Cuộc họp công bố kết quả kinh doanh quý 3 năm tài chính 2026 của FuelCell Energy (FCEL): Giá trị đơn hàng tồn đọng đạt 3,6 tỷ USD

Điểm tin chính

• FuelCell Energy (NASDAQ: FCEL) đã báo cáo doanh thu quý 3 năm tài chính 2026 đạt 33,0 triệu USD, giảm 29% so với cùng kỳ năm ngoái, do doanh thu từ sản phẩm giảm sau khi hoàn thành chương trình nâng cấp công suất Gyeonggi Green Energy.
• Lỗ ròng giảm xuống còn 45,3 triệu USD, tương đương 0,64 USD trên mỗi cổ phiếu, so với 91,9 triệu USD, hay 3,78 USD trên mỗi cổ phiếu ở cùng kỳ năm trước. EBITDA điều chỉnh chuyển biến xấu hơn, ghi nhận âm 36,7 triệu USD so với mức âm 16,4 triệu USD.
• Lỗ gộp tăng lên 24,5 triệu USD, bao gồm 17 triệu USD chi phí liên quan đến giai đoạn ban đầu 30 megawatt của Fit Energy do chi phí sản phẩm hiện tại và chi phí sản xuất chung vượt quá mức giá trong hợp đồng.
• Khối lượng đơn hàng tồn đọng về công suất đã cam kết và được trao đạt tổng cộng 3,6 tỷ USD tính đến ngày 31 tháng 7 năm 2026, bao gồm 1,3 tỷ USD khối lượng đơn hàng tồn đọng đã cam kết và 2,4 tỷ USD khối lượng đơn hàng tồn đọng về công suất được trao. Ban quản lý cảnh báo rằng công suất được trao không phải là đơn hàng tồn đọng chắc chắn hay doanh thu được đảm bảo.
• Công ty kết thúc quý với 737,3 triệu USD tiền, các khoản tương đương tiền và tiền bị hạn chế sử dụng. Ban quản lý cho biết kế hoạch mở rộng công suất sản xuất quy ra năm tại Torrington lên 500 megawatt vào tháng 6 năm 2028 đã được tài trợ đầy đủ nguồn vốn.
• Ban quản lý đặt mục tiêu đạt EBITDA điều chỉnh dương vào quý 4 năm tài chính 2027, tùy thuộc vào việc chuyển đổi đơn hàng tồn đọng, tiến độ giao hàng cho khách hàng, sản lượng sản xuất cao hơn và giảm chi phí sản xuất.

Kết quả tài chính cốt lõi

FuelCell Energy nắm giữ 658,1 triệu USD tiền và các khoản tương đương tiền không bị hạn chế sử dụng vào cuối quý, cùng với 79,2 triệu USD tiền bị hạn chế sử dụng. Tổng nợ và nghĩa vụ tài chính là 153,6 triệu USD, chủ yếu liên quan đến tài trợ dự án, các hạn mức của Ngân hàng Xuất nhập khẩu và nghĩa vụ bán và thuê lại.

Công ty đã huy động được khoảng 298 triệu USD tiền thu ròng từ việc bán cổ phiếu phổ thông trong quý, bao gồm 245,5 triệu USD từ đợt phát hành có bảo lãnh cam kết vào tháng 7 và 52,9 triệu USD thông qua thỏa thuận bán cổ phiếu trên thị trường tự do.

Kết quả kinh doanh và hoạt động

Điểm phát triển thương mại trọng tâm là thỏa thuận mua sắm thiết bị vốn của FuelCell Energy với Fit Energy. Thỏa thuận này có quy mô lên tới 380 megawatt qua bốn giai đoạn. Giai đoạn 0 đã cam kết có công suất 30 megawatt, được hỗ trợ bởi khoản tiền đặt cọc trước, dự kiến bắt đầu giao hàng trong quý 4 năm tài chính 2026 và hoàn tất trong năm tài chính 2027.

350 megawatt còn lại thuộc các Giai đoạn 1, 2 và 3 được tính vào khối lượng đơn hàng tồn đọng về công suất được trao. Fit Energy có toàn quyền quyết định tiến hành từng giai đoạn và các giai đoạn không nhất thiết phải triển khai theo thứ tự nối tiếp.

Tính từ đầu năm tài chính 2026 đến nay, các đề xuất đang hiệu lực đã đạt khoảng 10 gigawatt. Các cơ hội liên quan đến trung tâm dữ liệu chiếm khoảng 97% danh mục dự án tiềm năng trong quý 3 năm tài chính, theo ban quản lý.

Sau khi kết thúc quý, FuelCell Energy đã ký thỏa thuận giữ chỗ công suất 75 megawatt với một đơn vị vận hành trung tâm dữ liệu dùng chung (co-location) lớn cho một dự án ở Texas. Các bên đang nỗ lực đi đến các thỏa thuận chính thức, và ban quản lý cho biết các cơ hội hợp tác tiếp theo có thể xuất hiện với cùng khách hàng này.

Hoạt động sản xuất đạt tốc độ quy ra năm khoảng 37 megawatt trong quý. FuelCell Energy đã bổ sung thêm một ca làm việc tại nhà máy, đồng thời đang tuyển dụng nhân sự và mở rộng chuỗi cung ứng để đạt tốc độ quy ra năm 100 megawatt.

Công ty cũng đã hoàn thành việc nâng cấp công suất cho toàn bộ 42 mô-đun tại dự án Gyeonggi Green Energy ở Hàn Quốc. Ban quản lý cho biết dự án này tạo tiền đề cho tiềm năng thu về doanh thu dịch vụ và thay thế thường xuyên từ hệ thống thiết bị đã lắp đặt.

Trong lĩnh vực thu giữ carbon, FuelCell Energy đã giao và lắp đặt hai mô-đun pin nhiên liệu cacbonat đầu tiên tại phức hợp Rotterdam của ExxonMobil. Dự án thử nghiệm ở quy mô công nghiệp này được thiết kế để thu giữ hơn 90% lượng khí thải carbon tại nguồn, đồng thời tạo ra điện năng, nhiệt năng và hydro, theo ban quản lý.

FuelCell Energy cũng đã ký biên bản ghi nhớ với Siemens để phát triển hệ thống điện cân bằng nhà máy (balance-of-plant) cho các dự án vượt quá 100 megawatt. Giải pháp dự kiến sẽ kết hợp pin nhiên liệu với hệ thống lưu trữ bằng pin, bộ điều khiển lưới điện nhỏ (microgrid) và thiết bị trung thế.

Dự báo của ban quản lý

Ban quản lý kỳ vọng FuelCell Energy sẽ đạt tốc độ sản xuất quy ra năm 100 megawatt vào tháng 10 năm 2026, tạo đà cho công ty giao hàng ở mức đó trong suốt năm tài chính 2027.

Công ty có kế hoạch mở rộng công suất sản xuất quy ra năm tại cơ sở Torrington, Connecticut lên 500 megawatt vào tháng 6 năm 2028. Nhu cầu vốn ước tính vẫn duy trì trong khoảng 200 triệu USD đến 275 triệu USD. Dự báo chi phí vốn cho năm tài chính 2026 đã được giảm xuống còn 10 triệu - 20 triệu USD từ mức 20 triệu - 30 triệu USD do thời gian giao thiết bị, chứ không phải do thay đổi lịch trình mở rộng.

Ban quản lý đang hướng tới mục tiêu đạt EBITDA điều chỉnh dương trong quý 4 năm tài chính 2027. Việc đạt được mục tiêu đó phụ thuộc vào việc sản xuất ít nhất 100 megawatt mỗi năm, chuyển đổi công suất được trao và các cơ hội trong danh mục dự án thành các hợp đồng đã cam kết, đáp ứng tiến độ của khách hàng và giảm chi phí thông qua việc phân bổ chi phí chung và quy mô mua sắm.

Rủi ro và các điểm cần theo dõi

• Khối lượng đơn hàng tồn đọng về công suất được trao không phải là đơn hàng chắc chắn hay sự đảm bảo về doanh thu trong tương lai. Một phần hoặc toàn bộ số tiền 2,4 tỷ USD này có thể không được chuyển đổi, và thời điểm chuyển đổi có thể khác với ước tính của ban quản lý.
• Chi phí sản xuất hiện tại vẫn cao hơn mức giá hợp đồng theo giá thị trường ở tốc độ sản xuất 37 megawatt. Điều này dẫn đến khoảng 4 triệu USD chi phí giảm giá hàng tồn kho và 13 triệu USD khoản lỗ từ các cam kết mua hàng chắc chắn liên quan đến Giai đoạn 0 của Fit Energy.
• Mục tiêu EBITDA điều chỉnh phụ thuộc một phần vào các quyết định do khách hàng kiểm soát, bao gồm việc Fit Energy lựa chọn triển khai các giai đoạn sau cũng như lịch trình xây dựng và giao hàng của khách hàng trung tâm dữ liệu.
• Việc mở rộng sản xuất đòi hỏi phải tuyển dụng thành công, mở rộng quy mô chuỗi cung ứng, lắp đặt thiết bị và thực hiện các sáng kiến giảm chi phí.
• Doanh thu phát điện tiếp tục chịu ảnh hưởng do sản lượng của nhà máy thấp hơn. Dự án Groton 7,4 megawatt đã ngừng hoạt động trong cả quý và dự kiến sẽ được nâng cấp theo kế hoạch vào năm tài chính 2027.

Những điểm nổi bật trong phần Hỏi & Đáp với chuyên gia phân tích

Ban quản lý cho biết việc phát điện bằng pin nhiên liệu tại chỗ có thể giảm bớt sự phụ thuộc vào việc nâng cấp lưới điện truyền tải, máy biến áp và cơ sở hạ tầng tiện ích khác. Họ cũng đề cập đến độ tin cậy, độ ồn thấp, vận hành không tiêu tốn nước và tiềm năng tích hợp hệ thống làm lạnh hấp thụ để cải thiện hiệu quả sử dụng năng lượng của trung tâm dữ liệu.

Về lộ trình đạt lợi nhuận, ban quản lý nhấn mạnh rằng FuelCell Energy không phụ thuộc vào một hoặc hai khách hàng. Ban quản lý chỉ ra danh mục đề xuất khoảng 10 gigawatt là nguồn bổ sung tiềm năng cho khối lượng đơn hàng tồn đọng, đồng thời thừa nhận rằng việc chuyển đổi sang hợp đồng thương mại vẫn là điều bắt buộc.

Liên quan đến Fit Energy, ban quản lý làm rõ rằng các giai đoạn sau có thể được tiến hành trước khi việc triển khai 30 megawatt ban đầu hoàn tất. Tiến độ phụ thuộc vào việc Fit Energy hoàn tất các thỏa thuận với các khách hàng của chính họ.

Ban quản lý cũng giải thích rằng các thỏa thuận giữ chỗ công suất bao gồm các mốc thời gian để đi đến hợp đồng chính thức. Mối quan hệ hợp tác này giúp khách hàng tiếp cận công suất sản xuất trong tương lai, đồng thời giúp FuelCell Energy chủ động hơn trong việc lập kế hoạch sản xuất và vật tư.

Toàn văn Biên bản Cuộc họp Báo cáo Kết quả Kinh doanh

Toàn văn cuộc gọi công bố kết quả kinh doanh

Phần trình bày của ban lãnh đạo

Operator

Thank you for standing by. My name is Jaylen and I'll be your conference operator today. At this time, I would like to welcome everyone to the FuelCell Energy Third Quarter of Fiscal 2026 Financial Results Conference Call.

[Operator Instructions]

I would now like to turn the conference over to Michael Bishop, Chief Financial Officer. You may begin.

Michael Bishop

Thank you, Operator. Good morning, everyone, and thank you for joining us on the call today. This morning, FuelCell Energy released our financial results for the third quarter of fiscal year 2026, and our earnings press release is available in the Investors section of our website at www.fuelcellenergy.com. In addition to this call and our earnings press release, we have posted a slide presentation on our website. The webcast is being recorded and will be available for replay on our website approximately two hours after we conclude.

Before we begin, please note that some information that you will hear or be provided with today consists of forward-looking statements within the meaning of the Securities and Exchange Act of 1934. Such statements express our expectations, beliefs, and intentions regarding the future and include statements concerning our anticipated financial results, plans and expectations regarding the continuing development, commercialization, and financing of our fuel cell technology, our anticipated market opportunities, and our business plans and strategies. Our actual future results could differ materially from those described in or implied by such forward-looking statements because of a number of risks and uncertainties.

More information regarding such risks and uncertainties is available in the Safe Harbor Statement, in the slide presentation and in our filings with the SEC, particularly the risk factor section of our most recent Form 10-K and any subsequently filed quarterly reports on Form 10-Q. During this call, we'll be discussing certain non-GAAP financial measures, and we refer you to our website, our earnings press release, and the appendix of the slide presentation for the reconciliation of those measures to GAAP financial measures. Our earnings press release and a copy of today's webcast presentation are available on our website under the investor relations tab. For this call, I am joined by Jason Few, our President and Chief Executive Officer. Following our prepared remarks, the leadership team will be available to take your questions.

I will now hand the call over to Jason for opening remarks. Jason?

Jason Few

Thank you, Mike, and good morning, everyone. Thank you for joining us today. I am pleased to welcome you to our third quarter fiscal year 2026 earnings call. In the third quarter, we took an important step in the commercial development of FuelCell Energy's data center strategy. Rapid growth of AI and high-density computing is creating power requirements that the existing grid cannot address quickly enough. For data center customers, access to power has become a critical constraint on development. The AI economy will not be constrained by silicon. It will be constrained by access to electricity. We believe distributed generation will become an essential layer of AI infrastructure, enabling the grid to expand while allowing AI factories to deploy on commercial timelines rather than utility timelines.

Our FuelCell Energy Blocks are designed to address the constraint by providing clean, continuous, behind-the-meter power that can be deployed at the customer's site and scaled as demand grows. In the third quarter, we began to convert that value proposition into commercial commitments. We secured our first order for FuelCell Energy Blocks to supply baseload power for data center applications, increased committed backlog to $1.3 billion and added $2.4 billion of awarded capacity backlog, resulting in total committed and awarded capacity backlog of $3.6 billion as of July 31. Awarded capacity backlog is a new category reflecting multi-phase contracts and capacity reservations, which I'll let Mike detail in his remarks.

Subsequent to the quarter end, we closed a 75-megawatt capacity reservation agreement with a major co-location data center operator for a Texas project. We believe our utility-scale distributed generation platform is uniquely positioned to help accelerate AI infrastructure by reducing time to power, extending the existing electric grid with reliable behind-the-meter generation, and addressing many of the permitting and community challenges associated with large-scale power development. We expect to provide additional detail upon execution of definitive agreements. That is the central message for the quarter. We are moving from a growing pipeline to tangible commercial commitments while advancing the manufacturing capacity and operating capabilities required to deliver at scale.

At the outset, I want to talk about this commercial commitment. During the quarter, we signed a capital equipment purchase agreement with Fit Energy to supply power solutions for data center applications. It covers up to 380 megawatts across 4 phases, sized to the customer's deployment schedule. We received an upfront deposit on the initial 30-megawatt phase, which we expect to begin delivering in the fourth quarter, and the remaining phases are at Fit Energy's election. At the same time, our fiscal 2026 year-to-date pipeline has grown to roughly 10 gigawatts of active proposals, and it reflects our progress toward proving our value proposition for data centers, which now accounts for about 97% of the total third quarter pipeline.

I have said before that pipeline is a leading indicator, not a result, and I hold to that. The measure that matters is conversion, and Fit Energy is commercial proof that FuelCell Energy Block system can be the solution to some of the current public perception challenges facing data centers by providing scalable, clean, quiet, behind-the-meter power for data centers. The most important signal isn't that our pipeline is larger, it's that customers are buying differently. AI has made power availability a strategic decision rather than simply a utility decision.

In addition to our domestic backlog, we continue to execute on our existing global projects. In the third quarter, we successfully completed the repowering of the 42-module Gyeonggi Green Energy project in South Korea. This execution demonstrates our capability to manage complex utility-scale repowering projects overseas while maintaining strict operational standards. Furthermore, it validates our long-term technology replacement cycle, proving that our existing fleet represents a continuous source of service and product revenues as energy blocks reach their natural replacement intervals.

Important to converting our existing pipeline and backlog is our ability to scale. Because demand only matters if we can meet it. To support our increasing backlog, we are systematically expanding our manufacturing capacity. We are actively expanding our Torrington, Connecticut manufacturing facility, to support the multi-megawatt demand of the AI factory and data center markets.

Our immediate operational milestone is to increase our annualized production rate at Torrington to its current full capacity of 100 megawatts, with achievement of this milestone expected in October 2026. This near-term target represents a vital step toward our larger long-term goal of reaching 500 megawatts of annualized production capacity by June 2028, an expansion we are already investing in.

This expansion is progressing on schedule. During the third quarter, we finalized the comprehensive factory design, made significant equipment purchase commitments, and began the installation of a new high-volume tape caster that will dramatically increase our throughput. It is important to emphasize that this expansion is fully funded. We are executing this capital spend in alignment with our committed backlog to ensure disciplined capital allocation with the goal of meeting the high-volume requirements of global hyperscalers without building ahead of the market.

As we scale, one aspect of our fuel cells has come into particular focus, and that is the sourcing strategy for our materials. Our carbonate platform provides a powerful supply chain advantage, and it does not rely on rare earth minerals and is scandium-free, utilizing globally abundant commodity metals like nickel and steel rather than highly volatile critical minerals or those that are predominantly mined in potentially sanctioned countries. Our platform was designed around the abundant industrial materials, not scarce critical minerals. With over 90% of our supply chain is sourced domestically in the U.S., and approximately 93% of our FuelCell Energy Block components are reusable or recyclable through a take-back program, we offer our customers unmatched supply security in the current uncertain geopolitical environment.

Along with growing demand for FuelCell Energy power systems, our technology is being validated on a global stage by an increasingly diverse group of world-class blue-chip partners. We are proud to report that we have delivered and installed the first 2 carbonate fuel cell carbon capture modules at ExxonMobil's Rotterdam complex in the Netherlands. This delivery represents a pivotal operational milestone under our joint development agreement with ExxonMobil Technology and Engineering Company. This installation is the world's first industrial-scale demonstration of our jointly developed carbon capture technology, successfully moving it out of the laboratory and into a real-world application, addressing hard-to-abate low CO2 emissions from an industrial facility. This Rotterdam demonstration is expected to validate our fuel cells' performance under commercial operating conditions, positioning us as an essential technology partner for global industrial decarbonization.

During the third quarter, we also signed a memorandum of understanding with Siemens to design and supply the electrical balance of plant systems for our fuel cell installations. The primary goal of this collaboration is to accelerate physical deployment and lower the cost of large-scale commercial projects exceeding 100 megawatts. We plan to jointly develop integrated distributed energy systems that combine our clean fuel cells with battery energy storage, advanced micro-grid controls, and medium-voltage electrical equipment. By optimizing the electrical balance of plant, we can manage the full spectrum of power variability from minutes down to microseconds. We believe this integrated solution to be developed in collaboration with a global leader would provide the electrical reliability required to support critical, high-density AI data center workloads.

The opportunity in front of FuelCell Energy continues to grow. Our responsibility is straightforward: Execute. We are focused on converting commercial demand into contracted backlog, scaling manufacturing with discipline, and delivering for our customers. Those are the measures by which we should be judged, and they will remain our priorities as we work to build long-term shareholder value. With that, I'll turn the call over to our Chief Financial Officer, Mike Bishop, to provide a breakdown of our financial performance.

Michael Bishop

Thank you, Jason. Today I will walk through our third quarter fiscal 2026 financial results, which demonstrate our robust capital position alongside a transitional period for our top-line revenue. Total revenue for the third quarter of fiscal 2026 was $33 million, a 29% decline compared to $46.7 million in the third quarter of fiscal 2025. Breaking this total down, product revenue was $18 million, down from $26 million in the prior year quarter. This reflects fewer module deliveries to South Korea as we completed the repowering of Gyeonggi Green Energy fuel cell park, delivering all 42 modules committed under that program since 2024.

Service revenue was $2.4 million compared to $3.1 million a year ago. Generation revenue was $8.8 million, down from $12.4 million, driven principally by lower output from plants in our generation portfolio, including our 7.4-megawatt Groton project, which was out of service for the full quarter pending a planned upgrade that we expect to complete in fiscal 2027. Finally, advanced technology contract revenue was $3.8 million compared to $5.3 million in the third quarter of fiscal 2025.

We recorded a gross loss of $24.5 million in the third quarter of fiscal 2026 compared to a gross loss of $5.1 million in the third quarter of fiscal 2025. The primary driver was $17 million of charges recorded during the quarter, consisting of approximately $4 million to reduce the carrying value of certain inventories to net realizable value, and approximately $13 million for losses on firm purchase commitments. Both were recorded in connection with Phase 0 of our capital equipment purchase agreement, or CEPA, with Fit Energy due to the fact that our current product costs and manufacturing overhead exceed the contractual pricing established under that agreement.

We operated at an annualized production rate of approximately 37 megawatts during the quarter, which remains below the volume at which we expect our cost structure to align with market-based pricing for orders of this scale. These charges are expected to be limited to identified inventory and purchase commitments for Phase 0 and do not reflect our expectations regarding the overall economic value of the agreement.

The loss from operations was $46.7 million, a 51% decrease compared to an operating loss of $95.4 million in the third quarter of fiscal 2025. That improvement was primarily driven by the absence of the asset impairment and restructuring charges that heavily impacted the prior year period. Net loss for the quarter was $45.3 million compared to $91.9 million in the comparable prior year period, and net loss attributable to common stockholders was $45.3 million, or $0.64 per share, compared to $92.5 million, or $3.78 per share, in the prior year quarter. Per share improvement also reflects a higher weighted average share count of 70.4 million shares following our equity issuances over the past 12 months.

On a non-GAAP basis, adjusted EBITDA was negative $36.7 million compared to negative $16.4 million in the third quarter of fiscal 2025. That variance was primarily driven by Phase 0 charges I just described, which are not added back in our adjusted EBITDA reconciliation, rather than by any structural degradation in our core operating model.

Turning to our commercial progress, we are encouraged by the substantial expansion and evolution of our backlog. As of July 31, 2026, total committed and awarded capacity backlog was $3.6 billion, a significant step change. We have structured our commercial backlog into 2 distinct categories to give investors clear visibility: committed backlog and awarded capacity backlog. Committed backlog, which represents definitive non-cancellable agreements executed by the company and its customers, was $1.3 billion, up approximately 4.1% year-over-year. Awarded capacity backlog was $2.4 billion. Awarded capacity backlog represents commercial awards and capacity reservations where we have been selected as the supplier and the parties are advancing towards execution of definitive agreements.

For the third quarter, this category is driven by the 350 megawatts across Phases 1, 2, and 3 of our CEPA with Fit Energy, which was executed in June and provides for up to 380 megawatts in total product, commissioning, and service agreements, including the committed 30-megawatt Phase 0. Fit Energy may elect to proceed with Phases 1, 2, and 3 at its sole option, and no payment obligation arises with respect to a phase until Fit Energy makes an election to proceed with that phase.

I want to be clear that awarded capacity backlog is not contracted firm order backlog or a guarantee of future revenue. Amounts may not convert to committed backlog or to revenue in whole or in part, and the timing and amount of any conversion may differ materially from our current estimates.

We continue to maintain tight fiscal controls across the company. As summarized on slide 19 of the presentation, total operating expenses for the third quarter of fiscal 2026 were $22.2 million compared to $90.2 million in the third quarter of fiscal 2025. Looking at the details, administrative and selling expenses were $13.6 million for the quarter. Research and development expenses were $8.5 million for the quarter as we continue to invest in key product initiatives to support growth of data center opportunities.

This year-over-year reduction in operating expenses was primarily driven by the absence of $68.5 million of asset impairment and restructuring charges incurred during the third quarter of fiscal 2025. Excluding those 1-time historical charges, recurring operating expenses were essentially flat year-over-year with a modest reinvestment in research and development offsetting lower administrative and selling costs.

Now turning to the balance sheet and liquidity discussed on slide 21. We ended the quarter with the strongest cash position in our history. Total cash, cash equivalents, and restricted cash as of July 31, 2026, was $737.3 million, up from $440.9 million at April 30, 2026, the end of the prior quarter. Unrestricted cash and cash equivalents represented $658.1 million of that total, with the remaining $79.2 million in restricted cash and cash equivalents pledged as collateral for performance security and letters of credit. Our capital structure also remains straightforward. We carry no corporate convertible or high-yield debt, and our $153.6 million of total debt and finance obligations primarily consist of project-level financing, Export-Import Bank working capital facilities supported by our Korean deliveries, and sale-leaseback obligations.

This substantial capital buffer means that our manufacturing capacity expansion at our Torrington, Connecticut, facility is fully funded. We estimate the total requirement to expand Torrington to 500 megawatts of annualized production capacity to be between $200 million and $275 million with completion targeted for June 2028. The expansion is backed by approximately $298 million of net proceeds raised from sales of common stock during the quarter, consisting of $245.5 million from our July underwritten offering and $52.9 million under our open market sale agreement.

Looking ahead, we believe our strength in balance sheet and backlog expansion have established a clear path toward mid-term profitability. We are now targeting achieving positive adjusted EBITDA results in the fourth quarter of fiscal year 2027. We believe this target is supported by a series of operational and commercial catalysts. First, we have begun to increase our annualized production rate with the goal of achieving targeted annualized production rate of 100 megawatts in October 2026, up from approximately 37 megawatts this quarter, which should drive operating leverage over time.

Beyond that, reaching our adjusted EBITDA target will depend on several key factors, including conversion of our awarded capacity backlog into definitive revenue-generating committed contracts, alignment with customer delivery schedules, and continued execution of our manufacturing cost reduction initiative, as we benefit from higher procurement volumes. There can be no assurance that we will achieve these production rates, the conversion of awarded capacity backlog or the anticipated cost reductions within the timeframe currently expected.

In closing, we are executing our strategy with financial discipline, a fully funded manufacturing capacity expansion plan, and a sales pipeline that has grown to approximately 10 gigawatts in fiscal 2026 proposals, which we believe positions us to drive long-term value for our shareholders. Thank you for your continued support, and I will now hand the call back to the operator to open the line for Q&A.

Operator

[Operator Instructions] Your first question comes from the line of Julien Dumoulin-Smith of Jefferies.

Phần hỏi đáp

Ivana Ergovic

It's actually Ivana Ergovic for Julien. I just, kind of, had a question related to this deal announcement of 75 megawatts. If you could maybe give some, kind of, more details in terms of the timeline on any potential opportunities for expansion versus that deal. I think that, yeah, that would be my first question.

Jason Few

Thanks for joining us this morning. Yes, after the quarter or subsequent to the end of the quarter, we closed the 75-megawatt capacity reservation agreement for a major data center operator. We've not disclosed the timing of that, but we anticipate not only that opportunity, but follow-on opportunities with the same customer. And as you've seen, particularly in certain markets and including a market like Texas, where there's, you know, movement toward requiring bringing your own power. Our platform certainly sets up well to meet that requirement in Texas. And so we're excited about the opportunity.

We were working through the definitive agreement, and that'll really align the timeline from a delivery and execution standpoint. We see this as a continuation or how the model, the business model, is really evolving to put capacity reservations in place as customers really look to line up power while they're completing their designs for the architecture of the data center and securing their commitments from their offtake customers as well. So excited about this opportunity and look forward to executing.

Ivana Ergovic

Thank you. And in terms of a follow-up, I actually had a little bit of different question, kind of, related to your quarter results. I mean, there is a material increase in the cost of revenue, I guess, related to the Fit Energy deal. So I mean, how should we think about it? In the sense of the revenue recognition, I guess those would come with the deliveries in the fourth quarter and offsetting basically the cost of revenue that you booked in this quarter. Is that the right way to think about it?

Michael Bishop

Sure, Ivana. This is Mike, and thanks for joining the call. I'll take that one. So, as far as the Fit Energy Phase 0, yes, we have disclosed that we do expect to begin recognizing revenue on that order in the fourth quarter of our fiscal year with the balance of it being completed in fiscal 2027. On the cost side, what you've seen come through this quarter related to that order is really our legacy cost structure. As we sit here today, the company's operating at 37 megawatts of production volume, which has the cost higher than current market rates.

We expect that to normalize and be absorbed as we scale and get production rates up to 100 megawatts. As I said in my remarks, we do expect the company to get to adjusted EBITDA positive in the fourth quarter of fiscal 2027.

Ivana Ergovic

Can I just ask one more thing? In terms of the 100 megawatts, it seems that you should be able to, kind of, start producing at those levels by the year-end?

Michael Bishop

Yes. So what I believe your question was, where is our production rate going? We have announced that we are scaling our production rate up to 100 megawatts by the end of the fourth quarter of this year. And what that means is adding personnel, direct labor in our factory as well as scaling our supply chain so that we're positioned to be able to deliver at that level as we get into fiscal 2027. In addition to that, we're adding manufacturing capacity as well to go up to 500 megawatts of total capacity in this factory by June of 2028.

Operator

Your next question comes from the line of Manav Gupta of UBS.

Manav Gupta

I wanted to focus more on the Exxon power project. I mean, it looks like your cells are delivered. I'm just fundamentally trying to understand, are these 2 cells going to operate in a different way because their primary goal seems carbon capture? Can you help us understand how these 2 cells will be operating in a different way with Exxon, and what's the scope of expanding that partnership because, you know, Exxon is very bullish on carbon capture as a whole.

Jason Few

Manav, good morning, and thank you for joining the call, and thank you for the question. Yes, you are correct. The 2 modules have been delivered to Exxon at Rotterdam and are being installed. The primary focus of the application for those 2 modules is capturing carbon directly from the point source of emissions at the Exxon Rotterdam refinery or the Esso refinery, given that's the brand name they still use in Europe. It will demonstrate capturing 90% plus of the carbon while simultaneously producing power, thermal energy, and hydrogen, which is a unique capability to our platform, not only as a fuel cell provider, but a unique capability in terms of other carbon capture technologies.

The other big part of this demonstration, Manav, is to show our ability to capture CO2 from a low CO2 concentration stream of emissions, which is much harder to do. And that's another area where our technology accelerates in terms of our capability to actually capture low concentration CO2, which opens up the biggest aperture of market opportunity when you look across the industrial landscape. So everything from a company that does bottling and uses boilers to sterilize bottles all the way to the refinery application we're demonstrating here in Rotterdam.

We believe that demonstrating successfully our platform's capability, that, that will open up an opportunity to expand this technology more broadly to address carbon capture across industrial applications globally. And we think that there continues to be broad political support. I mean, if you look at the OBBBA, the actual, you know, incentives around 45Q actually improved. So we think that, you know, signals strong support, clearly strong support for carbon capture in Europe and Asia. And the product will function differently from a core focus being carbon capture versus our power generation. That being said, our core product and every product we ship today is carbon capture ready. So we have the ability to also decarbonize power generation by capturing the CO2 from the fuel that we use to power our energy block. But the core focus in Rotterdam is carbon capture.

Manav Gupta

Perfect. My quick follow-up here is in your opening comments, you talked about fuel cell and, you know, the time to power advantage and how the grid is not scaling up. What we have also noticed is that there's obviously something called LCOE, but increasingly, what the hyperscalers are finding out is by the time the electricity is delivered to you, it's not even close to LCOE, it's significantly higher. Now when you take that versus the benefits of on-site power generation, which is basically a spark spread, the cells actually start becoming a lot more economical and have a break-even of 7 or 8 years. So I'm just trying to understand from your perspective, can you also help us understand some of the other benefits of on-site power generation? Because on the screen, might look the LCOE is cheaper, but by the time the power gets to the data center, it's actually a lot more expensive. So if you could talk a little bit about that.

Jason Few

Sure, great question. I think if you look at the advantage of on-site power and LCOE, you're absolutely correct. If you look at what's really required to deploy new power generation in a constrained area, the amount of capital investment that has to go in from an infrastructure upgrade on the utility side, which may include high-voltage transmission, new local transmission, additional power electronics from transformers, et cetera, you can reduce a significant amount of that cost by doing on-site power generation. So not only do you get a lower LCOE, higher reliability, you're also going to get the ability with our platform to integrate absorption chilling so you can actually bring down the PUE of that data center, which is a core goal of a data center operator because they want to get more of the power to the compute, because that's really the business they're in, is powering compute.

The other big benefits are, we offer a low noise solution, and we can operate in water neutral. So a lot of the things that communities are complaining about today, we address with our technology, including not contributing to poor air quality, because although we use natural gas, we don't combust the fuel, which is another advantage. And so you take all of those things together, and I like to think about it more than just time to power, but it's really time to power on, because it's the time to deliver it, can you get it permitted, which is a big challenge today, and our platform really addresses a lot of those concerns. And so really time to power on, which is also time to revenue, and we think that creates a significant advantage.

Operator

Your next question comes from the line of Jason Tilchen of Canaccord Genuity.

Jason Tilchen

I guess to start, can you perhaps help us bridge the gap between the achievement of the 100-megawatt run rate that you, sort of, stated you expect at some point next month, to the Q4 of fiscal 2027, sort of, updated target for reaching EBITDA profitability and what some of the key factors, maybe expand on some of the key factors that you laid out in the press release that could help you achieve that rate.

Michael Bishop

Sure, Jason. This is Mike. I'll take that. So, again, as we said, we are hiring, we are ramping our supply chain to get our run rate up to that 100-megawatt annualized run rate in the fourth quarter of fiscal '26 and targeting adjusted EBITDA positive in fourth quarter of fiscal 2027. Between now and then, the key factors that will drive that is: One, continuing to convert our awarded capacity into committed backlog. As we sit here today, we are executing on the 30-megawatt committed backlog from Fit Energy Phase 0. We need to continue to convert that broader order as well as other opportunities. And of course, we talked about this new 75-megawatt capacity reservation agreement that was announced this morning. So that's another opportunity there to continue to convert.

So converting backlog, lining up with customer schedules, and then, of course, continuing down the cost reduction curve as we expand and we scale in the factory, we will absorb overhead and we will also get leverage from our supply chain. So those are the main drivers that will be occurring in the financial statements over the course of the next year plus.

Jason Tilchen

Okay, that's really helpful. And I guess the follow-up there is in terms of those factors you laid out, some of those are in your control and some of them are at the discretion of the customers like Fit Energy has the option to proceed with these deals. So I'm just curious maybe if you could help us understand on the, sort of, cost reduction side, if you feel like you are already or within close line of sight of achieving what you need to achieve to get to that point. And then what does the scenario look like where maybe Fit Energy is not ready to proceed with that phase in time to achieve it? Like, what are the -- what is the sliding scale look like in essence, if you're able to achieve some of those milestones, but maybe not all of them.

Michael Bishop

So, again, on the cost reduction side, we've been planning for this for a long time. We have a very well-defined cost reduction curve in front of us that we are executing on. And on the commercial side, we talked about a 10-gigawatt pipeline of opportunities. So we are not just relying on 1 or 2 customers, we see significant opportunities here across our customer base and fully expect to be able to convert additional pipeline into backlog over this time period.

Jason Tilchen

Okay, really helpful. And maybe we can sneak into one quick final one. I believe in the first question that was asked, it was around the bridge from 37 to 100 over the next, sort of, call it, 6 to 8 weeks. And you mentioned adding labor and increasing the supply chain to get to that run rate. Are those, sort of, boxes, have those been ticked already and it's just a matter of simply working through the next few weeks of just getting those people up and running? Or are you still in the process of finding that labor and making sure that supply chain is at the right point?

Michael Bishop

Yes, so there's multiple elements there. But yes, we've made considerable progress. We are hiring as we speak. We have added an additional shift in our factory. So you will see meaningful increase in our production rate come through this quarter as we described.

Operator

Your next question comes from Ryan Pfingst of B. Riley Securities.

Ryan Pfingst

Maybe just to start with a follow-up on the last one, for the target of positive EBITDA, in fiscal 4Q '27, can you frame that in terms of what that reflects or where you expect to be from an annualized production rate perspective at that point?

Michael Bishop

Thanks for the question, Ryan. So we would expect at least 100 megawatts of volume to support that, if not more. And again, that will be dictated to some extent by our customers in converting pipeline to backlog as well as customer delivery schedules. So being able to line up with our customer requirements. And as we described, we will absolutely have the capability in Torrington to get above 100 megawatts. We are targeting 500 megawatts of capacity by June of 2028. That does not get turned on like a light switch. That will come online over time as we unlock constraints. One of the big constraints that we've talked about is tape casting. That process is well underway and will be installed in next fiscal year, so that's something. That's a big unlock for us to create additional capacity.

Ryan Pfingst

Got it. And then on the Fit Energy deal, curious on your expectations for moving to the remaining phases. Do you see Phase 1 moving forward only after the initial phase is up and operating, or could that decision, in your view, come before then?

Jason Few

No, Ryan, this is Jason. No, they are not sequential or certainly don't have to be sequential. We -- the gating on those is just their ability to close out their agreements with their customer and that can happen at any time and so it's not a complete the first 30 megawatts and then, you know, they'll move forward or consider Phase 1, they're not a sequential set of events.

Operator

Your next question comes from the line of Noel Parks of Tuohy Brothers.

Noel Parks

I noticed that with the updated CapEx guidance, it's actually been nudged down a bit for the fiscal year. And I think there was a mention in the materials that some of that reflected equipment deliveries that would not be happening until after the fiscal year end. So I was just wondering if you could just elaborate a bit on that. And I'm assuming that would be deliveries for the, you know, primarily for the Torrington expansion.

Michael Bishop

Correct. This is Mike. So good read. The previous disclosure that we had around expected CapEx coming through the cash flow statement in fiscal '26 was $20 million to $30 million. We are now saying that that's $10 million to $20 million, just given timing, but it is, by no means, a reflection of execution, it is really a timing of receiving certain equipment based on our prior schedules. As we have described now, we are on track for the 500-megawatt expansion by June of 2028. We have committed $200 million to $275 million of capital to that and if you look at our purchase commitments disclosures, that is up significantly quarter-over-quarter and reflecting not only the increase in production rate, but also capital commitments that will be delivered in future quarters.

Noel Parks

Great, thanks. And also, early in the prepared remarks, I think there was a -- or maybe it was right at the start of Q&A, there was a comment about sort of your fuel cells, I guess the necessity of aligning with customer delivery schedules. And so I was just, sort of, thinking in this ramp up period on one hand, and then you have new agreements in negotiation and coming online. Just thinking a bit about how you manage that since, it looks like you're going to be kind of in anything but a steady state, sort of, trend for the next few years.

Jason Few

Yes, so this is Jason. So the way that you can think about this is we have visibility into our production capabilities. We have visibility into our expansion capacity ramp. And that information is closely tied to and fully understood by our business development team. And so when we're talking to customers about opportunities and schedule always becomes part of the question or conversation ultimately, we make sure that the commitments that we're making align to our view in terms of our ability to meet and deliver against that demand. And so even though you might sign an agreement for 100 megawatts of power, you really need to look at, well, what is the delivery schedule for that? Because first power could be something as small as 20 megawatts. And additional power comes on as they finish building out the data hall, or maybe as they add a second building for a second data hall. So you have to really look at the full build out from the customer perspective, and then we make sure that we align our commitments to our confidence on our scale-up and manufacturing capabilities.

Operator

[Operator Instructions]

Your next question comes in line of Christopher Ellinghaus of Siebert Williams Shank.

Christopher Ellinghaus

Mike, in your backlog slide, do you, sort of, envision adding more granular categories, sort of, as you convert pipeline?

Michael Bishop

Good morning, Chris, and thanks for joining the call. So obviously a big step change in our backlog disclosures this quarter compared to last quarter. We've added an additional category called Awarded Capacity Backlog, and what is in there as of July 31, 2026, the end of last quarter, is really the 3 phases of the Fit Energy contract that aren't currently committed. So this is Phases 1, 2, and 3 broken down between product and service. So we think we have a fair amount of disclosure there on the potential that will be converted into committed backlog.

As far as additional categories, we would expect, and we talked about the 75-megawatt capacity reservation agreement. We would expect that to go into the awarded capacity backlog as well, whether it's part of product or service or broken between the two, we'll disclose that in future periods. But that's the expectation, kind of, following industry practice. These days where folks are disclosing reservations whether they're called CRAs or something else but really reservations on future capacity which the company is being paid for. So we will be as transparent as possible as we add these additional opportunities into our backlog.

Christopher Ellinghaus

And Jason, I sort of liked the way you framed the pipeline versus conversion discussion. Is there -- I don't know how to put this, but is there any timelines or waypoints for some of the pipeline that we might look for you to announce any kind of agreement out of the pipeline?

Jason Few

Sure. No, great question, Chris. And thank you for being on the call. The way you ought to think about it is, you know, we talk about our sales pipeline and the 10 gigawatts we talked about today are, you know, tied to proposals that our business development team has with customers directly placed those proposals. When we talk about awarded capacity agreements, what we're talking about there is that we, along with the customer, have agreed to allocate manufacturing capacity in support of that customer's order.

Along with that awarded capacity, there is a date certain in which we and that customer have agreed that we're going to get to the definitive agreement and at that point that awarded capacity would convert into our backlog what Mike just talked about and what you find in our backlog are things that are committed firm project orders and that's the way we've, you know, reported backlog for probably the last 6 or 7 years in terms of that backlog number.

So what you should really look for is the velocity in which we take awarded capacity and convert that into committed definitive agreement, but in each of those awarded capacity agreements, there is a timeline tied to getting to that definitive agreement. And again, what it gives the customer -- the benefit for the customer is it reserves manufacturing capacity, right? And it gives them the ability to complete their final designs from an architecture perspective in terms of everything from our energy block all the way to the rack and how they're going to do that and to finalize their agreements with their customers. And it gives us better visibility into our planning, gives us the ability to make purchases on materials from a supply perspective, and there's financial consideration for providing that committed capacity reservation.

Operator

With no further questions, that concludes our Q&A session. I'll turn the conference back over to President and CEO Jason Few for closing remarks.

Jason Few

Thank you, Jaylen. And before we conclude, I want to leave you with this. The opportunity in front of FuelCell Energy is significant. AI and high-density computing are creating an urgent need for reliable, scalable power, and our technology is positioned to help meet that need. But opportunity alone does not create value, execution does. Our focus is clear. Convert awarded capacity and our pipeline into closed transactions, scale manufacturing with discipline, and deliver for our customers. We will align investment with commercial commitments, increase capacity responsibly, and scale to drive greater cost leverage across the business.

Partnership will be central to that execution. Our collaboration with Fit Energy, Siemens, and the customers reserving manufacturing capacity demonstrates what is possible when technology, capital, industrial capability, and customer demand come together around a shared objective. These are not simply commercial relationships, they are partnerships built to accelerate deployment, reduce execution risk, and create long-term value for everyone involved. We believe the market is moving toward us. Power has become a strategic constraint and FuelCell Energy has the technology, manufacturing platform and partnerships to help customers move from ambition to operation. Now we must execute with urgency, discipline, and consistency.

Before we close, I want to thank the entire FuelCell Energy team. Your commitment, resilience, and focus makes our progress possible. I also want to thank our customers, partners, and shareholders for your continued confidence and support. We appreciate your time today and look forward to updating you on our progress next quarter.

Operator

Thank you. This concludes today's conference call. You may now disconnect.