model it. There's some insourcing of materials to make sure that we have enough. From a capital standpoint, it's not much more than what you would expect from a maintenance capital program to continue to grow the business. It's more, as you had said lately, it's the key factor.
I should introduce John, who is our CFO for Capstone , John Juric. There was a question. There was another question. Yes.
I see that you moved exchanges recently. Any plans to go to the NASDAQ?
There is. John, you want to take that one?
Yes, we do want to get back onto the NASDAQ.
Oh, have him come up?
We're working towards that. We have a plan. We went through our Chapter 11 restructuring. We got removed from the NASDAQ as part of that. We got put on the clean sheets. We've moved up to the QX. Our next phase is we have to drive some additional profitability, help more quarters of profitability. We're going to go and apply for some of the exchanges. We're going to go to exchanges.
Would that be the NASDAQ or another? Would that other exchange be the NASDAQ or would it be the same thing?
We're evaluating that. We're not certain it'll be NASDAQ, but one of the same three, for sure.
Yeah.
Would that come with an IPO, or would it be able to do it without an initiative?
There's discussions about re-IPO. We're talking with the buyers of different ETFs, right? Part of it is re-IPO, raising capital, elevating our equity element. There are other methods to do that with business combinations that could be attractive for the things that we want to do strategically. It's all in discussions at this point in time.
Yes, sir.
Can you talk about some of your sales pipeline efforts post-execution? The turnover DC is pretty unique. Are you guys in talks with some hyperscalers?
Sure.
If I want to give you a lay of the land.
Okay. The question was, could I give a little bit of the lay of the land of our sales pipeline, especially in the data center space with hyperscalers and so forth? We've been at this a little over a year. When I first came to Capstone in March of 2024, that was the first question I asked. Why aren't we doing data centers? Our distributors back then said, oh, they need too much power. We can't do it. That is part of the reason why we invested in data center experts. I had some ideas from my background. We developed this reference design. From there, we've now been talking to a lot of the big hyperscalers, you know, ones that start with an O, one that starts with an E, the list goes on and on. Our pipeline is quite big. These data centers are very pragmatic.
You've got to continue to work and solve their technical concerns. The first objection a lot of them had was absorption chillers. They're starting to realize now absorption chillers are actually more reliable than electric chillers because you don't have a compressor. The other objection was power density. I remember we were talking to some folks up in the Bay Area that start with an E, and they said, we didn't think of using you for 75 megawatts because that's a lot of engines. When we explained that at 45 megawatts an acre, that's nothing. He literally wrote that in his notebook and said, I made a mistake. There are things like that that we've gone through over the year and a half. Now our pipeline is really starting to mature. We're seeing some very exciting opportunities. Some of them are four, five, ten megawatts.
Some of them are 50 megawatts. Some of them are 100 to 200 megawatts. When you start to look at a 100 megawatt project, and because we're delivering an engineered equipment package, that revenue stream is more than just microturbines. We have access to the chillers and the dry coolers and the battery energy storage that will be white labeled for us. We deliver a very nice revenue stream that makes that billion dollars happen very quickly if you land something like that. It's exciting. Did you have a follow-on?
Yeah, just a follow-on question. Are you actually in trials with any of the hyperscalers?
I'm sorry?
Are you in trials with any of the hyperscalers?
Are we in trials with any of the hyperscalers? We are talking to two that we are doing pilots with. We're about to do pilots with them. We haven't announced it because it's not finalized, and one of them is with the 800-volt DC pilot as well. Okay. Okay, we're here to answer. Yes, thank you for your time.