Energy

How Flow Rate and Head Set a Pumped-Hydro Plant's Round-Trip Efficiency

By Lucian — builder & engineer, LK Forge

Pumping water uphill and letting it flow back down through a turbine sounds like a clean energy round trip, but friction takes a growing bite as flow rate rises. We swept the Pumped Hydro Storage lab's model across flow rate and head for a 300 m head, 1.8 GL reservoir to measure exactly how much round-trip efficiency you give up for flow, and what head and reservoir size set instead.

 ·  5 min read  ·  measured from the flow/head benchmark

80.77%
round-trip efficiency at a gentle 60 m³/s flow
70.32%
round-trip efficiency at 420 m³/s, the fastest flow benchmarked
158.69 → 1033.96 MW
generator power, 60 m³/s vs 420 m³/s
8×
more energy capacity at 800 m head than at 100 m (same reservoir volume)

Why higher flow has a physical cost

Water moving through a penstock or tunnel loses head to friction, and that loss scales with the square of flow rate. A gentle 60 m³/s flow barely touches the available 300 m head; a 420 m³/s flow — 7.0× the flow — loses proportionally far more of it. On the way down, that lost head means less pressure driving the turbine; on the way up, the pump has to overcome that same lost head just to move the water at all. The lost energy doesn't do useful work in either direction — it becomes heat and turbulence in the water column, which is also why real plants size their penstocks to keep friction loss modest at rated flow.

round-trip % = (machine efficiency)² × (head − friction loss) ÷ (head + friction loss), friction loss growing with flow²

That's why every row in the benchmark below moves the same direction: nothing about the plant changed between flow rows except how fast the water moved, and nothing about the flow changed between head rows except how far it fell.

Round-trip efficiency falls with flow

Round-trip efficiency for the 300 m head plant, plotted against pump/generate flow rate. The decline is steady across the whole sweep, with no plateau.

0% 25% 50% 75% 100% 0100200300400 flow (m³/s) round-trip

Generator power rises the opposite way

Generator power for the same 300 m head plant, plotted against flow rate. Power climbs steadily even as round-trip efficiency falls — flow rate trades one for the other.

0 275 550 825 1100 0100200300400 flow (m³/s) MW

The exact numbers

Every row below is a direct sample from the lab's model: a 300 m head, 1.8 GL reservoir, 0.9 machine efficiency, no evaporation loss, for the flow sweep.

Flow (m³/s) Round-trip Generator power
60 80.77% 158.69 MW
120 80.07% 316.01 MW
180 78.93% 470.59 MW
240 77.35% 621.04 MW
300 75.37% 766 MW
360 73.02% 904.1 MW
420 70.32% 1033.96 MW

And the separate head sweep, at a fixed 1.8 GL reservoir and 180 m³/s generating flow:

Head (m) Energy capacity Discharge duration
100 490.5 MWh 2.78 h
200 981 MWh 2.78 h
300 1,471.5 MWh 2.78 h
500 2,452.5 MWh 2.78 h
800 3,924 MWh 2.78 h

Plant: 300 m head, 1.8 GL reservoir, 0.9 machine efficiency, DEFAULT_K friction coefficient, no evaporation loss. Generated 2026-09-11.

Try it yourself

Open the lab, set your own flow rate, head, and reservoir size, and watch round-trip efficiency and generator power move in real time.

Open the Pumped Hydro Storage lab →
Share this X Facebook Reddit

Related reading

Common questions

What is round-trip efficiency in a pumped-hydro plant, and why does it fall at high flow?

Round-trip efficiency is the fraction of the electrical energy used to pump water uphill that you get back out when it flows back down through the turbines. In this model it comes from three factors: pump and turbine machine efficiency (each applied once per direction), friction head loss (which subtracts from the usable head on the way down and adds to the head the pump must overcome on the way up), and evaporation loss during any hold between filling and generating. Friction head loss grows with the square of flow rate, so it stays small at gentle flow and grows fast at high flow. For this 300 m head, 1.8 GL reservoir, a gentle 60 m³/s flow round-trips at 80.77%, but pushed to 420 m³/s that falls to 70.32% — 10.5 percentage points of energy lost purely to moving the water faster, with the head, machine efficiency, and everything else unchanged.

Why does a high flow rate waste energy to friction instead of just moving water faster?

Because friction head loss in the penstock and tunnels scales with the square of flow rate, not linearly. Doubling the flow roughly quadruples the friction loss, so a small increase in flow costs a disproportionate share of the usable head. On the generating side that lost head means less pressure driving the turbine per cubic metre of water; on the pumping side it means the pump has to push against extra head just to overcome friction. The benchmark shows this directly: round-trip efficiency slides from 80.07% at 120 m³/s to 77.35% at 240 m³/s to 73.02% at 360 m³/s, a steady decline with no plateau across the sweep. Higher flow always trades some round-trip energy for the power and speed gained; the only question is how much.

What do head and reservoir size actually set — power, energy capacity, or both?

Different things, and independently. Head sets how much power a given flow rate can deliver: generator power for a fixed flow rises directly with head, since pressure per cubic metre of water scales with height dropped. Head and reservoir volume together set total energy capacity — a taller drop or a bigger reservoir both store more potential energy. In this model, raising head from 100 m to 800 m at a fixed 1.8 GL reservoir multiplies energy capacity by 8×, from 490.5 MWh to 3924 MWh. Discharge duration, by contrast, is set purely by reservoir volume divided by generating flow rate — at a fixed 180 m³/s flow it stays at 2.78 hours regardless of head, because head changes the power delivered, not how long the reservoir lasts at a given flow.