DOE Picked 31 SPARK Projects and Asked Recipients to Put Up $3.35 Billion. Only 1,500 of the 22,500 Miles Involve New Wire.

On September 24, the Department of Energy's Office of Electricity announced it intends to fund 31 grid-improvement projects across 26 states under SPARK — Speed to Power through Accelerated Reconductoring and other Key Advanced Transmission Technology Upgrades. The headline number is $5.25 billion in total project value. The federal share of that is $1.9 billion. The other $3.35 billion is money the recipients themselves have to find.

That ratio is the single most important fact in the announcement, and it is the one that got the least attention. Recipient cost share is 64 percent of the total package. For a program funded out of the Infrastructure Investment and Jobs Act, that is an unusually thin federal slice — and it tells you exactly what kind of applicant DOE selected for.

We covered the SPARK notice of funding opportunity when it dropped in March — see $1.9 Billion to Rewire America for the three topic areas and the concept-paper timeline, and Granted News for the original announcement. This is the other end of that competition: who won, what the award structure reveals about the scoring, and what the next 120 days actually determine.

Selections Are Not Awards, and the Gap Is Where Projects Die

DOE's language was careful. It "intends to help fund" these projects. Formal financial-assistance agreements are expected between October 2026 and January 2027, and individual project scopes and dollar figures remain subject to negotiation. DOE has not published a schedule for converting the 31 selections into executed agreements.

Grant seekers who have never been through a DOE negotiation phase tend to treat a selection announcement as a win. It is closer to an invitation to a due-diligence process that can last four months and routinely reduces the federal share, trims the scope, or restructures the milestone schedule. The National Environmental Policy Act review, the cost-share verification, the Buy America determination, and the go/no-go milestone structure all get settled here — not in the application.

The practical consequence is asymmetric. A large investor-owned utility that gets its federal number trimmed by 15 percent absorbs it out of the capital plan. A small recipient that built its financing around a specific federal percentage does not have that option.

The Small-Utility Provision Did Real Work

Kit Carson Electric Cooperative, which serves just under 30,000 member-owners across Taos, Colfax, and Rio Arriba counties in northern New Mexico, was selected for up to $38.1 million toward a $50.8 million rebuild of the 69-kilovolt corridor between Questa and Taos. That is 25.5 miles of line, and the cooperative's share is $12.7 million — a federal share around 75 percent.

Compare that to PPL Electric Utilities, selected for up to $71.5 million for the Montour Grid Resilience and Advanced Reconductoring Project: 29.3 miles of 230-kilovolt corridor in Pennsylvania. Similar mileage, a much higher voltage class, and a much larger utility. CenterPoint Energy was selected for $50 million.

The relevant mechanism is the cost-share reduction SPARK offered small utilities — 25 percent instead of the standard 50 percent. Kit Carson and Farmers' Electric both entered under that classification. A cooperative with 30,000 members cannot plausibly raise a 50 percent match on a $50 million transmission project; at 25 percent it can. The provision is the difference between the co-op being in the competition and not being in it.

This is worth internalizing as a general pattern. When a federal program publishes a reduced match for a defined class of applicant, that reduction is not a courtesy — it is the program office deliberately widening the eligible pool, and it usually means the program wants geographic or institutional diversity in the portfolio badly enough to pay for it. If you qualify for a reduced-match tier and you are not applying, you are leaving the most favorable terms in the program on the table.

There is a corresponding risk, and Kit Carson illustrates it. If DOE trims the 75 percent federal share during the October negotiation, a cooperative with that member base has very limited capacity to backfill. Applicants who won on a reduced-match tier should enter negotiation with a documented floor — the federal percentage below which the project is not financeable — and say so early rather than discovering it in month three.

Read the Mileage Split Before You Read the Capacity Number

DOE projects that recipients will reconductor or rebuild more than 1,500 miles of existing transmission line and deploy grid-enhancing technologies across nearly 21,000 additional miles, for a combined increase of more than 23 gigawatts of capacity serving roughly 100 million Americans.

Those two mileage figures are not the same kind of thing, and conflating them produces a badly wrong picture of what SPARK bought.

Reconductoring replaces the physical conductor. It adds firm, weather-independent capacity to a line that keeps its existing towers and right-of-way. It is expensive per mile and it is durable.

Grid-enhancing technologies — dynamic line ratings, power-flow controllers, topology optimization — do not add a conductor. They release headroom that already exists on the line by replacing conservative fixed seasonal ratings with real-time measurement. On a cool, windy day, a line can safely carry substantially more than its static rating implies. GETs let the operator use that margin. On a hot, still afternoon, the same technology may show less headroom than the static rating assumed.

So the 23 gigawatts is a blend of firm capacity from 1,500 miles of new wire and conditional capacity from 21,000 miles of better measurement. Both are legitimate and both are cheap relative to building new corridors. But a state energy office or a load-serving entity planning around SPARK-enabled capacity needs to know which category applies to the specific lines it cares about, because only one of them shows up reliably during a summer peak.

For applicants to future rounds, the mileage ratio — roughly 14 miles of GETs for every mile of reconductoring — is a scoring signal. DOE bought a great deal of low-cost, fast-deploying measurement capability alongside a smaller volume of heavy construction. A proposal that pairs a targeted reconductoring segment with a broad GETs deployment across the surrounding network matches the portfolio DOE actually assembled.

SPARK Reallocates Rather Than Adds

SPARK is the third round of GRIP, the Grid Resilience and Innovation Partnerships program, which has roughly $10.5 billion allocated over five years. The $1.9 billion in SPARK is money GRIP already held; it is not a new appropriation.

That matters for anyone tracking the program's future. A reallocation means the pool competing for the remainder of GRIP just shrank by $1.9 billion, and it means SPARK's selections were made against a budget that other GRIP priorities were also claiming. Applicants waiting for "the next GRIP round" should size their expectations against what is left in the five-year allocation, not against the headline program total.

It also means the administration made a deliberate technology bet with existing money. Reconductoring and GETs benefit generation of every fuel type equally — they relieve the interconnection queue for gas plants and wind farms without distinction. That fuel-neutrality is why a program of this size survived a budget environment that has been hostile to a great deal of IIJA energy spending.

What to Do in the Next Ninety Days

If you were selected. Assign one person to own the negotiation file and treat it as a live project, not a formality. Get your cost-share documentation — board resolutions, rate-case posture, third-party commitments — in a form DOE can verify without a second request. Identify your federal-share floor now. Confirm which portions of your scope are reconductoring and which are GETs, because the go/no-go milestones will likely differ.

If you were not selected. Request your debrief. DOE's Office of Electricity will generally discuss scoring, and the most valuable information in a debrief is not the score but which topic area the reviewers thought you belonged in. Applicants frequently submit a resilience project into a capacity topic area and lose on relevance rather than merit.

If you are a potential subrecipient or vendor. The 31 selections are about to become 31 procurements. Advanced-conductor manufacturers, dynamic-line-rating vendors, engineering firms, and workforce-training providers should be identifying which of the 26 states contain selected projects in their service footprint and making contact during the negotiation window, not after agreements are signed. Build America, Buy America requirements apply to this funding, which advantages domestic suppliers who can produce a compliance determination on request — and disadvantages anyone who cannot.

If you are a small utility that did not apply. Read the small-utility cost-share provision in the SPARK notice and check whether you would have qualified. A 25 percent match on a 69-kilovolt rebuild is among the most favorable federal terms available to a rural cooperative or municipal utility right now, and the class of applicant that provision exists to serve is systematically underrepresented in federal energy competitions — usually because nobody on staff has time to read a 90-page notice of funding opportunity closely enough to find the carve-out.

The selections announced on September 24 are real, but the numbers attached to them are provisional. The projects that get built are the ones whose sponsors treat October through January as the decisive phase.