Home › What we have to decide › Route: how do we add capacity?
Detail route 1

How do we increase generating capacity?

This route stems from one of the decisions on the general map: “How do we cover the nights while new generation arrives?”. Here we go one level deeper: why today's generation costs so much, what capacity is on the way, and what options there are to cover the shortfall.
Already covered on the general map and not repeated here: the age of the plants and the lack of reserve (Knot 2), imported fuel (Knot 1), rooftop solar and its credit (Knot 5 and decision 5), and the pace of federal grid funds (Knot 6).
700–850 MWis the generation shortfall that must be covered, according to LUMA (Sept. 2026).
~2,300 MWof new capacity on the way, between 2027 and 2030.
3,000 MWof firm capacity out for bid since October 2025.
Prepared from the work of the Chamber of Commerce's Energy and Water Committee. · Draft for discussion
1 · WHY TODAY'S GENERATION COSTS SO MUCH

Plants that burn more, with more expensive fuel

Old baseload plants

A steam unit around 40 years old burns on the order of 11,500 Btu per kWh; a modern combined cycle, about 6,800. That is close to 70% more fuel per kWh (typical industry values). The measured efficiency of Genera's steam plants, 28.7%, is equivalent to about 11,900 Btu per kWh.

Emergency turbines (peakers)

They cover the shortfall and the evening rise in demand. They burn the most fuel per kWh, and they burn diesel.

Expensive fuel and breakdowns

For October–December 2026, the fuel charge rose 5.95¢ per kWh: uncollected fuel from previous months, baseload units out of service and more diesel, gas that did not arrive, and forced outages.

Sources: typical values by technology (Committee estimate); EIA-923 via the Puerto Rico Energy Monitor; PREB, resolution of Sept. 30, 2026.
2 · THE AFTERNOON WITHOUT SUN

Midday sun and the evening ramp

A little-known mechanism. This is an illustration with orders of magnitude, not actual dispatch data.
Midday

Rooftop solar lowers the demand the grid sees by about 900 MW. Baseload plants are large and inflexible: they cannot go below a certain minimum or shut down and restart within hours, so fewer of them are scheduled.

Sunset

The sun goes down just as consumption rises: the demand the grid sees grows by about 900 MW in 3 hours. The baseload plants that were shut down do not come back in time; peakers burning diesel cover it.

Result

More peaker hours in the evening and a higher fuel charge. Without batteries or flexible plants, part of the savings from solar is lost in the evening.

How the kWh is producedFuel cost per MWh (order of magnitude)
Coal~$50
Gas in combined cycle~$90
Steam with bunker fuel~$140
Diesel peaker~$230
Illustrative: profile with the order of magnitude of demand and 1,666 MW of rooftop solar (LUMA, Sept. 2026). Costs: Committee estimate.
3 · WHAT IS ON THE WAY

About 2,300 MW, between 2027 and 2030

The shortfall is today's; the new capacity arrives in one to four years. Only approved or ongoing projects are counted.
ProjectMWWhenStatus
Utility-scale batteries (includes 430 MW from Genera at six sites)~995by Dec. 2027Underway
Utility-scale solar~6002027–2028Underway, with delays in procurement
Energiza: gas combined cycle in San Juan, the first new power plant in decades528Dec. 2030 (guaranteed date)Underway
New Genera peakers~244Late 2027Underway
Temporary generation: 400 MW at Aguirre (contract canceled in August 2026) and 200 MW of floating gas generation at Costa Sur (not approved)600—Not counted
Hostos: unsolicited proposal to the Public-Private Partnerships Authority645–7002031 (proposed)Not counted: no contract
Firm capacity procurement, open to gas, batteries and renewables with batteries (includes up to 600 MW at Costa Sur)3,000—Out for bid since Oct. 2025
The opposite risk, overbuilding, is already covered in decision 3 of the general map.
Sources: PREB (Oct. 2025; Energiza: joint motion of Sept. 8, 2026, NEPR-AP-2025-0003); PRFAA (Feb. 2026); New Fortress Energy (Jan. 2025); Financial Oversight and Management Board (June 2026); Mintz (2026); IEEFA; status of temporary generation and Hostos according to the Committee (Nov. 2026) and the press.
4 · THE OPTIONS TO COVER THE SHORTFALL

Speed, firmness or no fuel: each one costs something

Temporary generation (decision 3), rooftop solar (decision 5) and the wires network (Knot 2 and decision 1) already have their pros and cons on the general map. Here are the other three.

New peakers

~244 MW · Genera · gas or diesel
For
  • They start in minutes: they cover the evening ramp.
  • More efficient than current peakers.
Against
  • They only cover peak hours.
  • Expensive if they burn diesel instead of gas.
  • They arrive in late 2027.

Permanent gas

Energiza 528 MW + whatever the open procurement awards to gas
For
  • Firm power that replaces 40-year-old plants.
  • Gas is cheaper than bunker fuel and diesel.
Against
  • Capital, long contracts and years of waiting (2030 or later).
  • Tied to the price of gas and to few suppliers.
  • Risk of overbuilding (IEEFA).

Utility-scale solar and batteries

~600 MW solar + ~995 MW batteries
For
  • No fuel cost.
  • Batteries carry the sun into the evening and reduce peaker use.
Against
  • Years of delay in procurement.
  • They require more grid investment (LUMA estimates $900 million to $2.6 billion in transmission upgrades; part of them is needed on any path).
Sources: the Chamber's Energy and Water Committee; PREB; PR100 (DOE/NREL); LUMA IRP (Mar. 2026); IEEFA.
SOURCES FOR THIS ROUTE
Figures marked as estimates or illustrative are calculations with published data, not official figures. · Puerto Rico Chamber of Commerce · Draft for discussion