US Interconnection Queue 2026: Solar and Storage Wait Times

What Berkeley Lab Counted

Berkeley Lab's Queued Up report is one of the most cited yearly counts of generators waiting to connect to the US transmission grid. The 2026 edition, published in June, covers data through the end of 2025 from seven grid operators and about 50 utilities, which together represent roughly 98% of installed US generating capacity.

At the end of 2025, about 8,200 projects were actively seeking a connection. Together they represent 2,061 GW, made up of 1,312 GW of generation and 749 GW of storage. That is down 10% from 2,290 GW a year earlier, and it is still larger than the installed US fleet of about 1,374 GW.

Solar includes the solar side of hybrid plants, and storage includes estimated battery capacity at hybrids where the operator did not report it. About 99% of the storage is batteries.

The drop does not mean projects were built. More than 600 GW of new requests came in during 2025, but an estimated 750 GW or more was withdrawn, so the queue shrank. Of the 2,290 GW that was active a year earlier, only 53 GW reached operation in 2025, a throughput of about 2%. Berkeley Lab says reforms, market conditions, transmission constraints, permitting barriers and the looming end of some federal tax credits may all be contributing.

Where the Queue Sits

Texas alone holds 21% of active solar and 22% of active storage. California has 15% of storage and 8% of solar. Solar requests exist in every state, but most of the capacity is in the Southwest.

Some of the small numbers come from pauses rather than a lack of interest. ISO-NE and NYISO accepted no new requests in 2025 while they moved to cluster studies, and CAISO took none in 2024 or 2025. MISO capped its total queue size in 2025, and PJM paused review of new requests until 2026 while it worked through its backlog.

How Long Projects Wait

The median project that reached operation in 2025 took 61 months, a little over five years, from interconnection request to commercial operation. The same figure was 36 months for projects built in 2015 and 22 months for projects built in 2008. Berkeley Lab splits the wait into two parts. From request to a signed interconnection agreement, the median in 2025 was 45 months. From the agreement to operation, it was 31 months, up from 21 in 2008. Berkeley Lab says the second stretch is largely driven by things outside the queue, such as procurement, offtake, permitting and transmission work.

Size matters more than any other factor.

The step change sits at 20 MW, the line between FERC's small and large generator procedures. Projects under 5 MW typically signed their interconnection agreement in about 12 months, those from 5 to 20 MW in about 15 months, and larger projects in roughly 30 months.

Region and service type matter too. ERCOT, which runs a connect and manage model outside FERC's jurisdiction, averaged 24 months from request to agreement. Across the wider sample, energy-only service, which lets a plant connect and deliver on an as-available basis, averaged about 37 months, and network service, which makes a plant deliverable and eligible to count toward resource adequacy, about 41 months. Outside ERCOT, 81% of active capacity asked for network service.

These durations describe projects that were built, so they leave out the ones that gave up. Several regions have small samples for 2025.

How Many Projects Actually Get Built

Of the capacity that requested interconnection from 2000 to 2020, 13% had come online by the end of 2025. By number of projects the figure is 19%. About 75% of that capacity was withdrawn and 10% was still active. By capacity, only ERCOT completes more than 20% of what it receives, at 24%, followed by ISO-NE at 18%. CAISO and the non-ISO West are the lowest at 8% and 10%.

A signed agreement is not a guarantee either. Of the capacity that signed agreements from 2000 to 2022, 41% had withdrawn by the end of 2025 and 43% was operating. In the roughly half of 2025 withdrawals where the study phase is known, most left during the system impact or cluster study, with fewer late-stage withdrawals.

Hybrids, Fast Tracks and the 540 GW With Agreements

Hybrid plants are now a large share of the queue. They make up 49% of active solar (376 GW) and 48% of active storage (358 GW). In CAISO, 91% of active solar is in a hybrid, and in the non-ISO West it is 85%. In PJM and MISO, it is about 20%.

Roughly 540 GW already has a draft or signed interconnection agreement but has not reached operation. That includes 256 GW of solar, 161 GW of storage, 76 GW of wind and 45 GW of gas. This is the nearest-term pipeline, though the withdrawal rate after signing says not all of it will arrive.

A new set of fast tracks aims at resource adequacy: MISO and SPP's expedited resource addition studies and PJM's reliability resource initiative. They promise timelines as short as three months from request to agreement. MISO has 29 GW in its fast track, SPP 15 GW and PJM 8 GW, and gas dominates all three.

What Each Operator Is Doing in 2026

Most grid operators are moving to first-ready, first-served cluster studies under FERC Order 2023, with higher deposits and stricter readiness rules. Berkeley Lab says the effect will not show up in its data for some time, because the report predates most of the changes. For the rules behind them, see our guide to FERC interconnection queue reform.

Projects on the Distribution Grid

Queued Up covers transmission-connected generators only. Most rooftop and many commercial systems connect to a utility's distribution system instead, and Berkeley Lab published a separate study of those in September 2026. It covers more than 2.7 million applications to 188 utilities in 24 states, for projects up to 50 MW. The data leans heavily on California, New York, Arizona and New Jersey, and the timeline figures lean on Arizona and New York. The timeline here is application to permission to operate, which includes construction and inspection, so it is longer than the utility's study alone.

Small systems are quick. Solar under 30 kW had a median of 2 months to permission to operate from 2020 to 2025, and storage and hybrid projects overall had medians of 5 and 4 months. The picture changes with size. Solar from 1 to 5 MW had a median of 45 months, and storage from 1 to 5 MW had a median of 26 months. Berkeley Lab found that only 2.4% of 1 to 5 MW projects reached permission within two years, and that about 90% of 1 to 5 MW applications from 2020 to 2025 were still in process. Part of that is recent applications that have not had time to finish, so the 45 months should be read as an early signal, not a settled figure.

Median months from application to permission to operate. Over 1 MW figures rest on the few projects that reached permission, so most large projects in these states were still in process.

Upgrade costs are poorly documented. Berkeley Lab found usable cost data in only four states, Colorado, Connecticut, New York and Rhode Island, and New York is over 99% of it, with most New York projects paying nothing because of cost sharing. Elsewhere the numbers are higher: the median upgrade cost across all project sizes was $281 per kW in Connecticut, $173 in Rhode Island and $87 in Colorado, and the average was highest for solar between 5 and 10 MW at $383 per kW.

What This Means for Solar and Storage EPCs

Five points come out of the numbers.

First, a project over 200 MW that files today faces a 59 month median, which points to roughly 2031, though the median describes projects that were already built. An EPC working from a developer's queue position is looking at a multi-year horizon, and the construction window is the short part.

Second, the withdrawal numbers mean a queue position is not a project. With 41% of agreement capacity withdrawn and most known withdrawals coming during the study phase, it is worth asking where a project stands before committing engineering time to it.

Third, size changes the odds on both grids. Projects under 20 MW reach a transmission interconnection agreement in roughly half the time of larger ones, and distribution-connected systems under 50 kW typically finish within months.

Fourth, the 1 to 5 MW band is where timelines stretch most on the distribution side. Many community and commercial projects sit in it, and the 45 and 26 month medians for solar and storage, even with their caveats, argue for confirming the utility's study stage and construction window before fixing a delivery date. Our guide to US solar permitting and interconnection covers the utility steps.

Fifth, hybrids are close to half of the transmission pipeline, so storage design belongs in the first application, not as a later add-on.

Where Design Software Fits

A queue position starts with a defined project: a site, a capacity and a single line diagram. For developers testing several sites before they file, design software like Reslink supports ground-mount feasibility up to 1 GW and produces the layout, bill of materials and single line diagram from one set of site inputs. It does not predict study outcomes or queue timing.

See the full workflow → Book a demo

Frequently Asked Questions

Q1. How big is the US interconnection queue?

About 2,061 GW of generation and storage was actively seeking transmission interconnection at the end of 2025, across roughly 8,200 projects. That is down 10% on 2024 but still above the size of the installed US fleet.

Q2. How long does a solar project wait for interconnection?

On the transmission grid, projects built in 2025 took a median of 61 months from request to operation, and projects over 20 MW take about twice as long as smaller ones to reach an interconnection agreement. On the distribution grid, solar under 30 kW had a median of 2 months and solar from 1 to 5 MW had 45 months.

Q3. What share of queued projects get built?

By capacity, 13% of what was requested from 2000 to 2020 had come online by the end of 2025, and about 75% was withdrawn. Even after signing an agreement, 41% of capacity from 2000 to 2022 later withdrew.

Q4. Which region has the most solar and storage waiting?

The non-ISO West leads in total queue capacity at 567 GW, followed by ERCOT at 408 GW and MISO at 382 GW. Texas holds about a fifth of all active solar and storage requests.

Q5. Are rooftop and small commercial systems in the queue numbers?

No. Queued Up covers transmission-connected projects only. Distribution-connected and behind-the-meter systems follow utility processes, covered by a separate Berkeley Lab study, and large loads such as data centers have their own queues.

Q6. Is the queue getting shorter because reform is working?

Not clearly. Berkeley Lab lists reforms among several possible reasons for the drop, alongside market conditions, transmission constraints, permitting and tax credit timing, and says it is too early to measure their effect.

Q7. What is the difference between energy-only and network interconnection service?

Energy-only service lets a plant connect and deliver power as the grid allows. Network service makes the plant deliverable and lets it count toward resource adequacy, which is why most active capacity outside ERCOT asks for it, at a longer average study time.

Final takeaway

The US queue is shrinking, but mostly through withdrawals, and a project that gets through still waits about five years on the transmission grid. For solar and storage EPCs, the practical reading is that size sets the timeline in both datasets, hybrids are close to half of the pipeline, and a queue position is only the start of a project.

You May Also Like

• FERC Interconnection Queue Reform 2026: EPC Guide
• US Solar Permitting & Interconnection: 2026 EPC Guide
• US Commercial Solar and Battery Storage in 2026

Sources

• Lawrence Berkeley National Laboratory, Queued Up: 2026 Edition, Characteristics of Power Plants Seeking Transmission Interconnection as of the End of 2025, with Interconnection.fyi (June 2026)
• Lawrence Berkeley National Laboratory, Interconnection Timelines and Costs for Distributed Energy Projects in the United States (2000-2025) (September 2026)
• Lawrence Berkeley National Laboratory, Queued Up data and publication page, including the project-level dataset and interactive tools (2026)

This guide reflects Berkeley Lab's Queued Up 2026 edition, which covers data through the end of 2025, and its September 2026 distributed energy study. Operator rules and cluster dates have moved since, so check each grid operator's site before relying on a date.