FuelCell Energy (Nasdaq: $FCEL): AI Data Center Power, Carbon Capture, Hydrogen and the Execution Test Behind the Rally

01 August 6, 2026 update: a 12 million share warrant package tied to customer deposits, priced 27% above the market

On July 21, 2026 FuelCell Energy filed a prospectus supplement to its automatic shelf registration covering the resale of up to 12,000,000 shares issuable on exercise of warrants held by FIT Energy USA LP. The warrants were issued on June 22, 2026 and are structured in three tranches of 4,000,000 shares each, with a strike price of $26.44.

Two features separate this from an ordinary dilution event, and both matter more than the headline share count.

The warrants vest only if the customer pays. Vesting is tied to FIT’s payment of non-refundable deposits under a Capital Equipment Purchase Agreement signed the same day, covering phases 1, 2 and 3. Any tranche that has not vested by June 22, 2028 terminates automatically and is cancelled. Once a tranche vests it expires if not exercised within 24 months. In other words the dilution arrives only alongside equipment orders that have already been paid for in part, and disappears entirely if those orders do not materialise.

The strike is well above the market. At the August 7, 2026 close of $20.43, the $26.44 strike sits roughly 29% out of the money. The stock spent the first week of August drifting: $22.50 on August 4, $21.14 on August 5, $20.32 on August 6 and $20.43 on August 7, having traded as low as $18.69 intraday on August 7. For FIT to exercise, the shares have to appreciate materially from here. If all three tranches vest and are exercised, FuelCell would receive approximately $317.3 million in exercise proceeds. The company does not receive any proceeds from FIT’s subsequent resale of those shares, only the exercise price itself.

Do not confuse this with the July equity raise. They are two separate transactions that happen to involve similar share counts. The July offering was completed: 12,321,429 shares priced at $21.00, roughly $245.4 million of estimated net proceeds, cash already received. The FIT warrants are 12,000,000 shares that do not exist yet, may never exist, and would only be issued at $26.44 if a customer keeps paying deposits on an equipment contract. One is dilution that has happened; the other is dilution contingent on revenue.

02 July 22, 2026 update: Siemens strengthens the deployment architecture, while the equity raise was larger than the original headline

The latest official operating release remains the July 9 collaboration between FuelCell Energy and Siemens. The memorandum of understanding does not create a customer order or guaranteed backlog, but it addresses a practical bottleneck in large distributed-power projects: integrating fuel-cell generation with electrical balance-of-plant equipment, battery storage, microgrid controls and medium-voltage infrastructure.

Under the collaboration, Siemens is expected to design and supply electrical balance-of-plant systems intended to support rapid deployment of commercial projects exceeding 100 MW. The parties also plan joint engineering, integration and delivery work, plus pilot and solution-development initiatives that may include medium-voltage direct-current power delivery and modular electrical systems. Successful pilots are intended to create a path toward full-scale deployments, but the announcement remains an MOU rather than a binding megawatt purchase agreement.

The July offering also needs a more precise description. FuelCell Energy priced 10,714,286 shares at $21.00, and the underwriters fully exercised their option for another 1,607,143 shares on July 8. That brings the planned total to 12,321,429 shares. The SEC-filed materials estimated approximately $245.4 million of net proceeds after discounts and expenses and approximately 79,929,602 common shares outstanding after full exercise. The company said the proceeds are intended for manufacturing-capacity expansion, working capital and general corporate purposes.

The Korean financing has also moved beyond approval alone. The offering prospectus states that the first EXIM tranche was disbursed on June 30 and provided approximately $22 million of net proceeds after fees, expenses and reserves to support delivery of five 2.8 MW blocks to Gyeonggi Green Energy. A second tranche was expected in October 2026, subject to customary conditions.

As of July 22, no later operating press release had replaced the Siemens announcement. On the market side, FCEL closed at $21.64 on July 21 after three consecutive gains from the July 16 closing low, but remained approximately 42.9% below the June 30 high. That rebound improves the very short-term tape without confirming that the post-offering correction has ended.

Updated interpretation: Siemens improves the credibility of FCEL’s deployment ecosystem, the fully exercised offering materially strengthens liquidity, and the first EXIM tranche reduces some Korean-project financing risk. None of these developments yet proves profitable data-center revenue conversion.

03 Executive summary: the story is real, but the proof is still ahead

FuelCell Energy has become one of the market’s most volatile ways to express the AI-power bottleneck theme. The investment narrative is no longer limited to hydrogen enthusiasm or legacy fuel-cell projects. FCEL is presenting its carbonate platform as a source of continuous on-site electricity for data centers that cannot wait years for new transmission lines, utility interconnections or conventional generation capacity.

The commercial setup changed rapidly during June and early July 2026. Management highlighted a roughly 4 GW opportunity pipeline, introduced a standardized 12.5 MW power block, discussed manufacturing expansion in Torrington, signed a strategic framework with Fit Energy for up to 380 MW, advanced U.S. Export-Import Bank financing tied to equipment delivery for Gyeonggi Green Energy in South Korea, and then announced a Siemens collaboration focused on electrical integration for 100+ MW projects. Those developments helped push the stock to a 52-week high of $37.88 on June 30 and broadened the ecosystem around the data-center thesis.

The rally also exposed the structural weakness in the equity story. FCEL remains loss-making, reported negative gross profit in fiscal Q2 2026, relies heavily on external capital, and followed the surge with an underwritten public offering priced at $21.00. After the underwriters fully exercised the option, the planned issuance increased to 12,321,429 shares and the company estimated approximately $245.4 million of net proceeds. The market repriced that supply aggressively: FCEL later traded down to a $16.51 intraday low on July 17 before recovering to a $21.64 close on July 21.

The correct interpretation is neither “the AI story is fake” nor “every announced megawatt is future revenue.” The technology, customer problem and commercial interest are real. Siemens can improve system-level engineering and electrical integration, while EXIM and the equity raise improve financing flexibility. The unresolved issue remains whether management can convert non-binding or milestone-based frameworks into financed, permitted and delivered installations with acceptable margins. That conversion process—not headline pipeline size or partner-name recognition—is the central variable for the Stock Hub.

Merlintrader framing: FCEL is a high-volatility execution story. The upside narrative depends on data-center deployments, Korean shipments, carbon-capture progress and manufacturing scale. The downside narrative depends on delayed projects, negative margins and repeated use of equity markets before the commercial pipeline becomes self-funding.

04 Why FCEL matters now

The electricity constraint around AI infrastructure is creating demand for power technologies that can be deployed closer to the load. Large data centers increasingly need firm, around-the-clock electricity, but utility interconnection queues, transmission constraints and equipment shortages can delay conventional grid access. Fuel cells are being considered alongside gas turbines, nuclear projects, batteries, renewables and other distributed-generation solutions.

FuelCell Energy’s carbonate systems are not zero-emission in every configuration because many installations use natural gas. Their pitch is different: high-availability distributed power, low local pollutants, potential use of biogas, combined heat and power, hydrogen production and a pathway to capture concentrated carbon dioxide. That bundle may be useful where speed, footprint, resilience and emissions controls matter more than a single headline efficiency metric.

The company also has more than one route to value creation. Data centers are the fastest-moving market narrative, but FCEL retains exposure to Korean utility-scale installations, long-term generation assets, service contracts, carbon capture with ExxonMobil, hydrogen production and solid-oxide platforms. The diversity is strategically useful, although it also increases execution complexity and capital requirements.

The Siemens MOU matters because a fuel-cell power plant is only one component of a mission-critical data-center energy system. Customers also need switchgear, medium-voltage equipment, controls, storage integration and reliable delivery of electrical infrastructure. Combining FCEL generation with Siemens’ electrical balance-of-plant and integration capabilities could shorten design cycles and create a more complete commercial package. The limitation is equally important: the MOU does not identify a customer site, project value, deposit or binding equipment order.

The real question: can FCEL become a repeatable infrastructure supplier rather than a company that periodically announces large potential markets without producing sustained revenue growth and positive gross margins?

05 Company overview

FuelCell Energy is headquartered in Danbury, Connecticut and traces its history to Energy Research Corporation, founded in 1969. The company develops, manufactures, installs, operates and services fuel-cell platforms for distributed electricity generation, industrial decarbonization, hydrogen and carbon capture.

Its business model combines several revenue types. Product revenue comes from the sale of fuel-cell modules and balance-of-plant equipment. Service revenue is generated through maintenance and long-term support agreements. Generation revenue comes from projects owned or operated by FCEL under power-purchase arrangements. Advanced-technology revenue includes research, development and collaboration activity.

This mix has advantages and disadvantages. Generation assets can provide recurring revenue over long periods, while product sales may create larger but less predictable quarterly swings. Service economics depend on fleet reliability and contract structure. Research collaborations can validate technology but may not become large commercial revenue streams. Investors therefore need to look beyond a single revenue number and examine the quality, margin and durability of each category.

Jason Few has led the company as president and chief executive officer since 2019. The current strategic message is built around disciplined growth into power-constrained markets, but management’s credibility will ultimately depend on backlog conversion, manufacturing execution, project finance, gross-margin improvement and capital allocation.

06 Technology stack: what FCEL actually sells

Carbonate fuel cells and SureSource platforms

FCEL’s best-known systems use molten carbonate fuel-cell technology operating at high temperature. Fuel is converted electrochemically rather than burned in a conventional combustion cycle. The systems can supply continuous power with very low sulfur oxides, particulate emissions and nitrogen oxides compared with many combustion alternatives.

For data centers, the appeal is dispatchable on-site electricity. Unlike intermittent renewable sources, a fuel-cell plant can provide baseload output as long as fuel is available. The company also emphasizes direct-current architecture, modular installation, small land footprint relative to some alternatives and the possibility of using waste heat for cooling or industrial processes.

The standardized 12.5 MW power block

In 2026, FCEL began emphasizing a standardized 12.5 MW configuration intended to make commercial conversations easier and deployment more repeatable. Standardization matters because bespoke infrastructure projects can consume engineering resources, extend schedules and create cost variability. A repeatable block allows developers to think in increments that can be combined for larger campuses.

The commercial benefit is not guaranteed. Standardization must still be supported by reliable module production, balance-of-plant availability, permitting, fuel supply, financing and customer confidence in lifecycle cost. The product can shorten part of the sales cycle, but it does not eliminate infrastructure risk.

Carbon capture

FuelCell Energy’s carbonate platform can be configured to capture and concentrate carbon dioxide from industrial exhaust while generating electricity. The long-running collaboration with ExxonMobil is strategically important because it links FCEL to a major industrial partner and targets a market much larger than the current fuel-cell electricity business.

The partnership has generated development milestones and module shipments, but investors should distinguish technical progress from commercial deployment. The economic case depends on capture performance, system durability, project scale, carbon policy, storage infrastructure and customer willingness to fund installations.

Hydrogen and solid oxide

FCEL also develops hydrogen-production and solid-oxide technologies. Its Tri-gen configuration can produce electricity, heat and hydrogen from biogas or natural gas. Solid-oxide electrolysis offers another potential route into low-carbon hydrogen by using electricity and steam. These markets remain strategically attractive but capital intensive and highly sensitive to energy prices, policy support and competition.

07 The AI data-center pivot

The market’s 2026 re-rating of FCEL was driven less by the company’s historical business and more by a new thesis: grid-constrained AI facilities may require behind-the-meter or on-site power before utilities can provide conventional service. Management reported that a large majority of the company’s commercial opportunity pipeline was linked to data centers and AI-related power demand.

A pipeline is a sales funnel, not an order book. It can include early conversations, site studies, conditional frameworks, development agreements and opportunities that never reach financial close. Investors should therefore monitor movement through defined stages: customer selection, deposit, engineering notice to proceed, site control, interconnection or behind-the-meter approvals, fuel supply, project financing, manufacturing release, delivery, commissioning and revenue recognition.

The Fit Energy agreement is the most visible test. It contemplates up to 380 MW and includes an initial 30 MW component targeted for 2026 delivery. The initial deposit and milestone structure are more meaningful than a simple memorandum of understanding, but the total “up to” value should not be treated as guaranteed backlog. The market will need evidence that the first project remains on schedule and that later tranches become binding.

Best metric to watch: not the headline gigawatts, but the percentage of pipeline that advances into funded backlog and then into recognized revenue at improving gross margin.

08 Major commercial programs and strategic relationships

09 Timeline: the events that changed the 2026 story

FCEL introduced the standardized 12.5 MW power-block concept and described plans to expand Torrington manufacturing capacity toward 500 MW annually, subject to capital spending and commercial demand.

Fiscal Q2 results showed revenue of $35.6 million, a gross loss of $12.9 million and a large GAAP loss that included the Groton impairment. Management emphasized a roughly 4 GW opportunity pipeline and data-center demand.

The company announced the Fit Energy strategic agreement for up to 380 MW, including an initial 30 MW deployment target and milestone-linked economics.

The Russell US index reconstitution became effective. The event is now historical and should not be presented as a future catalyst.

EXIM approved approximately $49 million of financing for five 2.8 MW blocks for Gyeonggi Green Energy. The prospectus later disclosed that the first tranche was disbursed June 30, providing approximately $22 million net after fees, expenses and reserves.

FCEL reached a 52-week high of $37.88 after a rapid multi-week advance.

The company priced 10,714,286 shares at $21.00. The underwriters fully exercised the option for another 1,607,143 shares, bringing the planned issuance to 12,321,429 shares and estimated net proceeds to approximately $245.4 million.

Siemens and FuelCell Energy announced an MOU covering electrical balance-of-plant design, storage and microgrid integration, medium-voltage equipment, pilots and a path toward scalable 100+ MW commercial projects.

After falling to a $17.26 close on July 16 and a $16.51 intraday low on July 17, FCEL posted three consecutive gains and closed July 21 at $21.64. The rebound recovered the offering price but did not repair the larger decline from $37.88.

10 Financial position: cash is stronger, economics are not yet

FuelCell Energy’s fiscal second quarter ended April 30, 2026. Revenue was approximately $35.6 million, down about 5% year over year. The company reported a gross loss of approximately $12.9 million, an operating loss of approximately $77.9 million and a net loss of approximately $77.6 million. The quarter included a roughly $42.6 million impairment associated with the Groton project.

Adjusted EBITDA remained negative at approximately $17.1 million. The adjusted figure is more useful for comparing operating trends because it excludes the large impairment, but it still shows that the underlying business was not self-funding. Negative gross profit is especially important: increasing revenue does not automatically create value if project and service costs remain above recognized revenue.

Unrestricted cash and cash equivalents plus investments were approximately $373.2 million at April 30. Restricted cash increased the total cash pool, but restricted balances cannot be treated as fully available corporate liquidity. The company subsequently received approximately $22 million net from the first EXIM tranche for the Korean equipment program and raised a much larger amount through the July equity transaction.

The SEC-filed offering documents provide a clearer capital-structure reference than the original $225 million headline. FCEL offered 10,714,286 shares at $21.00, and the underwriters fully exercised their option for 1,607,143 additional shares. The full planned issuance was therefore 12,321,429 shares, representing approximately $258.75 million of gross proceeds. The company estimated approximately $245.4 million of net proceeds after underwriting discounts, commissions and expenses, with approximately 79,929,602 common shares outstanding after full exercise.

The offering improves liquidity and may support manufacturing expansion, working capital and project development. It also creates a materially larger equity base and a higher burden for per-share value creation. The next quarterly filing should confirm the final cash balance, period-end shares, actual closing effects and how much capital has already been committed to Torrington expansion or specific customer projects.

Balance-sheet conclusion: FCEL has more financial flexibility than it did before the 2026 rally, but the business is not yet economically self-sustaining. Cash reduces near-term survival risk; it does not remove dilution or execution risk.

11 Capital structure and dilution

Dilution is not a side issue for FCEL. It is one of the central components of the stock’s history and current valuation. During fiscal Q2, the company raised substantial capital through an at-the-market program. The July 2026 underwritten offering then added 12,321,429 planned shares after full exercise of the underwriters’ option, taking the prospectus estimate to approximately 79.93 million common shares outstanding after the transaction.

From the company’s perspective, raising capital into strength can be rational. Infrastructure projects require working capital, manufacturing investment, deposits to suppliers, engineering resources and project equity. A higher share price allows the company to raise the same amount of money with fewer shares than would have been required at lower prices.

From the shareholder’s perspective, the question is whether the capital creates value faster than the share count expands. If the proceeds finance profitable projects and convert pipeline into durable cash flow, dilution may be productive. If capital primarily funds recurring operating losses while commercial frameworks remain unconverted, per-share value can continue to erode.

What to monitor in every filing

• Basic and diluted weighted-average shares, period-end shares and any subsequent-event offering shares.
• Remaining ATM capacity, shelf registrations, warrants, preferred securities and equity compensation.
• Use of the approximately $245.4 million estimated net proceeds and whether spending is tied to identifiable projects, contracted demand and staged Torrington expansion.
• Project-level debt versus corporate equity financing.
• Cash burn after excluding restricted cash and one-time items.

12 Manufacturing and execution capacity

The Torrington, Connecticut facility is central to the data-center thesis. Management has discussed expanding capacity toward 500 MW annually, with potential investment of roughly $200 million to $275 million over a multi-year period. Adding capacity before orders are firm can create underutilized fixed costs; waiting too long can cause FCEL to miss customer delivery windows.

The optimal path is staged expansion tied to deposits, binding orders and project finance. Investors should watch hiring, shift additions, equipment commitments, supplier agreements and capital expenditures alongside backlog. Manufacturing announcements are more credible when accompanied by funded customer demand.

The Siemens collaboration could reduce part of the system-integration risk around large projects. Siemens is expected to contribute electrical balance-of-plant design and supply, including storage, microgrid controls and medium-voltage equipment. That can help FCEL present a more complete solution, but it does not remove the need for site control, permits, fuel supply, customer financing, equipment orders and profitable execution.

Quality and field reliability matter as much as nominal capacity. Fuel-cell stacks require replacement over time, and service obligations can create costs if performance is below assumptions. Fleet availability, service margins and warranty provisions should therefore be monitored together with new-unit shipments.

13 Competition and peer context

14 Basic technical analysis: where the stock stands

FCEL’s chart remains a high-volatility repair setup. The stock closed at $21.64 on July 21, approximately 42.9% below the June 30 high of $37.88. That close recovered the $21.00 offering price after three consecutive gains, but it followed a violent correction that reached a $17.26 close on July 16 and a $16.51 intraday low on July 17.

Momentum and volume

The post-offering decline was not a quiet consolidation. FCEL fell from the June 30 high to a July 17 intraday low of $16.51, a drawdown of approximately 56.4%. The stock then closed at $18.50 on July 17, $19.84 on July 20 and $21.64 on July 21. That sequence creates the first credible multi-session rebound since the offering shock.

The quality of the rebound is not yet decisive. July 21 volume was approximately 9.7 million shares, below the roughly 13.2 million-share 50-day average reported for the session. Price recovery on lighter volume can reflect reduced selling pressure, but it is less convincing than a breakout supported by expanding institutional demand. The chart still needs follow-through above nearby resistance and a successful higher low.

Approximate support zones

• $20.00–$21.00: first support and offering-price pivot. The July 21 close recovered this zone, but it must hold on pullbacks.
• $18.50–$19.50: secondary rebound support built around the July 17 and July 20 closes.
• $16.50–$17.30: major July swing-low zone. A breakdown below it would invalidate the current rebound structure.
• $15–$16: deeper support if the July low fails and the market retraces more of the data-center rerating.

Approximate resistance zones

• $22.20–$23.00: immediate resistance around the July 21 high, the July 14 range and the earlier rebound close.
• $24.50–$26.25: heavier post-offering supply and the next meaningful trend-repair area.
• $29.70–$31.90: major overhead resistance created by the pre-offering and early-July breakdown zone.
• $37.88: 52-week high and the maximum visible supply reference.

Technical attractiveness

At $21.64, FCEL has improved from the July lows but still does not offer a clean trend-following setup. The stock is attempting to build a rebound above the offering price while a large volume of recently issued shares and trapped buyers may create overhead supply.

For momentum-oriented traders, the setup would improve with a close above approximately $22.20–$23.00, follow-through toward $24.50–$26.25 and a later pullback that holds above $20–$21. For mean-reversion traders, the July low provides a visible invalidation reference, but the distance between support zones remains unusually large. A close back below $20 would weaken the rebound; a break below $16.50 would return the chart to active breakdown status.

Technical status: early rebound from a severe post-offering correction. Momentum has improved, but volume confirmation, resistance reclamation and a higher low are still required before the chart can be called constructive.

Technical levels are approximate chart zones based on data through the July 21, 2026 close, not recommendations or guaranteed support and resistance. They should be rechecked against the live chart because FCEL’s volatility can move levels rapidly.

15 Analyst positioning and market expectations

Analyst views became more constructive after the Fit Energy and EXIM developments. B. Riley upgraded FCEL and raised its target materially after the financing announcement, while Canaccord had already adopted a more bullish stance around the data-center opportunity. The Siemens MOU strengthens the industrial-partner narrative, but it does not by itself justify treating 100+ MW opportunities as contracted backlog. Other firms historically remained more cautious because of negative margins, dilution and the gap between pipeline and revenue.

Price targets should not be treated as intrinsic value. They are scenario outputs based on assumptions about project conversion, future margins, share count and capital spending. The July offering changes those inputs and may require analysts to update their per-share models even if enterprise value assumptions remain similar.

The useful question is not whether a target is above or below the current quote. It is what commercial conversion, margin and share-count assumptions are required to justify it. The fully exercised July offering materially increased the modeled equity base, while the Siemens MOU may improve project-conversion probabilities only if named pilots and binding customer contracts follow. FCEL’s valuation can change rapidly because a small change in probability assigned to hundreds of megawatts creates a large change in modeled revenue.

16 Ownership, sentiment and trading dynamics

FCEL attracts several different investor groups. Clean-energy investors focus on carbon capture, hydrogen and distributed generation. AI-infrastructure traders focus on data-center power scarcity. Event-driven traders follow offerings, index flows and analyst upgrades. Retail communities often focus on short interest, squeeze potential and comparisons with Bloom Energy.

Sentiment on Reddit, Stocktwits and X should be treated as non-professional market commentary. Bullish posts commonly emphasize the multi-gigawatt pipeline, the Fit Energy framework, EXIM support and the possibility that FCEL becomes a second major fuel-cell winner after Bloom. Bearish posts emphasize dilution, negative gross margins, legacy execution problems and the risk that “up to” megawatt announcements fail to become revenue.

Both sides identify real variables, but social sentiment is not evidence. The most reliable evidence will come from SEC filings, official customer milestones, shipment data, backlog, revenue mix, margin progression and cash-flow statements.

17 Upcoming catalysts and watchpoints

• Fit Energy initial 30 MW: engineering, manufacturing, financing and delivery milestones during 2026.
• EXIM second tranche: expected October 2026 funding, subject to conditions, followed by shipment and installation of the five 2.8 MW blocks for Gyeonggi Green Energy.
• Siemens collaboration: identification of a named pilot or customer, defined electrical-balance-of-plant scope and conversion from MOU to a binding commercial project.
• Fiscal Q3 2026 results: revenue, gross margin, cash burn, updated share count and pipeline-to-backlog conversion.
• Torrington expansion: capital-spending commitments, capacity additions and evidence that expansion is matched to funded demand.
• ExxonMobil carbon capture: module testing, demonstration economics and any move toward commercial deployment.
• Additional data-center contracts: preference for named sites, deposits, binding commitments and financing over broad “up to” frameworks.
• Capital markets: post-offering use of proceeds, remaining shelf capacity and any new ATM activity.
• Technical repair: hold above $20–$21, reclaim approximately $22.20–$23.00 and then challenge the $24.50–$26.25 supply zone.

18 Bull, base and bear scenarios

Bull scenario

Fit Energy advances beyond the initial 30 MW, EXIM-supported Korean deliveries occur on schedule, the Siemens framework produces a named 100+ MW-class deployment path and new data-center customers move into binding backlog. Revenue begins to accelerate, gross losses narrow and manufacturing expansion is funded by customer demand rather than recurring corporate dilution. The market starts valuing FCEL as a credible second platform in distributed data-center power.

Base scenario

The commercial pipeline remains large but converts gradually. Quarterly revenue is uneven, margins remain weak and the company uses its stronger cash position to fund several years of development. FCEL remains a volatile thematic stock whose valuation moves faster than its reported fundamentals.

Bear scenario

Framework agreements remain non-binding, the Siemens relationship does not produce funded customer projects, initial deployments require more capital than expected and revenue remains below the level needed for positive gross margins. The approximately $245.4 million estimated net equity raise extends runway but does not solve the operating model. Additional equity is eventually required, and the stock retraces more of the 2026 rerating.

19 Red flags investors should not ignore

Pipeline is not backlog

Opportunity pipelines can change quickly and may include projects without binding customer commitments. A multi-gigawatt funnel is strategically encouraging, but it should not be capitalized as if every megawatt will be delivered.

Negative gross margin

A company can grow revenue and still destroy value if project and service costs exceed revenue. Gross-margin progression is therefore more important than headline sales growth alone.

Repeated dilution

FCEL has repeatedly relied on equity financing. The fully exercised July 2026 transaction added 12.32 million planned shares and approximately $245.4 million of estimated net proceeds. It was rational from a treasury perspective but materially changed the supply-demand balance and may reduce per-share upside unless the capital generates profitable growth.

Capital intensity

Manufacturing expansion, project development and owned-generation assets require substantial capital before revenue is recognized. Delays can lock cash into projects for long periods.

Technology and service obligations

Fuel-cell stacks degrade and require replacement. Reliability problems or underestimated service costs can pressure margins and cash flow.

Natural-gas and policy exposure

Many carbonate systems rely on natural gas. Project economics can be affected by fuel prices, emissions rules, tax credits and changing definitions of clean or low-carbon power.

Competition

Bloom Energy has stronger commercial momentum in data centers, while turbines, nuclear, batteries and renewable-plus-storage systems compete for the same customer budgets.

The block below is a snapshot of the Stocktwits flow, with its date. These are opinions of retail traders and non-professional investors, not analyst research, and they measure attention and how one-sided positioning has become rather than anything about the business.

20 Merlintrader bottom line

FuelCell Energy has moved beyond being merely a legacy hydrogen-market speculation. The company has a differentiated carbonate platform, a plausible role in data-center power, meaningful Korean exposure, a strategically important carbon-capture partnership and, after the Siemens MOU, a stronger electrical-integration story for large distributed-energy projects.

The financing position is also materially stronger than the old $225 million headline suggests. Full exercise of the underwriters’ option lifted the planned issuance to 12.32 million shares and estimated net proceeds to approximately $245.4 million, while the first EXIM tranche supplied approximately $22 million net for the Korean equipment program. That capital lowers near-term financing pressure but raises the share-count hurdle and does not convert frameworks into revenue.

The next phase will be determined by execution. FCEL must convert the first 30 MW Fit Energy commitment, turn the Siemens framework into a named and funded deployment, complete EXIM-supported Korean deliveries, demonstrate manufacturing discipline and show that revenue growth can eventually produce positive gross margins. Until those milestones appear in filings, the stock remains a high-risk thematic vehicle rather than a derisked infrastructure compounder.

Technically, the July decline has produced an early rebound but not a confirmed base. The immediate chart question is whether $20–$21 can hold, whether FCEL can clear approximately $22.20–$23.00 and whether a later pullback forms a higher low above the $16.50–$17.30 swing-low zone. Fundamentally, the question is whether partner announcements, deposits and financing become shipped equipment, backlog and recognized revenue. Those two repair processes—technical and operational—will determine the next durable phase of the FCEL story.