Power transformer lead times have doubled and generator step-up transformer waits have grown 260% since 2022, the direct result of hyperscale data center buildout colliding with a supply chain still recovering from COVID-era disruptions – and while Siemens and Hitachi are each committing roughly $1 billion to US manufacturing expansion, those new plants will not deliver a single unit for years. That gap between demand and deliverable supply is now the single largest schedule risk for every utility, developer and grid operator planning new capacity this decade, and it is already forcing some buyers to turn to refurbished equipment just to keep projects moving.
Why Transformer Lead Times Are Exploding – and Why the Fix Is Slow
The numbers reported this week capture how sharply the equipment market has tightened. Since 2022, power transformer lead times have risen about 100%, generator step-up (GSU) transformers – the large units that connect power plants to the transmission grid – have seen waits grow 260%, and gas turbine backlogs have expanded anywhere from 50% to 250% depending on the model and manufacturer. These are not incremental slips. They represent a structural shift in how far in advance buyers must commit capital and how much schedule risk they must absorb.
The demand side is well understood. Hyperscale data center construction, driven by AI compute demand, has added a concentrated block of new load in specific regions – northern Virginia, Texas, the Southeast – that utilities did not forecast even three years ago. When a single data center campus can draw hundreds of megawatts, and multiple campuses land in the same utility service territory, the transformer order book fills quickly. What makes this different from a typical demand spike is the simultaneity: every major utility in high-growth regions is ordering at the same time, and so are the independent developers building the gas plants, solar farms and storage projects meant to serve that load.
The supply side explains why the market cannot respond quickly. Transformer manufacturing is not a commodity assembly line. Each large unit is engineered to order, and the industry’s capacity is constrained by specialized skills – core steel cutting, coil winding, vacuum drying, high-voltage testing – that cannot be scaled up in months. The two critical material inputs, grain-oriented electrical steel and copper, face their own supply chain pressures. And the testing facilities required to validate a large transformer before shipment are themselves a bottleneck. A manufacturer can add floor space faster than it can add qualified test engineers.
That is why the announced investments matter but will not relieve pressure soon. Siemens and Hitachi are each committing on the order of $1 billion to US production. As a rough benchmark, a modern large-transformer plant typically costs in the low hundreds of millions of dollars to build, so $1 billion could fund two to four facilities depending on scope and site. But even under accelerated permitting and construction, new plants take roughly three to five years from announcement to first shipment. In the interim, the order book keeps growing.
The turn to refurbished equipment is a telling market signal. Utilities and developers that cannot wait two to three years for a new transformer are sourcing decommissioned units, having them reconditioned and re-certified, and putting them into service. This is functionally similar to what happened in the semiconductor industry during its own equipment shortage – a secondary market emerged for used tools. Refurbished transformers carry real risks: unknown remaining service life, limited or no manufacturer warranty, and the need for extensive testing before energization. But for a project with a contracted commercial operation date and penalty clauses attached, a refurbished unit that ships in months can be the difference between on-time and never.
The Equipment Crunch Is Compounding the Interconnection Bottleneck
The transformer shortage does not exist in isolation – it is the second chokepoint in a two-stage problem. The first is the interconnection queue. Across the US, the backlog of generation and storage projects waiting for grid studies and approval is on the order of 2,000 gigawatts, far more than the entire existing US generating fleet. Policy attention has focused on queue reform, faster study processes and FERC’s recent interconnection rule changes. But here is the compounding effect: even if queue reform accelerates approvals by years, every approved project still needs a GSU transformer to connect. A project can clear the queue and still face a 260% longer wait for the equipment that physically ties it to the grid.
That sequencing problem has a direct financial consequence. Utilities planning generation to serve new data center load are now making procurement decisions three to four years ahead of need, which forces them to commit capital before they have certainty on load growth, regulatory approval or financing. Developers who signed equipment purchase agreements early are in a stronger position, but lead-time inflation means many of those contracts were priced before the market tightened – or were never signed at all because manufacturers stopped quoting firm delivery dates. This is a capital allocation problem as much as a supply problem.
There is also a parallel worth drawing with the broader electrification push. The IRA-driven buildout of solar, storage and EV charging infrastructure was already straining transformer supply before the data center surge. The incremental data center demand has effectively crowded out some of that earlier demand – manufacturers allocate production to the largest orders and the highest prices, and hyperscalers with deep balance sheets can pay premiums or pre-pay for capacity. Smaller utilities and rural cooperatives, which historically ordered in smaller volumes, are the ones most likely to see their delivery dates slip further. If this trend holds, the equipment shortage becomes a distribution-level problem as much as a transmission-level one, because those smaller buyers serve the residential and commercial electrification load that is growing steadily but less visibly than data centers.
Who Feels the Transformer Squeeze Most
- Utility planners: Put transformer procurement on the critical path of every generation and load-serving project now, not at the engineering stage. A three-year lead time means orders placed this year only arrive in time for 2027-2028 commercial operation dates; any project targeting earlier COD needs a refurbished or secondary-market strategy already in motion.
- Generation and storage developers: Treat the GSU transformer as the gating item in your project schedule, ahead of turbine delivery and interconnection approval. Lock in equipment contracts with penalty clauses for schedule slippage, and model a refurbished-unit fallback scenario for any project with a hard COD deadline tied to a PPA or tax credit milestone.
- Grid operators and reliability coordinators: The equipment shortage will show up in planning studies as delayed generation additions and transformer replacement backlogs. Retirements of aging transformers that would normally be replaced on schedule will now be deferred or run with increased risk, so factor longer outage durations and higher forced-outage probabilities into reliability assessments.
- Investors and lenders: Manufacturers with announced capacity expansions – Siemens, Hitachi and others likely to follow – have multi-year revenue visibility, but the real opportunity may be in the refurbishment and testing ecosystem emerging to fill the gap. Also treat transformer lead times as a due-diligence item: any project with a near-term COD and no confirmed transformer supply carries schedule risk that should be priced into financing.
What to Watch in Grid Equipment Markets
- New plant announcements: Watch for additional manufacturers to announce US capacity expansions beyond Siemens and Hitachi. If total announced investment reaches $3-4 billion within the next year, that signals the market expects sustained demand; if it stalls at the current level, the shortage will persist well into the 2030s.
- GSU lead time peak: The 260% increase in GSU waits may still be climbing. Track quarterly lead-time data from major manufacturers – if GSU waits stabilize or begin to fall, that is the first sign the supply response is catching up; if they keep rising, expect more project delays and more refurbishment activity.
- Refurbished equipment market maturation: Watch whether dedicated refurbishment players emerge with standardized testing and warranty offerings. The first company to offer a warranted, certified refurbished transformer at scale will capture a significant share of the urgent-need market.
- Price behavior in secondary markets: If refurbished transformer prices begin to track new-unit prices closely, that indicates the shortage is structural rather than cyclical. Wide price spreads between new and refurbished units would suggest the market is still in the early, opportunistic phase.
Bottom Line
The transformer shortage is not a supply chain blip – it is a multi-year structural constraint that will shape which projects get built, when they reach commercial operation, and who captures the value of the AI-driven load boom. The $1 billion investments from Siemens and Hitachi are a necessary first step, but they will not close the gap before the end of the decade. The practical response for anyone planning grid infrastructure is to assume transformer lead times will not improve before 2027, plan procurement accordingly, and treat refurbished equipment as a legitimate – if imperfect – tool for keeping critical projects on schedule.
Read the full report at <a
Note: facts and figures attributed above to reflect that outlet's original reporting. Broader context, cross-sector connections, and forward-looking scenarios reflect independent analysis by our editorial team.
About this article: Drafted by Energy Ai with AI-assisted research and writing based on public reporting, then reviewed under our editorial process before publication.
Leave a Reply