5. Hydroelectric Power

After two modules of furnaces, this one is a relief. No fuel. No flame. No chimney, no ash, no Carnot ceiling. Just water that was lifted into the sky by the sun, dropped on a mountain as rain, and is now trying to get back to the sea.

We simply put a turbine in its way and charge it a toll.

Potential energy, cashed in

Module 1 introduced potential energy: energy stored by position. Water high up has it; water at sea level has spent it. A hydro plant is a machine for collecting the difference.

The whole thing rests on two numbers:

• Head — how far the water falls, in metres. This sets how much energy each litre carries.
• Flow — how many litres arrive per second. This sets how many litres you are collecting from.

Power is essentially head × flow (times gravity, times water’s density, times efficiency). Which means the same power can come from two completely opposite situations — and that is the answer to our hook question:

• A Himalayan stream with a 500-metre drop and a modest flow.
• A Gangetic river with a 6-metre drop and a colossal flow.

Both can be a hundred-megawatt station. But in the first, water arrives as a screaming jet; in the second, as a slow ocean. You cannot use the same machine for both.

The penstock in that diagram is worth a moment. It is the steep pipe from reservoir to turbine, and by the bottom the water inside is under enormous pressure. A ruptured penstock is one of the more spectacular failure modes in engineering, which is why they are so heavily built and so carefully instrumented.

The turbine gallery

Three designs cover essentially all the world’s hydro, and each is a different answer to “how is the water arriving?”

Pelton — very high head, small flow. Nozzles fire jets of water at up to 150 m/s onto a wheel rimmed with double-cupped buckets. Each bucket splits the jet, turns it back on itself and lets it fall away almost motionless — having handed over nearly all its energy. Pelton wheels spin in air, not submerged; they are the mountain specialists.

Francis — medium head, medium flow. The world’s workhorse, and the most common turbine on Earth. Water enters a spiral casing, is aimed inward onto a runner by a ring of guide vanes, and exits along the axis, giving up energy by both pressure and momentum. Those vanes are not fixed: they swivel together like the slats of a venetian blind, and engineers call them wicket gates. Open them wider and more water reaches the runner; close them and the machine throttles back. That ring is how a Francis unit changes its power output, and it is a large part of why a hydro station can follow demand minute by minute. Its range of head is enormous — from around ten metres to several hundred — which is why most big dams use one.

Kaplan — low head, huge flow. Essentially a giant ship’s propeller running backwards. It has the same ring of swivelling guide vanes as a Francis, and then adds a second control on top: the propeller blades themselves can be adjusted in pitch while running. Two adjustments instead of one — engineers say a Kaplan is double-regulated, where a Francis is single-regulated. That second adjustment matters: a low-head river plant sees flows swing wildly between monsoon and dry season, and a fixed-pitch propeller would be efficient at only one of them.

All three routinely exceed 90% efficiency — dramatically better than any thermal plant. There is no Carnot limit here, because this is not a heat engine. Nothing is being burned; motion is simply being handed from water to metal.

Dams, rivers, and what a reservoir really is

Two ways to build:

Run-of-river takes the river as it comes, diverting part of the flow through turbines with little or no storage. Cheaper, far less submergence — but you generate what the river gives you, when it gives it. In India that means a great deal of power in the monsoon and much less in April.

Storage (dam and reservoir) blocks the valley and creates a lake. And here is the insight worth carrying into Part D: a reservoir is not just a water store, it is an energy store. Every cubic metre held behind that wall is potential energy, banked, waiting for you to decide when to spend it.

That makes hydro uniquely precious to a grid operator. A coal plant takes hours to warm up. A hydro station opens a gate and reaches full output in under a minute — no warm-up, no fuel to grind, nothing to preheat. When demand jumps or a plant trips, hydro is what catches the fall.

India’s dams

Hydro supplies about 53 GW of India’s installed capacity and produced roughly 160 TWh — just under 9% of generation — in FY 2024-25.

Bhakra Nangal, on the Sutlej, is the emotional origin story. Built through the 1950s and 60s, it was the project Nehru called a “temple of resurgent India”: irrigation for Punjab, Haryana and Rajasthan, plus power for an industrialising north. Whatever one thinks of big dams today, Bhakra’s canals are part of why the Green Revolution had water.

Tehri, on the Bhagirathi in Uttarakhand, is the modern, complicated one. The complex totals 2,400 MW — a 1,000 MW conventional station, the 400 MW Koteshwar plant downstream, and a 1,000 MW pumped-storage scheme. It is also a case study in costs: the reservoir submerged the old town of Tehri and dozens of villages, displacing tens of thousands of people, and the project was contested for decades over seismic risk in a young, earthquake-prone mountain range.

The Northeast holds the largest untapped potential — the Brahmaputra basin especially — and also the sharpest questions, involving seismically active terrain, downstream effects in Assam and Bangladesh, and Chinese dam-building upstream on the same river.

An honest ledger for large hydro looks like this. On one side: low-carbon electricity for a century, irrigation, flood moderation, drinking water, and unmatched grid flexibility. On the other: submerged land and forest; displaced communities whose resettlement has often been slow, inadequate or both; blocked fish migration and altered sediment flows that starve downstream floodplains; methane from decaying vegetation in some tropical reservoirs; and silt gradually filling the reservoir, stealing the storage that justified building it. Both columns are real. Which weighs more depends on the specific river, and — uncomfortably — on whether you live upstream of the wall or under the water.

The original giant battery

Now the idea that will dominate Module 11.

Take two reservoirs, one high and one low. When electricity is cheap and plentiful — the middle of a sunny day, or 3 am — run the turbines backwards as pumps and push water uphill. When electricity is scarce and dear — that 7 pm peak — let it fall back down and generate.

That is pumped storage, and you lose energy on the round trip: roughly 70–80% comes back. Module 1’s heat tax, collected twice.

So why bother? Because the grid does not value energy; it values energy at the right moment. A unit at noon and a unit at 8 pm are physically identical and economically very different. Pumped storage buys cheap and sells dear, and in doing so keeps the lights on at the moment they would otherwise go out.

And it is not a curiosity. Pumped hydro remains the overwhelming majority of the world’s installed electricity storage — vastly more than all the grid batteries combined. Every lithium-ion megawatt-hour that gets built is joining a category that water has quietly dominated for a century.

🔥Why hydro is the grid's best friendAdvanced · optional

Three properties make hydro worth far more than its share of generation suggests.

Ramping. Output can swing from zero to maximum in well under a minute, repeatedly, without damaging anything. Thermal plants hate that; hydro shrugs.

Black-start. After a total grid collapse, restarting is a circular problem: a coal plant needs megawatts of electricity for its own pumps, fans and mills before it can produce any. If the grid is dead, where does that come from? A hydro station needs almost nothing to begin — the energy is already stacked behind the dam. It opens a gate, generates, and energises the transmission lines that wake the thermal plants. When Module 10 covers the 2012 blackout and the restoration that followed, remember whose hand was on the gate.

Inertia and frequency response. A hydro station’s spinning mass helps stabilise grid frequency, and its governor can adjust output within seconds to correct a wobble. This matters increasingly as inverter-based solar — with no spinning parts at all — takes a larger share. Module 10’s 🔥 block returns to this.

The strategic point: as a grid gets more solar and wind, the value of a plant is decreasingly about how many units it makes and increasingly about how fast it can change what it is doing. By that measure, hydro is the most valuable thing on the system.

🔥🔥The hydropower equation, and what 'specific speed' really decidesDeep · optional

The power available. A mass m of water falling height h releases mgh joules. Divide by time and use mass flow rate ρQ (density × volumetric flow):

P = ρ · g · Q · H · η

with ρ ≈ 1,000 kg/m³, g = 9.81 m/s², Q in m³/s, H in metres, and η the overall efficiency (~0.9).

Try it. Q = 100 m³/s through H = 100 m:

P = 1000 × 9.81 × 100 × 100 × 0.9 ≈ 8.8 × 10⁷ W ≈ 88 MW

Notice the symmetry: doubling head and doubling flow both double the power. That is precisely why two utterly different rivers can host equally powerful stations — and it is the mathematical form of our hook question.

Specific speed. Turbine selection is not aesthetic; it is governed by a dimensionless-style parameter, specific speed:

N_s = N · √P / H^(5/4)

where N is rotational speed. Low N_s (high head, low flow) → Pelton. Medium → Francis. High N_s (low head, high flow) → Kaplan. Choose wrongly and you get cavitation, vibration and rapid destruction rather than merely poor efficiency.

Cavitation deserves its own note, since it is the characteristic hydro failure. Where water accelerates, its local pressure drops; if it falls below the vapour pressure, the water boils at ambient temperature, forming bubbles. Those bubbles then collapse violently on the blade surface, each implosion hammering the metal. Over months this chews visible craters into steel. Avoiding it constrains how deep the turbine must sit relative to the tailwater and how aggressively blades may be shaped — one of the real limits on hydro design.

What comes next

Water, gravity and a gate — the most elegant machine in the course, and the one the grid leans on hardest in a crisis. Module 6 goes to the opposite extreme of sophistication: a fuel so concentrated that a pellet the size of a pencil eraser replaces a tonne of coal, held in a chain reaction on a leash. And when we get to the turbine hall, you will recognise it exactly — because it is a coal plant’s, with a different kettle.