AI Data Centres Hunt Used Jet Engines as New Gas Turbines Face Six-Year Delivery Timelines

AI Data Centres Face Critical Power Infrastructure Bottleneck

The explosive growth of AI data centres has triggered an unprecedented hunt for power generation equipment, with companies scouring global markets for used jet engines, second-hand turbines, and mothballed machines. New gas turbine delivery timelines now stretch six years on average in the United States, with discussions firmly focused on the 2030 to 2031 window, according to Wood Mackenzie1

1

. This timeline creates a fundamental mismatch with an industry racing to bring massive computing facilities online far sooner.

Global gas turbine orders have surpassed 100 gigawatts, dramatically exceeding estimated annual manufacturing capacity of just 60 to 70 gigawatts2

2

. Industry giants Siemens Energy, GE Vernova, and Mitsubishi Heavy Industries—who dominate the market—all confirm that the AI boom has driven orders well beyond their production capacity1

1

. While global manufacturing capacity is expected to grow to around 87 gigawatts by 2028, this still falls drastically short of the projected 129 gigawatts in turbine power demand1

1

.

The Secondary Market Heats Up Amid AI Power Rush

"That secondary market is very, very hot right now," said Ana Amicarella, CEO of EthosEnergy, an equipment services provider. "We also have seen an escalation on the prices of what people are demanding"1

1

. The emergence of this robust secondary market attests to the profound impact data centre power consumption has had across industries, transforming what was once a niche market into a critical supply channel.

Companies are pursuing unconventional power solutions through multiple channels. AI infrastructure firm Crusoe ordered 13 aeroderivative gas turbines from US-based ProEnergy in April—turbines based on jet engines made by GE Aerospace—with a combined capacity of 650 megawatts for deployment at US AI data centre projects1

1

. Darayus Pardivala, who leads the US aftermarket business of Sulzer, noted receiving requests every few weeks asking to find used gas turbines for data centres1

1

.

Cost Advantages Drive Adoption of Refurbished Turbines

The financial incentive for second-hand turbines is substantial. Well-maintained used units sell for approximately $1.5 million to $1.7 million per megawatt, while refurbished turbines command about $1.3 million per megawatt1

1

. This represents a significant discount compared to new turbines, which cost roughly $2 million to $2.5 million per megawatt according to Aurora Energy Research1

1

.

For smaller units, the discount can be even more dramatic. Two 14.5-megawatt turbines made by Solar Turbines recently sold for $12.5 million, or around $430,000 per megawatt, according to Dan Poole, general manager at ReflowX. New equivalents would have cost about $35 million and required a five-year delivery timeline1

1

.

Growing Power Demand Reshapes Energy Markets

The scale of AI data centre power demand is reshaping traditional energy markets. US data centres alone account for nearly 17 gigawatts of demand this year, a figure that investment bank Jefferies estimates could more than triple by 20281

1

. The International Energy Agency projects global data centre electricity consumption will nearly double from 485 terawatt-hours in 2025 to 950 in 2030, with AI-focused facilities expected to triple their consumption over the same period2

2

.

Data centres and their operators have emerged as major customers in a gas turbine market that almost exclusively served traditional utilities until recently. The global turbine supply market is worth approximately $25 billion annually, though reliable estimates of the secondary market's contribution remain elusive1

1

.

Challenges and Limitations of Alternative Solutions

While the secondary market offers faster deployment, it cannot fully solve the power infrastructure crisis. "The market is already constrained, with limited excess supply, so it is unlikely to offset a meaningful share of the delays," said Noor Qureshi, an associate at Enverus Intelligence Research1

1

. Beyond turbines, companies face additional obstacles including air permits, grid access, and environmental permitting requirements.

Refurbished turbines and converted jet engines carry inherent risks. Overhauled equipment can have unclear maintenance histories, limited remaining operational life, and weak long-term servicing support. Converted jet engines raise questions about their suitability for continuous operation as power generators1

1

. The Environmental Protection Agency subjects stationary turbines and engines used by data centres to federal air-pollution standards, adding regulatory complexity2

2

.

Power-market research firm Cleanview identified approximately 90 gigawatts of proposed on-site generation at US data centres in mid-2026 analysis, yet only about 2 gigawatts was actually operating at that time2

2

. This massive gap between planned and operational capacity underscores the formidable challenges ahead for powering AI's expansion.