管理層發言
Greetings. Welcome to Atlas Energy Solutions, Inc. Second Quarter 2026 Earnings Call. Operator Instructions: Please note, this conference is being recorded. I will now turn the conference over to Kyle Turlington, Investor Relations. Thank you. You may begin.
Hello, and welcome to the Atlas Energy Solutions Conference Call and Webcast for the Second Quarter of 2026. With us today are John Turner, President and CEO; Blake McCarthy, CFO; Tim Ondrak, President of Power; and Bud Brigham, Executive Chair. John, Blake and Bud will be sharing their comments on the company's operational and financial performance for the second quarter of 2026, after which we will open the call for Q&A. Before we begin our prepared remarks, I would like to remind everyone that this call will include forward-looking statements as defined under the U.S. securities laws. Such statements are based on the current information and management's expectations as of this statement and are not guarantees of future performance. Forward-looking statements involve certain risks, uncertainties and assumptions that are difficult to predict. As such, our actual outcomes and results could differ materially. You can learn more about these risks in the annual report on Form 10-K filed with the SEC on February 24, 2026, and our quarterly report on Form 10-Q for the first quarter and current reports on Form 8-K and other SEC filings. You should not place undue reliance on forward-looking statements, and we undertake no obligation to update these forward-looking statements. We will also make reference to certain non-GAAP financial measures such as adjusted EBITDA, adjusted free cash flow and other operating metrics and statistics. You will find the GAAP reconciliation comments and calculations in yesterday's press release. With that said, I will turn the call over to John Turner.
Thanks, Kyle. For the second quarter, Atlas generated revenue of $293.2 million and adjusted EBITDA of $49.5 million, which represents an EBITDA margin of approximately 17%. Blake will cover the financial detail later on the call. Before I get into the quarter, let me lay out how our Power business is organized because we get a lot of questions about it. We have two divisions. Oilfield power sells generation to oil and gas operators across many basins, and we expect that fleet to exit this year with 180 megawatts to 200 megawatts deployed, the majority of which are under long-term agreements. Long-term behind-the-meter power sells primary permanent power to large-scale users, principally data centers. We signed our first contract in that division this quarter, and our Global Framework Agreement with Caterpillar supports its growth. The second quarter was highlighted by the execution of our first behind-the-meter contract, a 120-megawatt power purchase agreement with a subsidiary of an investment-grade technology infrastructure provider. The economics are as follows: total project capital is approximately $190 million. We expect the contract to generate approximately $55 million of adjusted free cash flow on an annualized basis once the permanent facility is operating. This is a cash-on-cash payback of less than 3.5 years. These economics are specific to this contract, this counterparty and this site and should not be applied to future projects. Just as important, the capital required to build this facility sits inside the capital guidance we gave you last quarter. We are not raising our previously announced capital budget to fund this growth. The site is in Socorro, Texas, and here's the full sequence. We have completed construction of a 26-megawatt facility that is powering the site today through the customer's construction and testing phase. We began installing Atlas equipment for the permanent plant in the third quarter. Commissioning begins in the fourth quarter. The 120-megawatt facility electrifies at the end of the first quarter of 2027, and we began recognizing revenue on this new system in the second quarter of 2027. That sequence is the whole point. This project demonstrates Atlas' full solution approach to the behind-the-meter market, providing customers with a one-call option to solving their power procurement issues in every phase of a project's life cycle from powering the pivotal early-stage ramp-up to providing power through the life of the facility. The commercial opportunity set for the permanent behind-the-meter power continues to expand rapidly. As demand for compute capacity explodes with the evolution of AI, we have seen the urgency of our commercial negotiations rise. As the frontier models grow in complexity and ability, the need to accelerate access to token generation rises. The priority placed on access to power is best exemplified by the hyperscalers continuing to build out their internal procurement teams where they have added seasoned power professionals that have accelerated and focused contract negotiations. Over the past few months, we have seen the nature of the power deals we are pursuing evolve from both a size and duration perspective. When we entered into our framework agreement with Caterpillar, we assumed it would take eight to ten projects to fully contract the capacity we place on order. We were very confident we could do so, but we expected it would require a significant commercial effort. As our commercial capabilities have become more apparent with prospective customers, we have found the scale of the projects on which we are engaged has grown significantly in magnitude. Thus, it is becoming increasingly more likely that two to four projects could contract our remaining uncommitted capacity compared with our previous assumption of eight to ten projects. Additionally, we have seen an increasing appetite, even need from our prospective customers for longer-tenor contracts. With grid access becoming increasingly difficult to secure on any predictable timeline, prospective customers are embracing island power as a long-term solution for their sites. Initially, we were looking at 10-year contracts as a sweet spot, but we are increasingly seeing a desire for 15- to 20-year terms from our prospective customers as they look to de-risk power supply to their facilities for the long term. With potential customers looking to execute on the prerogative of procuring long-term island power quickly, we are finding that our strategy of providing a full-service solution from early engineering through full life cycle maintenance and operations is gaining traction. Prospective data center customers are looking to us to solve the entirety of the power problem. They don't want to provide specifics on engineering or equipment. They want to provide a load quantum with productive load swings and a reliability metric. They want us to provide a system that achieves their goals in the shortest amount of time and with the partner they can trust to operate for the long term. Lastly, we are not seeing potential deals shifting to the right. There is a sense of urgency to get projects moving forward. We expect the industry to see additional contract activity over the coming weeks and months, and we believe Atlas is well positioned to compete for those opportunities. Outside of the 120 megawatts attached to our recently announced projects, we have an additional 120 megawatts arriving at the end of this year and another 350 megawatts scheduled for delivery over the course of 2027. The availability of this 470 megawatts for deployment in 2027 with a clear line of sight on a steady ramp phase lines up well with the requests we are seeing from our prospective customers, putting Atlas in a strong competitive position. We believe our contracted backlog has the potential to evolve meaningfully in the second half of this year. We will announce new contracts when they are signed. A question we often get is, what are the competitive advantages for Atlas and Power? And why is Atlas the right partner in long-term private power solutions? Atlas has built over $1 billion worth of infrastructure projects, which includes many of the lowest cost mining facilities in West Texas, along with a first-of-its-kind 42-mile conveyor with the Dune Express. We designed, engineered, constructed, powered and now operate these projects. Large complex construction projects are in our DNA. That significant expertise is being called on in the markets today as our power customers are facing new challenges that demand precise execution. As customers like data center operators increasingly look to private power, they demand confidence that their power provider has the depth and experience to deliver on time and on budget. As our power business grows, you will see how our engineering and execution expertise transforms the way companies attain the critical power they need. The equipment you choose and the construction partners you pick will matter. Innovation is core to the Atlas culture. We're the only company to dredge mine in the Permian, the first to deliver frac sand autonomously in the Permian, the only company offering multi-trailer sand deliveries, and we built the longest sand conveyor system in North America to change the way sand is delivered in the Northern Delaware Basin. We plan to bring that same outside-of-the-box mentality to the private power market. In our oilfield power assets, we continue to see strong contracting momentum. This quarter, we signed additional contracts and agreements for current and future megawatt placements and now expect total oilfield megawatts deployed to exit the year between 180 and 200 megawatts, the majority of which will be under long-term agreements. We are being very deliberate about the placement of our current oilfield fleet, prioritizing longer-tenor agreements over near-term deployments. We anticipate continued desire from our customer base to sign agreements to secure their power needs as traditional utility power timelines continue to slip. We are also seeing growing interest in microgrid systems utilizing Atlas' battery technology. With this technology, we can hybridize customer locations to increase reliability. Our hybrid technology combines our affordable generators, a robust battery system and a control software that manages site loads and power availability, which leads to a more robust service for our customers and our operating efficiencies for Atlas. Demand for private power is robust, and we believe we are still in the early stages of a major infrastructure growth cycle. We have the right equipment, the right strategy and most importantly, the right people. I will turn the call over to our CFO, Blake McCarthy, to expand on what we are seeing in our Sand and Logistics segments.
Thanks, John. If I were to pick a word to describe the current West Texas sand and logistics market, it would be nuanced. Macro conditions have improved significantly over the first half of the year. And while geopolitical events continue to drive volatility, we believe that the supply-demand balance for crude oil has been dramatically altered and the fundamental floor for oil prices has been lifted. We are obviously not the only ones to share this view as the Permian rig count has grown 17% since the start of the Iran conflict. This growth has been led by the private operators who are the most apt to respond to changes in the commodity price backdrop. Assuming our view is correct, we would not be surprised to see the public E&Ps ramp activity next year once they have refreshed their capital budgets. Despite the growth in rig activity, completion activity has nearly remained static, which is what drives demand for our business. This is due to a number of factors. For one, operator DUC inventories were already thin in early 2026, with oil placement activity highly aligned with completion schedules. As pad sizes have grown and even with today's efficient drilling operations, it takes extended time and planning for new pads to be constructed and wellbores to be placed to enable simul-frac operations. In addition, gas takeaway capacity in the Permian remains an issue. While this will be partially alleviated as more than 4.5 Bcf of incremental pipeline capacity comes on in the back half of this year, it has certainly put a cap on the activity of a few operators year-to-date, particularly in the Delaware Basin. As we mentioned on our last call, we don't expect frac fleet additions beyond the marginal view we saw in April until later this year as operators attempt to gain comfort with the strip amidst the volatility. However, based on recent customer conversations, we expect activity levels to begin ramping moderately in Q4 in a calendar seasonal way as customers look to hit 2027 running. Pressure pumpers are displaying discipline in not bringing incremental equipment to market until pricing improves on their currently utilized fleets. And with the lack of readily available equipment and crews on the sidelines, there's a significant lag of approximately two months between when a customer can contract the fleet and when completion operations actually begin. Touching on the supply-demand balance of the actual sand market, it is our view that the market is much tighter than current pricing and market sentiment would suggest. Sand is the ultimate commodity and then a little bit of oversupply quickly leads to the price falling to the marginal cost of production for the industry. And inversely, just a little undersupply can lead to a rapid spike in prices. While nameplate capacity would suggest that there's still quite a way to go before the industry comes into balance, we believe these figures could radically overstate the true productive capacity of the industry. Over the past three years, maintenance CapEx has been an afterthought to a broad swath of the market. And based on recent spot sales to customers experiencing nonproductive time due to waiting on sand, it appears incremental production in the Permian is still limited. This trend is likely to become more apparent as the broader industry moves closer to full utilization. We're beginning to hear more anecdotes of competitor facilities struggling operationally as they attempt to ramp production and it's causing us to reconsider our earlier math that it's going to take four to six net completion crew additions for the market to reach tight conditions. We believe the market is rapidly approaching a period of true capacity discovery. We think it's time to force the issue. Nonproductive time or NPT is likely to become a hot-button issue for the industry before that point is reached, driven not by sand supply, but by truck availability. Trucking rates have stabilized at much higher levels and with continued price increases in the national over-the-road freight market, driver shortages in the Permian have become more acute. To add on top of the higher hauling rates, the spike in diesel prices has dramatically increased the overall cost of hauling sand. While Atlas is partially insulated from some of this inflation due to the advantages of the Dune Express and our use of autonomous trucks, we are beginning to see our competitors who have been loath to raise logistics pricing negatively impacted operationally from these developments. In June alone, we took over two wellsite jobs mid-completion as competitors simply could not secure drivers at the rates they were offering. We expect this trend to become more common in the back half of the year. And with oil prices where they are, delays in monetizing resources in the ground become significantly more punitive to operators. To be blunt, we expect operators will need to pay higher rates to avoid NPT related to both sand and trucking beginning in Q3 and accelerating in Q4, which we expect to benefit Atlas. This commercial strategy is intended to reinforce the value of execution reliability. Atlas provides a superior level of execution reliability to our clientele, enabled by the investments we have made in our plants, our logistics infrastructure and most importantly, our people. However, at times, we can become victims of our own success. When we do our jobs well enough, customers can begin to take that level of service for granted. For more than a year, we've been willing to price our services at levels where our customer base can enjoy the operational efficiencies of the Atlas network and realize price savings, a strategy that has resulted in us gaining market share. We believe we have done what was needed to rationalize the market. It is our belief that many of our competitors' minds have been severely impaired operationally and the market needs a period of true capacity discovery. Thus, at this point, we are choosing to hold the line on pricing on certain tenders in the market. Some customers may prioritize the lowest cost option on paper, which, in our opinion, will highlight the difference between the service providers who can deliver and those who simply cannot. We expect this will test both the industry's true productive capacity and operators' tolerance for nonproductive time. Second quarter sand volumes were approximately 5.6 million tons, which was below our expectations as rig moves and completion schedule changes negatively impacted volumes in late June. July volumes recovered nicely to approximately 2 million tons. Full third quarter volume expectations remain a bit up in the air due to our aforementioned commercial strategy as well as some scheduled breaks and customer completion schedules. The current expectations range from approximately 5.3 million to 6 million tons, which is admittedly a wide range. However, we believe this near-term uncertainty is necessary to properly set the stage for the more important contracting season at year-end. It's worth noting that our completion schedule for Q4 is already positioned for a strong close to the year, representing the highest volume quarter of the year on an already allocated tons basis as some key customers are positioning themselves to close the year with gathering momentum. Our last mile team set a quarterly record for shipments at 6 million tons. During the second quarter, we made more than 4,600 autonomous deliveries, up 70% from the first quarter. Our partnership with Kodiak has begun to result in significant productivity gains, which we expect to accelerate as new operational milestones are reached that will expand the operational footprint trucks are able to reach. We are targeting operations on public roads by the middle of next year, subject to regulatory and operational milestones. Additionally, we set quarterly volume records down the Dune Express. Moving to our financials. Second quarter 2026 revenue was approximately $293.2 million. Total proppant sales volume was flat sequentially at 5.6 million tons. Our average sales price for proppant for the second quarter was approximately $17.70 per ton. Second quarter cost of sales, excluding DD&A, were $221.3 million, consisting of $66.1 million in proppant plant and logistics equipment operating costs, $1.4 million from power equipment costs, $140.7 million of service costs, $8.8 million in rental costs and $4.3 million in royalties. For the second quarter, our per-ton proppant plant operating costs were approximately $12.39, including royalties, down from the first quarter. OpEx per ton for the third quarter is expected to be flat to down, depending on total volumes as our plant operational efficiency initiatives continue to bear fruit. Our logistics business posted strong sequential improvement in the second quarter on the back of record volumes, an improving rate environment and strong utilization of Dune Express. Q2 logistics margins were 14%. For the third quarter, margins are expected to stay solidly in the double digits. Our power business also reported strong sequential growth with improved utilization in our oilfield power fleet and the start-up of operations at our new facility in Socorro, Texas. Q3 contribution from this business is expected to display continued improvement as we deploy larger portions of that fleet under long-term agreements. Q2 adjusted cash SG&A, excluding extraordinary litigation expenses and other nonrecurring items, was $24.3 million. SG&A is expected to average approximately $22 million to $24 million for the third quarter, excluding legal fees from litigation and contracting activities. Growth CapEx for the quarter was approximately $131.5 million, the majority of which was tied to our initial Caterpillar power generation equipment order. Maintenance CapEx was $14.6 million. CapEx for the second half of the year is budgeted to be approximately $200 million, which keeps our full year capital spending inside our full year 2026 guidance range of $350 million to $375 million. The vast majority of that, approximately $175 million to $190 million is attached to the build-out of our private grid power business. It's worth noting that approximately $110 million of second half growth CapEx is connected to the build-out of our already contracted facility in Socorro that will begin generating meaningful cash flow in Q2 of '27. As a reminder, we expect this to generate approximately $55 million of adjusted free cash flow per annum. The remainder relates to purchase obligations under our Caterpillar Global Framework Agreement, which we announced in March. This is not new spending. It is the fulfillment of an order already on the books. Maintenance spending for our legacy business is expected to take a step down as we have completed the majority of our larger initiatives at plants. Maintenance capital spending for our sand and logistics business is expected to average approximately $5 million to $7.5 million per quarter in the second half of the year, supporting the free cash flow generation ability of that business. On the heels of our successful convertible issuance in April, the combination of Atlas' available liquidity and the positive free cash flow from our sand and logistics business is more than enough to satisfy our upcoming capital needs. Looking ahead to the third quarter, overall sand and logistics sales volume remain the biggest barrier, while we expect continued improvement in our production costs in power. The combination of planned customer breaks and exercising more discipline on outstanding tenders is expected to result in a temporary step back in overall volumes in order to drive longer-term price improvement. For Q3, we currently expect EBITDA in the range of $30 million to $45 million. As mentioned earlier, we expect the fourth quarter to show meaningful sequential improvement based on already allocated volumes and customer completion schedules that have been communicated to us with current expectations, matching or exceeding Q2 results. I will now hand the call back to John.
Thanks, Blake. I want to reiterate Blake's comments about the shift in our commercial strategy. This is an intentional strategic decision. We are holding price on certain sand tenders rather than chasing volume, and we are willing to trade near-term volumes to do it. We believe this will drive the market to realize the rationalization in productive capacity and logistics capability that has transpired across the West Texas sand industry and serve as a catalyst for a pricing recovery. I will now hand the call off to our Executive Chairman, Bud Brigham, for some closing remarks before we turn the call over to Q&A.
Thank you, John. In late July and early August across most of the country, everyone gets excited about the upcoming football season. At this point in the year, every football team is still undefeated. I call it the talking season. But in the case of power, it seems the market hasn't appreciated the fact that Atlas has moved beyond the talking season. We've already begun putting points up on the scoreboard. It reminds me a bit of the cynicism surrounding our first company, Brigham Exploration, when we were pioneering horizontal drilling and multistage fracking in the Bakken nearly 20 years ago. Most people did not believe horizontal fracking would work in oil. Over the next five years, we not only proved them wrong, we led the way, delivering superior production and economic performance. About five years later, we did it again with Brigham Resources in the Permian. Eight years ago, we faced the same skepticism about the viability of local sand when we started Atlas. We went on to build the largest state-of-the-art frac sand plants in the country and meaningfully improved economics for Permian operators. Then again, just three years ago, many said we couldn't build North America's largest conveyor system to move proppant 42 miles into the heart of the Delaware Basin. Of course, we did. We love these challenges. We're very, very good at them. Nobody builds large-scale energy infrastructure as successfully as Atlas. And here we are again this time with an extraordinary opportunity in private power. Even with our first contract, the skeptics are once more out in force. That's fine. We've been here before. I have complete confidence in our team, our strategy and the partners we've chosen. We look forward and are excited to changing the narrative. Thank you for joining us today. I'll now turn the call over to the operator for Q&A.
分析師問答
Operator Instructions: Our first question is from Jim Rollyson with Raymond James.
John, your commentary around the pace of data center deals was interesting and maybe stands out a little bit from at least the color we've heard so far this quarter. Can you just expand about what you're seeing in terms of project size, scope, timing, all those kinds of things? I thought it was interesting that placing that capacity with two or three or four customers is a departure, but I'd love to just get a little more color there and what we can expect coming through the second half.
Yes. Sure, Jim. Thanks for the question. Obviously, we're not seeing every deal that's out there, but what we are seeing is that there's an intense urgency from our potential customers to get contracts signed and things are moving on. I guess things are moving in order to get those projects derisked and those timelines derisked. The counterparties we're negotiating with are looking to move as quickly as they can, and that makes—obviously, that's in months and not years. Obviously, these deals don't happen overnight. They're billion-dollar deals that take very long—very long durations, it takes a long time to negotiate. And they are also running these deals in parallel with these data center lease negotiations, which are pretty complicated as well, if not more complicated. So when you get all these contracts together and once we get our contract negotiated, there's multiple things that have to happen before a contract could be signed. So there's a lot of moving parts there. But what we have been seeing is we have been seeing an urgency to get deals signed. We've been seeing the size of these deals increase. We've been seeing the tenor of these deals increase from—when I'm talking about what we were seeing, say, six months ago. So obviously, very positive from our standpoint. We haven't really announced any—we won't be announcing any deals until we have a contract signed, but we are working on those.
Got it. Appreciate that. And as a follow-up, maybe for Blake, can you expand a little bit, Blake, on the volume guidance, like pretty wide range, obviously, for 3Q? And I guess just trying to understand how much of that is the customer breaks versus kind of you electing to maybe hold the line on volumes to get better pricing going into next year? And kind of tied to that is how comfortable you are with the 4Q implied ramp?
Yes. Yes. That's a great question. I was expecting that one. I think it's a completely fair characterization that it's quite a bit of variability there. So yes, as you pointed out, there's a number of moving pieces in Q3. So first, we do have some key customers that are taking short crew breaks during the quarter as they prepare to ramp up in Q4. That probably represents, say, 50% of the variance. And so that includes some customers that are transitioning pumping providers as they secure new equipment and they move to bigger frac designs. With the ramp in activity for some of these key customers, we are on course for a very strong Q4, and that's without incremental volume wins. However, the other impact is going to come from our shift in our commercial strategy. And that's the one where there is a bit of question mark, and that's why you have that wide range. So for more than a year, we've been following the playbook, which is, say, you're the low-cost provider of commodity and service. So when the market is oversupplied, yes, you've got to price it at the marginal price of production for the industry or slightly below in order to gain market share and force the market to rationalize. We've done that, and we've caused a lot of pain in the market. Most of our competition, they've laid off crews. They've cut maintenance spending to zero to the point where some of them even have padlocks on front gates. So however, just because the mine is kind of limping along, doesn't mean that it's theoretical sand that isn't being bid in the projects. So all of this like theoretical nameplate capacity, what I call zombie mines, is being bid into customer tenders for sand. And it's creating this perception that there's still an ample oversupply of sand in the market. And we just simply don't think that's the case. But we think that doesn't really matter until our customer base thinks it too. So as long as Atlas is willing to match the competing bids in the market, customers can continue to hammer on price while still enjoying access to our service and our execution reliability. And so it's not until we've drawn a line in the sand and let them go test the waters elsewhere that you shine a light on the market and what's reality. So we have to create a catalyst for that light to get shone on what these mines can actually produce, what competing haulers really have to charge to deliver the sand and who can really orchestrate all the different moving pieces to run these to get sand on site because we make it an afterthought, and we've made it easy. It's a little bit of a victim of our own success. When you do your job well, sometimes it gets taken for granted. And so our expectations are that NPT, which we have tried to make a thing of the past in West Texas, is about to become a pretty big issue for some operators. So at the end of the day, this is going to allow us to obtain more value for our products and services. We think that it's going to set us up very well for RFP season for 2027 and the execution of the strategy and it's going to give us a hammer when it comes to negotiations.
Our next question is from Stephen Gengaro with Stifel.
So can I start with sort of the CapEx question? I mean we hear a lot from companies this quarter as far as kind of CapEx per megawatt deployed in the power business. And I think we're hearing numbers like around $1 million for the generating equipment per megawatt and maybe like $1.6 to $1.7 for sort of all-in balance of plant. What are you guys seeing? And are you seeing kind of inflation in those numbers?
Thanks, Stephen. I'm going to let Tim jump in with all the details because he's the guy at the coal face, but just to lead, like this is something that we've been pretty vocal about for some time and that we—it's really customer and project dependent. So, hey, like answer for me these questions: what's the load profile the customer requires for their objectives? What type of system resiliency and reliability metrics do they insist on? Those all have knock-on impacts to the overall cost of the system and therefore the price, which is why we've always said from the beginning that the best lens through which to review these projects is unlevered project IRR because the variables on the front end are apt to change, and thus the cash flow stream has to change, too. Tim, do you want to jump into the deeps?
Yes. I think Blake gave some good color and to answer that question. The CapEx ranges we're seeing on projects can be anywhere from $1.5 million a megawatt to $2.5 million a megawatt. And again, those are really informed by what is the system intended to do versus cost inflation. It's really scope inflation. And I think we're seeing hyperscaler teams getting a little more in the weeds on what is engineering asking for versus what is procurement willing to put forward. And there's a lot of difference in a system that's designed to five nines versus three nines. And so we're seeing a little bit more thought go into what actually works, what's deliverable and what's cost efficient in that case. So as they build out those teams, we're getting better answers from them on what they're willing to live with from reliability, availability and what they need on load steps.
Got it. Okay. That makes sense. When we think about—and you talked a little bit about sort of your balance sheet liquidity and kind of how you fund the growth. Just remind us your planned deployments of power over the next couple of years and how you think about paying for that? And obviously, given the dynamics you just mentioned, it's going to vary a little bit by which projects are signed. But how do we think about that?
Yes, yes. So during Q2, we did make some large payments for the initial order of Caterpillar generators we're receiving this year. So as I've said in the prepared remarks, we still have another $200 million of CapEx planned for the back half of the year and 90-plus percent of that is going to the rest of the Caterpillar deliveries, including down payments on our 2027 orders and ancillary equipment for our currently under-construction deployment in Socorro and some other longer lead time items for other projects. So thinking about the liquidity following the convertible raise in April and the Q2 Caterpillar payments, we currently have approximately $168 million of cash on the balance sheet and approximately $125 million of undrawn capacity on our ABL. Additionally, I think it's key to reiterate that the CapEx for our sand and logistics will now truly reflect the low capital intensity nature of that business. So we're effectively done with the major CapEx projects we have planned for that business this year. So CapEx for that business steps down to that $5 million to $7.5 million range for that business moving forward. Additionally, the CapEx cycle for oilfield power business has also matured. So both of those businesses are going to start spitting off cash. So we're more than good when it comes to our near-term obligations. That's not to say we won't need incremental capital for the projects we are currently negotiating. And we have already, in fact, made significant equity investments into those prospective projects. So funding for those projects is most likely to come in the form of debt financing, which we won't be putting on the balance sheet until we have hard contracts with great counterparties in hand.
Our next question is from Doug Becker with Capital One.
It seems like you're having some good success in the oilfield power side of the business. Just curious if there's any consideration to deploy some more capacity into that, presumably higher shorter-term returns. But the power contracts, the longer-term data center-related contracts can take a while to finalize. Just wanted to get your thoughts on that balancing data center versus maybe some shorter-term oilfield work.
Yes. I think we'll continue to deploy assets into the oilfield power space. We've become a lot more selective about that over the last six months. We want to deploy those with some tenor. We want to deploy them where we've got some density. And that's how we pick up efficiencies and continue to operate that business. But it's a great business. The levers to scale that business up are much shorter than the levers to scale a business that's supporting industrial power data centers. And so to answer your question, it really continues to be opportunity driven: where we've got good relationships with good customers that have an outlook for meeting those assets for a period of time that we like, we'll continue to deploy into that space.
That sounds good. And maybe switching gears to the logistics business. Margins in March were kind of the mid-teens, finished below 13% for the full second quarter. Everything seems to be lining up to really favor the Dune Express. So kind of curious why we're not seeing margins maybe improve more than just kind of solidly into double digit going forward this year.
Yes. For the Q2 moving pieces, you did have a lag in terms of as I talked about: third-party carrier rates continue to march up, the over-the-road national freight market continues to strengthen and that starts to pull, suck drivers out of West Texas. And so your cost—there's a lag in your cost going up on the third-party carrier rates and as you start to amend your own hauling rates. So that actually is a tailwind on pricing. As you look ahead to the second half, it's more a knock-on effect of that—again, there's kind of some flex in that guidance, and that's related more to the volume guidance on the sand side, where obviously, there's a fixed cost absorption piece of that. So it's—again, it's loose guidance based upon more like, 'Hey, we're drawing this line in the sand.' We're starting to move rates. And we think that there might be a quarter of kind of an initial reaction and then finishing off with a strong Q4.
But we are seeing tightening in the trucking market, absolutely.
Our next question is from Scott Gruber with Citigroup.
I appreciate the pricing discipline here as demand improves. But I'm trying to get a sense of what this could mean for your average pricing. There was something like a $4 a ton spread between some of the contracts you had coming into the year and more recent sales. So just thinking through, like if the market comes to meet you at your line, does your average pricing as you head into '27 kind of stay flat around the $18.50 you posted in 2Q? Is it trending higher? Just trying to get a sense of where the realized price could go given the dynamics you have in the book today.
So thinking about '27 pricing, obviously, we're not going to guide that yet, but what we're trying to do—as you've touched on—we are positioning ourselves to strengthen our position come RFP season where, as I said, you shine a light on the true productive capacity of the market. We do have a significant portion of the book turning over for '27. So in the event that we are able to achieve the pricing move, the increases that we're looking for, that would result in accretive pricing to the average price of sand.
Got you. So the simple way to put it, like—I'm not trying to pin you down on exact numbers around your strategy. But your strategy, if successful, if the market comes to meet you, that would be to move higher in your realized pricing heading into next year? Is that fair?
Yes.
Yes, yes.
Okay. Okay. Just wanted to clarify that. I appreciate it. And then the progress with Kodiak on autonomous trucking is good to see. Maybe you could just provide some more color. You mentioned 100 trucks on the road in a year or so. Kind of when does this start impacting the financials? What do we see in terms of your cost base with 100 trucks running? And then as you scale it up, ultimately, what could this mean for your financials?
I guess, really, the first thing that we need to do to start seeing the benefits from that is we really need to start expanding the horizon for which these autonomous trucks serve. They're serving in a couple of heat zones right now and activity moves in and out of those heat zones. But in order to maximize the benefit on autonomous delivery, we're going to need to maximize the number of well sites we can serve with that. So number one, the first thing we needed to do is we needed to be able to get those trucks into other heat zones.
Broaden the area.
Just touching on Doug's earlier question, right, there was that lag effect in the cost of drivers going up and then the actual accretion into your hauling rates. And so you basically eliminate that variability with autonomous trucks, where as the trucking market tightens, you keep that cost more stable and it becomes accretive. It also de-risks our logistics operations as the number of crews expands; one of the bigger risks is, 'Can you get enough third-party drivers? Can you get enough trucks?' And that becomes less of a question mark when you've got more of the autonomous trucks on the road.
With all the initiatives coming around with commercial driver's licenses and foreign CDLs, that's going to make a big impact on the trucking market going forward.
Our next question is from Michael Scialla with Stephens Inc.
Blake, you said you wouldn't bring debt on the balance sheet unless you have contracts to underpin the new power requirements. I guess I want to see how you're thinking about absolute debt levels or net debt or leverage, however you want to look at it? What kind of targets are you looking to stay below as you build out the power business?
Yes. I think on the front end of this build-out, the leverage ratios do start to blow out before those first key projects start spitting off cash. I think that's pretty common across the space. Once you get these projects online, though, they significantly start to delever themselves very quickly. You're looking at cash-on-cash payback on these projects in the five- to six-year range. And so it's—you're probably looking at standard project financing like equity to debt ratios of anywhere from 30% to 40% equity, 60% to 70% debt. So when you think about those cash-on-cash paybacks, you quickly—you get three times. And that's—the oilfield services guy in me is like, 'Well, that's too much debt.' But when you have these contracts that are 15, 20 years with hard concrete terms, that's actually a very strong leverage position.
Understood. A big trade-off there. I guess I also wanted to see on your third quarter guidance, EBITDA guidance, you obviously have a range in there. And I assume most of that range is because of the range you talked about on sand production, having a pretty wide range with that. Can you break down the EBITDA, how much you're anticipating from sand versus logistics versus power?
I mean, if you think about it, 100% of that variance is going to be coming from the sand and logistics side. So power EBITDA will be up slightly as we get a full quarter of deployment of that temporary small facility that we put in place for the construction phase of the Socorro project, and then continued gains on the oilfield power side. The sand and logistics are very closely correlated, and so again, on that variance, about, say, 40% of that is due to some customers having communicated to us like, 'Hey, we're going to be taking a short break'—there's 30- to 45-day breaks in the quarter, in front of their ramps into Q4 and into 2027. The rest of that delta is coming from the unknown response that we're planning for in terms of the shift in commercial strategy. And that really sums that up.
Our next question is from Chuck DeVore, private investor.
I know that we don't often like to consider things like public policy and politics, but I noticed that Texas Governor Greg Abbott announced a pause for data center construction insofar as data centers that connect to the grid. There's a lot of resentment in rural Texas. We got a midterm coming up. The other thing that happened this week was that the Lieutenant Governor and the key Chairman of a Senate Committee called on ERCOT and the PUC to delay the construction of the 765-kilovolt lines heading into the Permian. This would seem to demand additional thermal generation in the Permian if you're not going to build those lines. Are things like these two public policy changes likely to increase the demand for your power services in the Permian?
Yes. Thanks, Chuck. Obviously, these are pretty new as far as policies go. We don't agree with them, but yes, this is a potential tailwind for Atlas and any company that has additional power or has power assets available that can be immediately deployed over the next couple of years. I think this is going to be a huge tailwind for—look, I mean, we are already seeing companies—I think a lot of the hyperscalers have already come to grips with this and are already starting to provide or look for their own behind-the-meter power solutions. And I think we're just going to continue to see that. I think anything like this is also going to push a number of companies out into West Texas that are going to start locating data centers and their projects out there. But yes, this is a very big tailwind, and we've been seeing this. I think you're starting to see the urgency from a number of customers and users that are looking for power, and we really turn into those companies that have access to that power.
Chuck, just to follow up on John, we've really seen this coming for about 12 months now. We've touched on it on a few prior calls—the political headwinds with respect to the grid. Our prospective customer base totally gets it, where they're no longer thinking about the grid as the default solution. I think that's why you've seen the increase in the tenor of the contracts: if you rewound 18 to 24 months ago, people were talking about five- to seven-year contracts. Now, we're talking 15 to 20 because they're just not going to even think about the grid. When you look at the backlog and the capital requirements that have to post up on the front end of that with still question marks around it, it's just easier for them to say, 'We're going to take care of this ourselves.' On top of that, they get higher reliability and significantly better resiliency with a system that's designed for their uses. When you think about that, you're not getting three or four nines on the grid; you're lucky if you get 98%. So the combination of those factors makes behind-the-meter solutions much more compelling and permanent.
Yes. And that's precisely the reason why we signed the Global Framework Agreement with Caterpillar. We realized that this power demand behind the meter was coming and the winners are going to be the ones that have the ability to deploy.
Our next question is from Alexa Breno with Goldman Sachs.
We wanted to ask, as we think about the commercial pipeline of opportunities on the power side, can you just talk about what that split could look like in terms of the oil and gas and then some of those larger projects?
Yes. So the majority of our pipeline is going to be larger projects. And in oil and gas, we know that universe. We've got roughly 40 megawatts that we could deploy into that space. And those are megawatts that are on the ground or in process. In the larger power space, our opportunity set is roughly 8 to 10 gigawatts. That was the number on our last quarterly call. There have been projects that have moved in and out. Keep in mind, that's pre-Governor Abbott's announcement yesterday. But those projects have grown in scale from the discussions that we had six to nine months ago. We're seeing the nodes that customers are coming to us to solve in that kind of 500-megawatt and up range. And I think in a lot of those conversations, that's a start. That's kind of what they need to get off the ground and some of those campuses have designs on one, two or even three gigawatts of behind-the-meter power for all the reasons that John and Blake just talked about.
Okay. That's very helpful. And then maybe just a follow-up. Can you talk a little bit more about this kind of pricing piece that you're talking about on the logistics side? What type of margins are you targeting there? You kind of talked about a commercial strategy and part of the reason the 3Q guide might be a little lower than we anticipated. So how do we think about what you're targeting and what 4Q could potentially look like, assuming those numbers come to fruition?
Yes. It's more about a bigger picture than just targeting a specific margin profile. If you look at the nameplate capacity for the Permian and add up all the mines out there based on what they were built to produce, it creates the perception of ample sand. But even at our own facilities where we've been spending significant maintenance CapEx, there is a delta between nameplate capacity and effective productive capacity. And our facilities are among the best maintained in the Permian. Many other facilities have been running at low utilization for the last two years and are held together with minimal maintenance. That nameplate capacity becomes theoretical sand and still gets bid into tenders and used as a hammer on pricing. We believe only a few grains of undersupply are needed for pricing to move materially. When incremental pricing turns, it moves quickly and has a big impact on the income statement. On the logistics side, it's all tied together. Traditional trucking is a very tight market. We've been insulated by the Dune Express, which reduces our reliance on third-party trucks. If someone is moving everything by trucks, they require many more trucks and face larger operational challenges, which will result in more inflation on truck rates. If operators aren't willing to pay more, they won't get trucks. This is a grand experiment, but we think it's going to result in strong findings and should support improved pricing and margins as the market discovers true productive capacity.
Thank you. This will conclude our question-and-answer session. I would like to turn the floor back over to management for closing comments.
Yes. Thank you, everybody, for attending the call or sitting on the call. I'll just close by saying we like where we are, where we sit today with Atlas. We have two good businesses and real advantages in both and a team that knows how to execute. There's work in front of us, and we're clear-eyed about it, and we're going to keep our heads down, do the work and let the results speak for themselves. We look forward to reporting our third quarter numbers. Thank you.
Thank you. This will conclude today's conference. You may disconnect your lines at this time, and thank you for your participation.