How an electric grid works

Article summary

An electric grid is not a tank: it is generation, high-voltage transmission, distribution, and load/demand (customers), balanced at every instant.
Volt (voltage), watt (power at an instant), and watt-hour (energy accumulated over time) are different quantities — swapping one for another is the most common reading error.
"Capacity" appears in distinct senses: maximum generation, physical line limit, contingency margin, economic congestion, and connection queue — it is not a single number.
Congestion means power cannot reach a market within safe limits; in most cases this is a local economic problem, not synonymous with a national shortage of energy.

Grid layers

The EIA describes electricity delivery as a layered system: power plants generate, high-voltage transmission lines move that energy over long distances, substations and transformers bridge between stages, and distribution lines — at lower voltage — carry electricity to homes and businesses. [1] The DOE/CESER organizes the same chain into three segments — generation, transmission, and distribution — and notes that end users can also inject energy when they have distributed resources, such as residential solar. [6]

The DOE defines the “grid” as that interconnected group of lines and equipment that moves electric power at high voltage between points of supply and points of delivery, stepped down to lower voltage before reaching the customer — whose residential use is measured in kWh. [5] The FERC Reliability Primer sums up this interconnection as “essentially one large machine,” with three main functions (generation, transmission, distribution), and calls individual customers load: the parts operate in continuous electrical balance, with a real-time balance between demand and available resources. [8] In Brazil, the ONS describes the SIN as a hydro-thermal-wind system with hydro predominance, organized into four subsystems (South, Southeast/Midwest, Northeast, and most of the North), whose transmission network allows transferring energy between regions and taking advantage of hydrological diversity — a picture of generation and transmission, not of the distribution layer. [10] The EPE, in the transmission chapter of the PDE, defines the Rede Básica (Basic Network) as the facilities with voltage equal to or above 230 kV, whose function is to carry energy from power plants to major load centers, integrate the system for stability, and reduce bottlenecks between submarkets. [11] In the glossary of the Statistical Yearbook, the EPE also separates energy load — the volume of energy required from the generating system — from demand — the average power injected or required over an interval — a distinction that is easily lost in everyday conversation. [12]

Volt, watt, and watt-hour are not the same thing

The EIA draws a precise distinction: watt (W) measures power at an instant; watt-hour (Wh) measures energy used over time — “a watt-hour is the energy of one watt used for one hour.” The scale continues in kW, MW, and GW, and it is in kWh that utilities bill residential consumption. [2] The EIA’s own teaching example helps fix the idea: a 40 W light bulb turned on for 5 hours consumes 200 Wh, or 0,2 kWh — power multiplied by time equals energy. [2]

Voltage (volt) is a different quantity: the EIA’s material for children describes it as the “electrical pressure” on the lines, and transformers adjust that pressure — higher for long distances, lower and safer inside homes and businesses. [14] The DOE/NREL material explains why with basic physics: P = V × I, so for the same power, higher voltage means lower current; since losses scale with the square of the current, transmitting at the highest practical voltage reduces losses over long distances, and substations convert between voltages along the way. [13] The voltage ranges cited by the DOE/CESER primer — power plants generally between 5 and 34,5 kV, typical transmission at 115 to 765 kV in the US — are examples from that country, not universal constants of electrical engineering. [6]

Generation capacity ≠ generation

The EIA separates two ideas that lay discourse tends to merge: generation capacity is the maximum electrical output a generator can produce under specific conditions; generation is the amount of electricity actually produced over a period. The numerical bridge is simple — 1 MW of capacity operating at that level for 1 hour produces 1 MWh; at half capacity over the same period, it produces 0,5 MWh. [3] [4] Operators dispatch power plants so that the right amount of electricity is delivered at every instant and demand stays balanced — “how much a plant can generate” and “how much it generated in the month” are different readings of the same asset. [4]

It’s also worth noting that more electricity is generated than is sold, because part of the energy is lost as heat along transmission and distribution, in addition to use that occurs directly behind the meter. The EIA confirms this fact qualitatively; no aggregate percentage is used here, because that specific figure has not yet been verified in this cycle of the dossier. [4]

“Capacity” has several meanings

In grid jargon, “capacity” is not a single number — it’s a label that changes meaning depending on context. The DOE/NARUC transmission primer shows how this works in practice: physical properties and safety factors limit a line’s capacity, and operators deliberately leave unused margin to cover contingencies. The example cited is a line between Quebec and New England capable of about 2.000 MW, but operated near 1.400 MW for contingency reasons. [7] In other words: the same line has a physical capacity, an operated capacity with a safety margin, and a commercial capacity — and none of them is “how much energy exists in the country.”

There is also a fourth meaning, which this text only names without going into depth: the generation connection queue, i.e., projects awaiting interconnection studies and agreements before they can generate. The IEA describes grids as the “backbone” that delivers power to homes, businesses, and industry, and reports at least 3.000 GW of renewable projects in connection queues — of which about 1.500 GW already advanced — with the grid acting as the bottleneck: new grid assets typically take 5 to 15 years to plan, permit, and complete, against 1 to 5 years for renewable generation projects. [9] This generation queue — who enters it, what “advanced” means, why MW in the queue is not the same as MW ready to operate — is the subject of another text in this series, because it mixes processes and timelines that deserve their own space.

Congestion is not a national blackout

The DOE/NARUC transmission primer defines congestion as the situation where power cannot reach its market because the transmission system lacks sufficient capacity to carry it. When local generation still meets demand within reliability standards, this congestion is framed as an economic problem — not necessarily a reliability one — in areas described as “transmission constrained.” [7] The DOE’s National Transmission Needs Study reaches the same distinction from another angle: transmission congestion is the economic impact of operating within physical safe-flow limits; a thermal limit on a line or transformer can force power to be rerouted, with more expensive generation being dispatched while cheaper generation is curtailed elsewhere. The physical limit and the economic consequence are distinct things. [15]

This distinction matters because “lack of delivery capacity” at a specific point in the grid does not equate to “there is no energy generated in the country.” These are different questions: is generation missing at the right instant? Is generated power unable to reach where it’s needed? Or is the generation project still in a connection queue, awaiting studies? The sources in this dossier clearly separate the first question from the second; the third, about connection queues, is left for another text in this series.

Analogies that help (and the ones that mislead)

Two analogies appear in the official primers themselves and are useful when applied carefully. The first: the transmission system works “much like a highway system that enables a new commerce in power” — a good image for transfer capacity and energy trade between regions. [7] The second: “power flows over transmission lines much as water flows through a network of interconnected pipes” — if one path is saturated, the power seeks another physical route (loop flow), which may not match the contracted or scheduled path. [7]

The risk lies in pushing these analogies beyond what they explain. An “empty lane” on a highway doesn’t mean a blackout — it just means unused transfer capacity at that moment, which is normal and, in most cases, deliberate (the contingency margin mentioned above). Treating idle grid space as synonymous with an energy crisis is exactly the kind of reading these analogies, well applied, help avoid.

Uncertainties

There is no single official page, among the sources consulted in this cycle, that brings volt, watt, and watt-hour together in the same diagram — the triad here is composed from separate EIA and DOE materials.
The aggregate percentage of transmission and distribution losses (in the US or globally) is not verified in this dossier; only the qualitative fact that more is generated than sold due to heat losses.
A single operational definition of “available capacity” or “margin” for connection purposes is not fixed here — it depends on the local operator and is treated, with its own jurisdictional sources, in the text on connection queues and reinforcement.
The layer map for Brazilian distribution, as detailed as the EIA material for the US, still depends on ANEEL/PRODIST rules not fully covered in this cycle; what this text uses from Brazil covers generation and transmission via the ONS and the EPE.

References

EIA — How electricity is delivered to consumers. U.S. Energy Information Administration. Accessed August 09, 2026.
EIA — Measuring electricity. U.S. Energy Information Administration. Accessed August 09, 2026.
EIA — Capacity vs. generation FAQ. U.S. Energy Information Administration. Accessed August 09, 2026.
EIA — Electricity generation, capacity, and sales in the United States. U.S. Energy Information Administration. Accessed August 09, 2026.
U.S. DOE / Office of Electricity — Electricity 101. U.S. Department of Energy. Accessed August 09, 2026.
U.S. DOE / CESER — Electric Transmission and Distribution Backgrounder. U.S. Department of Energy. Accessed August 09, 2026.
U.S. DOE / NARUC — Electricity Transmission: A Primer. U.S. Department of Energy. Accessed August 09, 2026.
FERC — Reliability Primer. Federal Energy Regulatory Commission. Accessed August 09, 2026.
IEA — Electricity Grids and Secure Energy Transitions (executive summary). International Energy Agency. Accessed August 09, 2026.
ONS — O que é o SIN. Operador Nacional do Sistema Elétrico. Accessed August 09, 2026.
EPE — PDE, capítulo Transmissão de Energia Elétrica. Empresa de Pesquisa Energética. Accessed August 09, 2026.
EPE — Glossário, Anuário Estatístico de Energia Elétrica. Empresa de Pesquisa Energética. Accessed August 09, 2026.
U.S. DOE / NREL — Electricity Grid Basics (text version). U.S. Department of Energy. Accessed August 09, 2026.
EIA — Science of electricity, Energy Kids. U.S. Energy Information Administration. Accessed August 09, 2026.
U.S. DOE — National Transmission Needs Study. U.S. Department of Energy. Accessed August 09, 2026.