Part 1

Energy Fundamentals

The mental model the rest of the guide rests on — how raw resources become useful work, why power is not energy, and the six-layer architecture of the modern energy system.

118 min read

The Energy System: From Primary Energy to Useful Work

The mental model that makes every other chapter in this guide make sense — how raw resources become the motion, computing, heat and light that run the modern economy.

Modern civilization is an energy conversion system. This chapter builds the shared map the rest of the guide depends on: the difference between power and energy, the three conversion architectures behind thermal, solar and wind generation, why electricity is a carrier rather than a source, the six layers running from resource to useful work, and the arithmetic of storage duration, capacity factor and energy density.

Power vs energyConversion chainsCapacity factorEnergy density
Read chapter 1
222 min read

Electricity: The Physics Behind the Grid

Charge, voltage, current, power and frequency — the five quantities that determine the size of every cable, transformer, battery and inverter in the modern energy system.

Saying “electricity” hides an enormous amount of engineering. This chapter unpacks what electricity actually is and why the grid is built the way it is: Ohm’s law and the arithmetic of power, why I²R losses force transmission to high voltage, how transformers step voltage up and down, the AC/DC conversions now running through every solar plant and EV, what frequency reveals about grid balance, and why real, reactive and apparent power determine the size of every piece of equipment on the network.

Ohm’s lawI²R lossesPower factorThree-phase AC
Read chapter 2
326 min read

Power, Energy, Capacity & Efficiency

The vocabulary that separates a nameplate spec sheet from a bankable energy asset — MW vs MWh, capacity factor, C-rate, round-trip efficiency, and why “how big is it?” is never a complete question.

A 500 MW solar plant, a 100 MW / 400 MWh battery and a 90% capacity factor all sound like they measure the same thing, and none of them do. This chapter works through the quantities professionals actually trade on: how MW and MWh combine into storage duration, why capacity factor makes $/MW comparisons meaningless, how efficiency and capacity factor are independent axes, the Carnot ceiling on any heat engine, round-trip efficiency and where the missing energy goes, C-rate and the Ragone trade-off between energy and power density, data-centre load and PUE, DC/AC clipping on oversized solar arrays, usable versus nameplate battery energy, and how all of it feeds into LCOE and LCOS.

MW vs MWhCapacity factorRound-trip efficiencyLCOE and LCOS
Read chapter 3
428 min read

The Economics of Energy

CAPEX, OPEX, WACC, NPV, IRR, LCOE and merit order — why the cheapest technology on paper does not always win the project, the market, or the grid.

Physics sets what is possible; economics decides what gets built. This chapter is the finance half of an energy engineer’s education: what CAPEX really includes beyond the equipment invoice, how fixed and variable OPEX behave differently, why the capital-intensive versus fuel-intensive split determines whether a technology fears interest rates or commodity cycles, how WACC is assembled and why two points of it can outweigh a decade of efficiency gains, discounting, NPV and the three failure modes of IRR, LCOE and everything it hides, capture rates, the merit order and marginal pricing, price cannibalisation and the duck curve, LCOS and why storage is never financed on arbitrage alone, revenue stacking, PPAs and merchant risk, DSCR-based debt sizing, the compounding cost of delay, learning curves, why manufacturing and project economics are near-opposites, commodity exposure, curtailment and capacity credit, and how policy instruments differ in bankability.

WACC and NPVLCOE and LCOSMerit orderRevenue stacking
Read chapter 4
530 min read

The Electrical Grid: From Generation to Consumption

The world’s largest coordinated machine — how generation, transmission, distribution, protection and control combine into a system that must balance itself every single second.

The grid connects power plants, solar farms, batteries, factories, data centres and homes into a system with no single owner and no single location, which nonetheless has to stay electrically coordinated every second. This chapter builds the architecture underneath everything covered so far: why the wires store nothing and what that forces, the six conversion paths that all end in AC, synchronous machines versus inverter-based resources and the inertia problem, why transmission exists and when HVDC beats HVAC, substations and the transformer lead-time crunch, selective protection, distribution and the prosumer, why infrastructure is sized for the peak rather than the average, ramp rate and the evening cliff, grid strength in weak networks, congestion and curtailment arithmetic, interconnection queues as the real bottleneck, islanding and microgrids, data centres as grid-scale loads, storage at every layer, the BMS/PCS/EMS/SCADA distinction, N-1 economics, reliability versus resilience, virtual power plants, and why the network rather than the generator may be the binding constraint on the transition.

Transmission and distributionPeak vs energyCurtailmentGrid control
Read chapter 5

Part 2

Power Generation

How electricity is actually made — the physics, engineering and economics of each generation technology in turn, and what decides whether a given plant is worth building.

632 min read

Solar Power: From Photons to Gigawatts

From a single photon striking a silicon wafer to a gigawatt-scale power plant — the physics, manufacturing, engineering and finance that together make solar the fastest-growing source of electricity on Earth.

A solar plant is millions of cells, thousands of modules, inverters, trackers, transformers and a grid connection — and the physics is the part that is already settled. This chapter builds the complete chain: the photovoltaic effect and why silicon won on manufacturability rather than performance, the eight-stage supply chain, cell architectures from PERC through TOPCon and HJT to IBC, why most incident energy is lost to bandgap physics, the solar resource in DNI, GHI and DHI, capacity factor and why nameplate MW means little, DC/AC ratio and deliberate clipping, everything an inverter does beyond conversion, tracking and balance of system, the eight-stage loss cascade to a 78–84% performance ratio, the 14% temperature derate that makes the hottest sites not the best sites, soiling and degradation, EPC and the project lifecycle, why solar’s near-pure CAPEX structure makes WACC decisive, the four different costs that get quoted interchangeably, supply-chain concentration, solar plus storage, and capture price — the reason the marginal value of solar falls as solar gets cheaper.

PhotovoltaicsPerformance ratioDC/AC ratioCapture price
Read chapter 6
734 min read

Wind Power: From Kinetic Energy to Gigawatts

From a moving parcel of air to a gigawatt-scale offshore array — the aerodynamics, control systems, resource statistics and plant-level engineering that turn an invisible fluid into dispatchable-looking electricity.

Solar converts a photon flux arriving uniformly across a surface; wind converts the kinetic energy of a fluid that must physically pass through a rotor, and that single difference changes everything. This chapter builds the chain from moving air to a grid-connected wind farm: the cube law and why a 10% better site holds 33% more energy, swept area and the case for ever-larger rotors, air density as the overlooked third variable, the Betz limit at 16/27, power coefficient and tip-speed ratio, lift rather than drag and why blades are twisted, the four regions of the power curve and why turbines deliberately spill energy above rated, wind shear and the economics of hub height, turbulence and IEC turbine class, geared versus direct-drive drivetrains, pitch and yaw as protection systems, DFIG versus full-converter generators, specific power as the dial that sets capacity factor, Weibull statistics and the P50/P90 gap that decides debt capacity, wake effects — the loss mechanism solar does not have — layout optimisation and wake steering, onshore versus offshore, floating foundations, HVDC export cables, the gross-to-net loss cascade, and why wind’s capture price falls over multi-day windy periods rather than a daily midday block.

Cube lawBetz limitWake lossesP50 vs P90
Read chapter 7
834 min read

Nuclear Power: From Fission to Firm Electricity

Not a harvested energy flow but energy released from inside the atomic nucleus — the physics, safety architecture and economics of the highest-energy-density, highest-capital-intensity technology in the entire energy system.

Solar and wind harvest flows that already exist; nuclear releases energy locked inside atomic nuclei at a density that makes every chemical fuel look like a rounding error. This chapter builds the chain from binding energy to a grid-connected gigawatt: mass defect and E = mc², why once-through uranium carries roughly 21,000 times more energy per kilogram than coal, fission and the chain reaction, why “critical” describes normal operation rather than danger, the neutron economy and why enrichment is a sovereignty decision as much as a technical one, PWR versus BWR versus PHWR, control rods and scram — and then the concept that governs the entire discipline: decay heat, which leaves a scrammed 3 GW-thermal core producing 192 MW at shutdown and still 7.6 MW a week later. From there the chapter turns to defence in depth, why roughly 75% of nuclear’s levelised cost is capital and financing while under 10% is fuel, how extending a build from eight to fourteen years raises cost by 19% through compounding interest alone, the capacity factor advantage and why comparing capacity factors across technologies misleads, load-following and its hidden cost, synchronous inertia as a distinct value stream, data centre demand, the SMR bet on trading economies of scale for economies of volume, the learning curve problem, geological disposal, proliferation, and India’s three-stage thorium programme.

Decay heatCriticalityCost of capitalSMRs
Read chapter 8
930 min read

Hydropower: From Water Cycles to Grid-Scale Energy and Storage

The original renewable and the original grid-scale battery — why a reservoir is simultaneously a power plant, a water asset and the most valuable flexibility resource on most electricity systems.

Every generator covered so far makes electricity; a reservoir also decides when. This chapter builds the chain from rainfall to a dispatched megawatt: hydropower as solar energy once removed, the governing equations E = mgh and P = ρgQHη, and why head and flow multiply together but cost completely differently — a 300 MW plant needs about 34 m³/s at 1,000 m of head but 2,265 m³/s at 15 m, sixty-seven times the water for identical output, which is why turbine choice between Pelton, Francis and Kaplan falls out of site geography rather than preference. From there: why hydro escapes the Carnot penalty that caps thermal plants near a third, the difference between reservoir and run-of-river that makes identical nameplate ratings worth entirely different amounts, capacity factor and the two very different situations it conflates, black start, and pumped storage as Chapter 4’s arbitrage implemented in gravity. The economics section works through why round-trip efficiency imposes a required price ratio rather than a required spread, why solar has migrated the cheapest hours from overnight into the middle of the day, and why batteries are the cheap way to buy power while pumped hydro is the cheap way to buy energy. The chapter closes on the constraints that actually decide projects: multi-purpose water obligations that outrank electricity, sedimentation that erodes energy capacity while leaving power capacity untouched, resettlement and consent, a historical flow record that climate change has made unreliable, and transmission distance.

Head and flowPumped storageRound-trip efficiencySedimentation
Read chapter 9
1032 min read

Coal, Gas and Combustion: The Incumbent and Its Economics

The only generation technology whose fate is decided by a price it does not control — why the cheapest plant to build is so often the most expensive to run, and why a fleet can become uneconomic without any of its costs changing.

Every technology in this part has been measured against one incumbent: a fire under a boiler. This chapter builds combustion from the chemistry up. Burning fuel is oxidation, so a thermal plant is a heat engine and pays the Carnot penalty in full, leaving exactly two levers — fire hotter, or reject colder. That explains the whole efficiency ladder: subcritical coal at 36% against an ultra-supercritical unit at 45%, where the Carnot ceiling barely moves and almost all the gain comes from capturing more of it; an open-cycle gas turbine stuck near 38% despite firing a thousand degrees hotter, because it dumps its exhaust at 550 °C; and combined cycle at 62%, which fixes nothing about the fire and everything about the cold end. From there the chapter turns to the numbers the industry actually trades on: heat rate as 3.6 ÷ efficiency, fuel cost as heat rate times fuel price, and the merit order that lets a marginal gas plant set the price every generator receives. The screening curve then separates what runs tonight from what gets built — peakers below a 17% capacity factor, CCGTs from there to 73%, coal only above 73%, which is the story of the last decade compressed into one crossing point. The chapter closes on emissions derived rather than quoted: 950 kg CO₂/MWh from subcritical coal against 325 from a CCGT, what a 2.3% methane leak rate does to that comparison, why 90% capture lands near 131 kg CO₂e/MWh rather than near zero, and the four-line calculation that gives the carbon price at which coal switches to gas.

Heat rateMerit orderScreening curveCarbon intensity
Read chapter 10

Part 3

Storage and Flexibility

What a grid does when generation and demand refuse to line up — the technologies, control systems and economics that move electricity through time rather than through space.

1134 min read

Energy Storage: From Batteries to Grid-Scale Flexibility

Transmission moves electricity through geography; storage moves it through time — the electrochemistry, control architecture and economics that turn a variable resource into a dispatchable one.

Every generator in Part 2 answers the same question: how is a megawatt-hour made? Storage changes the question. It creates nothing — it moves energy that already exists from a moment when it is abundant to a moment when it is valuable, and that single capability turns solar and wind into something a grid operator can dispatch. This chapter builds storage from the cell up. Duration — energy divided by power — is the number that decides what an asset is for, and the number most often missing from the announcement: two systems both described as “100 MW batteries” can be a one-hour frequency asset or an eight-hour energy asset, and they are not substitutes. From there: the duration ladder that spans eleven orders of magnitude and explains why no technology covers all of it; why the cathode decides almost everything about a lithium-ion cell, with NMC carrying about 1.6× the energy per kilogram and LFP surviving roughly 2.4× the cycles; why a vehicle that must accelerate its own battery makes the opposite choice from a grid asset that never moves; and why flow batteries, which scale power and energy independently, are the mirror image of lithium-ion. The chapter then works through the arithmetic that financial models get wrong: round-trip efficiency as a permanent multiplier that imposes a required price ratio rather than a spread, depth of discharge as a design choice that buys three times the cycle life at 50% rather than 100%, and the usable window — nameplate times SOC range times state of health times efficiency — as the only energy figure that belongs in a model. It closes on the parts that decide whether the asset survives: the weakest-cell rule and balancing, the BMS/PCS/EMS hierarchy and the BMS’s absolute veto, thermal runaway and why chemistry alone does not solve safety, and the cost stack in which the cell is only about 45% of an installed system.

DurationRound-trip efficiencyDepth of dischargeBMS, PCS and EMS
Read chapter 11
1232 min read

Long-Duration Energy Storage: Beyond the Four-Hour Battery

Why almost every grid battery is sized for four hours, what breaks past that wall, and the two-axis cost model that decides which technology wins at which duration.

Almost every grid battery being built today is sized for about four hours — not because of physics, but because that is where lithium-ion is cheapest and because four hours bridges a midday solar surplus to an evening peak. Grids also fail across a week of weak wind and a fortnight of monsoon cloud, and for those a four-hour battery is not undersized but irrelevant. This chapter is about everything past that wall. It rests on one idea: every storage system has two independent capital costs, one that scales with power and one that scales with stored energy, so cost per kilowatt-hour of capacity is power cost divided by duration plus energy cost. The first term shrinks with hours; the second is a floor. That single structure explains why a blended $/kWh figure is meaningless without a duration attached, why lithium-ion is cheap on power and dear on energy while hydrogen is the reverse at roughly 13× the power cost and 58× cheaper per kilowatt-hour stored, and why the crossovers are arithmetic rather than opinion — flow overtaking lithium-ion near 7.7 hours, pumped hydro near 14, hydrogen near 100. Splitting Chapter 11’s $165/kWh installed cost into $200/kW and $115/kWh reproduces it exactly at four hours, which is the check that the model is honest. From there: flow batteries, mechanical storage, thermal storage that skips a conversion entirely when the end use is heat, and hydrogen as the only realistic seasonal store despite returning 30–40%. Then the economics that decide what actually gets built — why efficiency matters in proportion to cycling frequency, why self-discharge and calendar life eliminate candidates before price is discussed, why duration caps revenue frequency at 8,760 ÷ 2D cycles a year, why a $27/kWh system cycled five times a year delivers capital at $216/MWh against $30/MWh for a battery cycling daily, and why a first-of-a-kind risk premium can turn a 21% hardware advantage into an 11% annual disadvantage.

Power cost vs energy costCrossover durationMissing moneyBankability
Read chapter 12

More chapters are being written. This series is planned to run to 80 chapters. New chapters appear here as they are published.