Stranded Gas
Stranded gas is the fastest path to on-site power.
Productive natural gas with no pipeline to market — converted to electricity at the wellhead, at the point of demand, for the buyers racing to secure power: AI compute.
Molecule to megawatt, at the wellhead.
Access
Tap stranded gas at the wellhead in basins the grid can't economically reach.
Marcellus · Permian · LouisianaDeploy
Modular, emissions-permitted generation on-site — recycled engines rebuilt to spec.
With Waukesha-PierceConvert
Combust gas to electricity at the source. Initial PA permit targets up to 15 MW.
Application in progressOfftake
Sell power to co-located data centers, compute, and miners — or under long-term PPAs.
Distributed GenerationElectricity, not chips.
Generation sits at the molecule.
Application in progress.
Why gas strands in the first place
Stranded gas is not a geology problem. The molecules are there, the wells produce, and the resource is commercially useful the moment it reaches a buyer. What is missing is the pipe. A well drilled ten miles from a gathering system, or connected to a system already running at capacity, produces a fuel with no route to market — and the cost of building that route is frequently larger than the value of the gas it would carry.
Geography, not geology
Midstream capacity is built toward volume, and volume follows the largest fields. Wells at the edge of a basin, on older leases, or behind constrained takeaway are last in line for connection and first to be curtailed when a system fills. Their production is physically identical to gas trading at the hub; the difference is entirely in the interconnect.
The economics of a missing pipeline
Gathering line construction is priced per mile and recovered over years of throughput. For a small or declining position, the arithmetic rarely closes: the capital required to reach the system exceeds what the connected volumes would return. Operators are left holding a productive asset with a negative delivered value.
What happens by default
Where the gas cannot be sold and cannot be stored economically, the common outcomes are curtailment, venting, or flaring. Satellite observation of U.S. flaring identifies the Permian as the leading basin by flared volume, with the Bakken among the largest [11]. In the most constrained hubs, prices have gone below zero — Waha averaged near $0/MMBtu across 2024 and traded negative on roughly 42% of days [9].
The idled-well overhang
Alongside stranded production sits a large inventory of wells shut in because their standalone economics no longer justify an operator's attention. These are not dry holes. Many carry remaining reserves that a lower-cost, on-site conversion route could make useful again, which is why PowerLink's acquisition activity has focused on undercapitalized field positions rather than premium acreage.
Turning a molecule into a megawatt at the source
PowerLink's approach inverts the usual sequence. Instead of moving gas to where power is generated, generation is moved to where the gas already is. The pipeline is replaced by a wire, or by a load that sits on the pad. Capabilities described below reflect the platform as designed; what is deployed varies by site and by stage.
Generation sized to the resource
Wellhead volumes support generation measured in single-digit to low-double-digit megawatts, not the hundreds of megawatts a utility-scale plant assumes. Modular units let capacity be matched to what a pad actually produces, added as volumes justify it, and relocated when a field declines — an operating profile a fixed central plant cannot offer.
Rebuilt reciprocating engines
PowerLink's stranded-gas generation is built around recycled reciprocating engines rebuilt to specification with Waukesha-Pierce. Rebuilt units shorten procurement against a market where standard large power transformers alone were averaging roughly 128-week lead times as of Q2 2025 [2], and they tolerate the variable, wet, and inconsistently treated gas that wellhead service produces.
Fuel conditioning and treatment
Wellhead gas is not pipeline gas. Liquids, water, and variable heating value all have to be handled before combustion, and the conditioning package is part of the deployed system rather than an afterthought. Treatment requirements are a site-level engineering question and are assessed per pad.
Permitting and emissions
On-site generation is an emissions-permitted activity, and permitting is the gating item on schedule far more often than equipment is. PowerLink's initial Pennsylvania application targets generation of up to 15 MW and is in progress. No assurance exists that this or any other application will be granted, or granted on the timeline sought.
Who buys power at the wellhead
Behind-the-meter generation is only worth building if there is a buyer standing next to it. The buyers exist because the alternative route to power has become slow and expensive.
Co-located compute
AI training and inference are concentrated, high-density loads that can be sited where power is rather than where fiber is cheapest. For an operator, the binding constraint is interconnection: U.S. projects reaching commercial operation in 2024 spent a median of roughly 55 months in the queue [1]. On-site generation is designed to shorten that path by not joining the queue.
Load balancing
Bitcoin mining load is interruptible in a way most industrial load is not, which makes it useful as a flexible offtaker while firmer contracts are developed. It is a bridge use for otherwise idle capacity, not the objective of the platform.
Long-term power purchase agreements
Where a site supports it, the intended structure is a term PPA with an industrial or data-center counterparty. Capacity value has repriced sharply in constrained markets — PJM's 2025/2026 base residual auction cleared at $269.92/MW-day against $28.92 the prior delivery year, about +833% [3] — which is the market signal a decentralized generator is responding to.
Why the buyer is willing
Roughly 2,600 GW sat in U.S. interconnection queues at year-end 2023, and about 2,290 GW at year-end 2024 [4]. A buyer facing that queue, and the transmission timelines that commonly run 10–15 years [13], is evaluating on-site generation against waiting — not against a cheaper grid connection that is actually available.
Two closed Louisiana acquisitions expand controlled inventory and field-level execution leverage.
Recent closings move 57 wells and roughly 2,000 acres of Louisiana mineral rights into PowerLink — adding near-term workover opportunities and a larger base for disciplined field development.
Across two closed acquisitions, May and June 2026.
Louisiana mineral rights across both fields.
Identified at close, plus new drill targets.
Llanada Operating LLC
100% Louisiana acquisition with greater operating control.
Approximately 800 mineral acres with 16 wells, including saltwater disposal infrastructure. One well is currently producing, roughly eight wells are strong workover candidates, and the field can support one to two new production wells.
Lake St. John Field
41-well portfolio with low-rate production and development upside.
Approximately 1,200 mineral acres acquired from a small regional operator. Multiple wells are producing today, with upside from targeted workovers, improved field operations, and new drill targets.
PowerLink is converting undercapitalized field positions into a larger, more controllable operating footprint with identifiable reactivation, workover, and new-drill inventory.