Coming up on Path to Zero.
We believe that this technology can let you make your own electricity cheaper than you can buy from most of the grid across the US.
Welcome to Path to Zero, the podcast about clean energy and the journey toward a low carbon future. Your host is Tucker Perkins, president and CEO of the Propane Education and Research Council. Now, here's Tucker.
Hello, Path to Zero listeners. Today we're talking about one of the biggest challenges in energy: how to deliver reliable, flexible power as demand surges from AI data centers and electrification. Our guest is Thomas Healy, founder and CEO of Hyliion. The company has made a bold shift from trucking to modular power generation, focusing on delivering scalable, fuel-flexible energy where it's needed most. Thomas, welcome to Path to Zero. I'm looking forward to the conversation.
Thanks for having me on, Tucker, as always. Great seeing you and looking forward to sharing more with your viewers about what Hyliion is up to.
Well, let's start right there. I mean, not only is it almost every conversation today around producing power, but you're producing power, I think, in a very unique way for people who really are interested in getting power, but being very concerned around environmental. So talk about maybe Hyliion, talk about this KARNO generator. Kind of give us a primer in your everyday world.
Absolutely. So you're right, power is the discussion that everyone's having right now. Data centers are leading the charge for sure. But even past that, like industrial applications like oil and gas or even just commercial buildings, right? Everyone is seeing their power rates go up and it's happening fast, big percentage changes each year. And so people are starting to look at, well, if I can't get the electricity I need from the utility, can I make it myself behind the meter? Or if the electricity I'm buying is just too expensive, can I make it cheaper myself as well? And that's where Hyliion comes in. We've developed a new way of making electricity, and the technology that really enables behind the meter or distributed grid power generation. So think of, you know, a commercial building every hour. Most commercial buildings have air conditioning systems, just like there's a box outside making that cold air. We want to put another box right next to it that's actually taking fuels like propane or natural gas or hydrogen or diesel and making electricity. Now you think of a normal generator, you usually think of like emergency standby power. Our business model is this is prime power. If you own one of these things, this is the way you should make electricity and remove your dependency from the grid and make your own electrons. So that's the very high level. Obviously, we can dive deep into it, though.
And let's talk about something very interesting about your technology, because I believe you used the word flameless. Certainly we don't talk about combustion like we do in a traditional generator. And that leads to a really significant benefit for emissions, which, by the way, maybe the unspoken secret about so much of the power generation that's going on today. Much of it is fueled by diesel power, and the emissions from it are relatively bad. Tell us about what flameless, and maybe explain to the listeners about this unique combustion method.
Sure, absolutely. So I'll maybe just start with like when people hear generators or power generation, you usually think of an internal combustion engine, a turbine, maybe a fuel cell. With ours, it's actually based on Stirling engine technology. So, little history lesson here. If you go back 200 years ago, there was the steam engine. That's what was being prevalently used. There was also this idea or this invention of this thing called the Stirling. Even 200 years ago, it was known that Stirling was the superior way to make power. It's got the ability to be super efficient, very low maintenance. The kind of the two main buckets that you look at when you're looking to make power. And the problem has been, though, over the last 200 years, it hasn't been heavily utilized. There's a few use cases here and there, but for the most part it was way too difficult to manufacture. So what we've done is we said, all right, superior way to make power was invented 200 years ago, hasn't been used because of manufacturing. Well, there's been a major breakthrough in manufacturing thanks to 3D printing or additive manufacturing. So what we do is we took this 200-year-old concept, brought it to today and said, well, let's 3D print the really complex parts. And then that way we can actually bring forward this superior solution. And so that's why we're able to get like power plant level performance out of a very small box that can go at your facility.
To your question though, around the flameless side. So the whole concept of a Stirling is you have trapped gas and all you want to do is heat that gas up and cause it to expand, and then you cool it down and cause it to contract. So heat and cool and heat and cool and you do that like 20 times a second. And we use that motion to actually make the power. Now the reason that we call it a flameless oxidation process is because we have a proprietary way of taking that fuel in and actually reacting it to make heat. Now, what we do is we suppress oxygen levels, we keep the temperature right at a sweet spot where it has extremely low emissions. And it's actually this unique reaction process called flameless oxidation, where if you could actually be on the inside of our KARNO power module, you actually couldn't see a flame. But all the metal around you is glowing because it's making heat. So it's this really, really cool concept that allows for extremely low emissions and fuel flexibility.
You know, it's just fascinating. And you're seeing the market come to you. Now on paper, when people come to you, are they interested in your KARNO power module? Are they interested in the fact that you have such a wide range of available fuels you can use? Are they interested in you because of your emissions? Or maybe all of the above?
Yeah, it truly is all the above and it depends on who the customer is, right? So to go through some of the use cases, like data centers. Data centers, the initial question is how fast can I get power. And then the second question is what's my cost per kilowatt hour. Since they're consuming so much power, they want to make sure it's low cost electricity to be financially effective. Another bucket is the military. We work closely with the Office of Naval Research. So the Navy, we're also expanding programs with other branches of the government. All of our work with them actually started on the low maintenance aspect. So the Navy is actually putting us first on a fully unmanned autonomous ship. It's like this 200-foot ship, no crew on board. I got to board the ship a few months ago. It's fascinating. All it is is these engine bays where our technology goes in and then fuel all around, it's stored in the hull. And then on the deck of the ship, they can put whatever weaponry or monitoring systems they want. But what's unique about it is our system has an extremely low maintenance profile, and it doesn't have any oils, any lubricants.
So when you think about like a diesel engine, conventionally, normally you have to have someone there that's constantly doing oil changes, versus with us that doesn't exist. And that's really important when you don't have a crew on board to maintain the engine. Then we have other applications in California where it's all about emissions, right? It's about making sure you meet the air permits and the air quality district mandates. And so that's where our emissions benefits are big. And then there are other applications where it's the fuel flexibility. They want to be able to switch from diesel to natural gas. So I'm going on and on. But you can see like this technology has a lot of different benefits, but it really depends on what the use case is as to what the main driver for their adoption is going to be.
We talked about fuel flexibility. You mentioned earlier diesel fuel, natural gas, propane, and hydrogen. Is that pretty much the list?
So we've actually identified over 20 fuels. The list goes on. I mean, dimethyl ether, JP8, ammonia. I mean, it's a long list that we can run.
And that begins to give you tremendous flexibility as you think about different countries. I don't know if people realize how much fuel varies. Fuel in Africa, gasoline in America, relatively different than diesel fuel in Europe or South America. So all of a sudden, if you're using liquid fuels, you've got a lot of options, but it provides for you to be a solution really across the globe, right?
Yeah, very true. And it also just provides this flexibility of if you're someone that is going out and buying this power plant, hopefully it's going to be in existence or in operation for the next ten, 20, 30 years. Right? I mean, these are long-term capital purchases. I don't think any of us can bet on what the fuel of choice is going to be over, even maybe the next handful of years, as opposed to the next ten, 20, 30 years. And so, you know, a few years ago, there was a huge push towards hydrogen. Now that has pulled back. Great.
Obviously there's the kind of fuels we conventionally go to of natural gas and propane and diesel, but maybe there's other fuels that we're not yet really considering are on the forefront. So if you can go make a capital investment that you can look at and say, look, no matter what fuel we have availability to or no matter what fuel is the lower cost fuel of this time, I'm going to be able to run on it. That provides you a pretty secure bet on your capital investment versus if you were to bet on diesel engines and diesel prices like we've seen recently, have skyrocketed. Well, maybe that's going to be financially no longer the right choice. And you're going to be looking at doing another capital investment in another technology. So it also provides you, like you mentioned, Africa, the fuel can change even in the U.S. like an oil and gas site. There's different purity levels from one wellhead to the next. And for us, we can usually in most instances handle those fuel differences versus not all engines can.
I know you have several sizes of power outputs, but let's talk a minute. You find that you are a solution for the data centers, even the hyperscalers. Do you see that as a market for you?
Yeah. So, maybe just for your viewers' reference. So we have a 200-kilowatt module that's kind of our base system. And then we also have an 800-kilowatt, 1.6, 2.4, 3.2 MW system. And it can scale in increments of 800 kW. And the reason for that is, and this is very unique to us as well, is that 200-kilowatt four-shaft core that we call it, our KARNO core. That's really our building block that we then just stack those together to get the amount of power you need. So same idea as what like a Tesla has in its floorboard. You've got thousands of these small battery cells all strung together to give you the amount of power you want. We're taking the same concept and applying it to power generation, where you have this 200-kilowatt core and you just stack them to get the power you need.
So for big data centers, I mean, the really big ones, we're talking about like a gigawatt of power or high hundreds of megawatts of power. That would be a ton of our system. So more the discussions we're in on that end is we can be your transient handler. So if you think of AI compute, it's actually extremely spiky in terms of the amount of power it draws. A lot of this is new to data centers, by the way. Like data centers in the past had very consistent power levels. Now it's like the edge of a sawtooth, where it changes drastically, and you need systems to be able to handle that. Well, turbines aren't very good at handling peaks and valleys. And so what we're in discussions about is buy a turbine for your base, 800 and 900 MW of power, and then buy KARNO or a technology like us for your top 100 MW of power, and we'll be able to handle those transients. Or if you go to the other side of like 10-megawatt, 50-megawatt data centers, then you would just be able to buy KARNOS to be able to handle that much. So we're not saying we're like the end all, be all answer. We definitely see other technologies as the right part of the mix, but we do think we hold a pretty unique value proposition to be able to help many of the different use cases.
I think most people read in the paper every day about data centers and their impact on power, positive or negative, their impact on pricing. But I don't think very many people really realize the whole transformation of the electric grid as we learn to interact with it differently, to bring on new cars, we bring on more air conditioners, we bring on more industrial development, and all of that creates a massive opportunity, not only at the high end, as you say, 500 MW to a gigawatt, but on the low end as well, 50 to 300 kW. How would you describe your ideal customer when someone calls you and says, hey, this is what I'm wanting from you. What are the attributes of an ideal customer for you?
Yeah, I would say for the beginning here. And just for viewers' reference, we're at a point where we just started initial customer deliveries last year in '25. We're expanding deliveries in '26. And then '27 is where we're looking to start to scale. And so, it's always good to just level set. So, you know, if a gigawatt data center came to us tomorrow and said, we want to, you know, in '26, '27, do all of it with KARNOS, it's not going to happen. We just don't, we're not there. So what we're trying to do in the beginning here is find customers who aren't solely competing just on cost per kilowatt hour. So as we talked about, the Navy has that pain point of wanting low maintenance, or oil and gas wanting to have more flexibility on the fuels that they can run, or data centers. We also see there's a benefit of being able to run on pipeline natural gas and site emergency diesel fuel as well, in case your pipeline has an issue to just bolster the reliability of the whole system.
So those are the customers that we look at in the beginning innings here, where we're like, this is awesome. This is perfect because yeah, you want to experience our low cost benefit, but that's not the pure reason you're buying. You want this other suite of benefits as well, or the emission side like we're talking about. And then long term, I mean, my vision for the company and the technology is we will just compete on purely cost per kilowatt hour. Right? I mean, we believe that this technology can let you make your own electricity cheaper than you can buy from most of the grid across the US. So if that's the case, why wouldn't you put one of these behind the meter and make your own power, control your own reliability, and control your own kind of destiny on what the cost per kilowatt hour is going to be?
