Skip to content
WattCurious

Pelton vs Turgo vs Cross-Flow Turbines

Updated 21 September 20267 min readWind & Hydropower

An impulse turbine converts the whole of the available head into the speed of a free jet, then extracts energy by turning that jet around. Nothing downstream of the nozzle is pressurised, which makes these machines simple, tolerant and easy to maintain. The three common designs differ in one thing — the geometry by which the runner meets the jet — and that single difference sets which heads and flows each can work with, and how gracefully each copes when the stream is low.

Key takeaways

  • An impulse turbine turns all the head into jet velocity first, so the runner spins in air at atmospheric pressure.
  • A Pelton bucket reverses the jet almost completely, which is why it extracts so much of the energy available.
  • The Turgo takes the jet at an angle so water enters one side and leaves the other, letting a given runner swallow more flow.
  • A cross-flow turbine passes water through the runner twice and can be split into part-width sections, which is what makes it forgiving of varying flow.
  • On a real stream the flow changes all year, so efficiency at part flow usually matters more than peak efficiency.
On this page
  1. All the pressure becomes speed first
  2. Pelton: reverse it almost completely
  3. Turgo: take the jet at an angle
  4. Cross-flow: through the drum, twice
  5. Matching the machine to the site
  6. Part flow is what decides it

All the pressure becomes speed first

The defining feature of an impulse machine is where the pressure disappears. In the nozzle, entirely. By the time water reaches the runner it is a free jet travelling through air, and the runner is turning in atmosphere rather than running full.

All the pressure becomes speed, in the nozzleA penstock delivering pressurised water to a nozzle, where the pressure converts into the velocity of a free jet. The jet strikes a runner spinning in air at atmospheric pressure, is turned around by the blade, and falls away at much reduced speed. Labels mark that the whole pressure drop happens in the nozzle, that the runner casing is not pressurised, and that the energy extracted is the change in the water's momentum as the blade reverses its direction.The pressure disappears before the water reaches the runnerpenstock — under pressurenozzlefree jet — at atmospheric pressurerunner, spinning in airspent water falls away, slowcasing — not pressurisedAlong the pathpressurevelocitypenstocknozzlejetafter the runnerThe energy extracted is the change in the jet's momentum — so the more completely the blade reverses it, the more it takes.That single sentence is what separates the three designs: each is a different answer to reversing as much of the jet as possible.
Pressure becomes velocity in the nozzle; the runner then turns that velocity around. Nothing after the nozzle is under pressure, which is why these machines are so tolerant.

Two consequences follow. The casing does not have to contain pressure, so the machine is simpler, lighter and easier to open up. And the energy extracted is the change in momentum of the jet — so the more completely the blade reverses the flow, the more of the jet's energy is taken.

That single sentence explains the three designs. Each is a different answer to "how do I reverse as much of the jet as possible, while letting the spent water get out of the way?"

Pelton: reverse it almost completely

The Pelton answer is a ring of double bowls with a sharp ridge down the middle.

The Pelton bucket and its splitterA Pelton runner in section with a jet striking one bucket. The bucket is a pair of hemispherical bowls divided by a central splitter ridge. The jet meets the splitter, divides into two halves, and each half is turned through nearly one hundred and eighty degrees so that it leaves almost directly back along the incoming direction. A note marks that the spent water must fall clear before the next bucket arrives, and a second marks that the split keeps side forces balanced on the shaft.The bucket, from abovesplitter ridgejetturned back, almostthe way it cameand the same belowrunner rim, buckets around itWhy the splitterTurning the jet through close to 180° takes nearlyall the momentum available — which is why a goodPelton is the most efficient of the three at itsdesign point. The split also balances side thrust.And the constraintThe spent water has to fall clear of the runnerbefore the next bucket arrives, which limits howlarge a jet a given runner can accept.So: high head, and modest flow.A lot of speed, in a small stream of water.Multi-jet machines get around the flow limit by adding nozzles rather than by enlarging one — and shutting jets off is howa Pelton copes with part flow, keeping the remaining jets at full velocity instead of throttling all of them.
The splitter halves the jet and each half is turned right around. Reversing the flow almost completely is what extracts so much of its momentum.

Turning the jet through close to one hundred and eighty degrees takes nearly all the momentum available, which is why a well-made Pelton is the most efficient of the three at its design point. The splitter also balances the side thrust, so the bearings only carry the intended loads.

The constraint is the spent water. It has to fall clear of the runner before the next bucket arrives, which limits how large a jet a given runner can accept. Pelton machines therefore favour high head and modest flow — a lot of speed in a small stream of water.

Turgo: take the jet at an angle

The Turgo changes one thing: the jet arrives at an angle to the runner plane, so water enters one face and leaves the other.

The Turgo takes the jet at an angleA Turgo runner with the jet arriving at an oblique angle to the plane of the runner. Water enters at the front face of a blade, is turned, and exits from the back face rather than returning the way it came. Alongside, a Pelton is shown for comparison with the jet arriving in the runner plane and the spent water having to fall clear on the same side. A note marks that because the Turgo's exit path is clear of the inlet, it can accept a larger jet for the same runner diameter.Pelton — in and out on the same sideTurgo — in one face, out the otherrunner planejet, in the planespent water leaves on thesame side it arrivedSo the jet cannot be made much larger —it would run into its own spent water.runner planejet, at an angleand out through the back faceThe exit never blocks the inlet —so a larger jet fits on the same runner.What that buysMore flow for a given runner diameter — or thesame flow at a higher rotational speed, which canremove a gearbox between runner and generator.Medium head, more flow than a Pelton its size.And what it costsThe jet is not turned quite as completely, so lessof its momentum is recovered and peak efficiencysits a little below a comparable Pelton.It is a different answer, not a cheaper one.
In and out through different faces. The spent water never has to get out of the incoming jet's way, so a given runner can swallow a larger jet.

Because the exit path is clear of the inlet, the flow constraint that limits a Pelton is relaxed. A Turgo of a given diameter accepts a larger jet, so it takes more flow — or runs faster for the same flow, which can remove a gearbox between the runner and the generator.

The price is that the jet is not turned quite as completely, so peak efficiency is a little below a comparable Pelton. In exchange, the machine occupies the middle ground: medium head, more flow than a Pelton of the same size.

Cross-flow: through the drum, twice

The cross-flow is a different shape entirely — a wide drum with blades around its circumference, fed by a rectangular jet across its full width.

The cross-flow drum, and its divided inletA cross-flow turbine in section. A rectangular nozzle directs a sheet of water across the full width of a drum-shaped runner. The water passes inward through the blades on one side, crosses the empty centre of the drum, and passes outward through the blades on the opposite side, giving up energy on both passes before leaving. A separate view along the shaft shows the inlet divided into a one-third and a two-thirds section, each with its own guide vane, so the machine can run either section alone or both together.In section — the water crosses the runner twiceempty centrerectangular nozzlefirst pass: inward through the bladessecond pass: outward through the far sideand awayAlong the shaft — the inlet is dividedone thirdtwo thirdseach section has its own guide vanesmall onlylarge onlybothThree sensible operating widths, instead of one.A stream at a third of design flow runs through the smallsection at its proper velocity, rather than through a largenozzle barely open.Lower peak efficiency, far simpler constructionA drum, straight blades, a rectangular nozzle. Energy is taken on both passes, and neither pass reverses theflow as completely as a Pelton bucket does.The simple inlet and blades also tolerate more of what a stream carries — which matters more than efficiency on a silty site.
Water crosses the runner twice, giving up energy on the way in and again on the way out. The divided inlet is what lets the machine match a stream that changes.

Water passes through the blades on the way in, crosses the hollow centre and passes through again on the way out, taking energy on both passes. Peak efficiency is lower than the other two, and the construction is far simpler — a drum, straight blades, a rectangular nozzle.

The feature that matters is the divided inlet. Split the nozzle into a third and two thirds, each with its own gate, and the machine has three sensible operating widths instead of one. A stream at a third of its design flow can be run through the small section at close to its proper velocity rather than through a large nozzle barely open.

Matching the machine to the site

Head and flow, established by a micro-hydro site survey, select the machine before anything else does.

Which machine suits which site
PeltonTurgoCross-flow
HeadHighMedium to highLow to medium
Flow for a given sizeLowModerateHigh
Jet reversalNearly completePartialTwo passes, partial each
Peak efficiencyHighest of the threeClose behindLower
Part-flow behaviourGood with multiple jetsGoodVery good with a divided inlet
Runner speedLower for a given headHigher — may avoid a gearboxLower, wide runner
ConstructionPrecise cast bucketsPrecise cast bladesSimple fabricated drum
Tolerance of debris and siltSensitive at the nozzleSensitive at the nozzleMore forgiving

Design tendencies, not thresholds. The boundaries overlap considerably and a specific manufacturer's range will not match these generalisations exactly.

The last row deserves emphasis in a small scheme. A nozzle is a precision orifice and silt abrades it; a cross-flow's rectangular inlet and simple blades put up with more of what a stream carries, which is why intake design and settling matter more for the first two. It is the same survivability-against-output trade that decides wave and tidal devices, where the water is rougher still.

Part flow is what decides it

Here is the conclusion that surprises people who compare peak efficiencies.

Efficiency across the whole range of flowsEfficiency plotted against fraction of design flow for three machines. A single-jet Pelton peaks highest but falls away sharply below about half flow. A Turgo holds up better across the middle. A cross-flow with a divided inlet peaks lower but stays nearly flat from full flow down to about a quarter, because sections can be shut off. Beneath, a flow duration curve shows how much of the year a stream actually spends at each fraction, with most of the year well below full flow.Efficiency against fraction of design flowPelton, one jetTurgocross-flow, dividedefficiency¼½¾fullthe Pelton collapses herethe cross-flow does notAnd this is where the stream actually spends its yearhoursmost of the year is spent down here¼½¾fullWhat a machine delivers annually is its efficiency at each flow, weighted by the hours the stream spends there.Illustrative shapes. A flatter curve can beat a higher peak — which is why peak efficiency is the last thing to look at, not the first.
Illustrative shapes. Peak efficiency is a single point; a stream spends the year spread across the whole axis, and the area under the combination is what you actually harvest.

A stream is not at its design flow for most of the year — the flow duration curve says so plainly. What a machine delivers annually is its efficiency *at each flow*, weighted by how many hours the stream spends there. A machine that holds 70 per cent of its rating across a wide range can beat one that reaches a higher peak and then collapses below half flow.

This is why multi-jet Peltons exist: shutting off jets is the Pelton's version of the cross-flow's divided inlet, keeping the remaining jets at full velocity rather than throttling all of them. And it is the same principle that governs the Betz limit's cousin in wind, where a turbine is judged on the whole distribution of wind speeds rather than on its best one.

The honest summary: let head and flow narrow the field, then choose on part-flow behaviour and on how much of what the stream carries the machine will tolerate. Peak efficiency is the last thing to look at, not the first — the same reasoning that makes round-trip stability matter more than peak efficiency in storage, and for the same reason: what you keep across a year is not what a data sheet reports at one operating point.

Frequently asked questions

What makes a turbine an impulse turbine?

That the pressure drop happens entirely in the nozzle, before the water reaches the runner. The runner spins in air at atmospheric pressure and is driven by the momentum of a free jet. A reaction turbine, by contrast, runs full of water with pressure dropping across the runner itself, which is a different machine with different sealing and siting requirements.

Why does a Pelton bucket have a splitter down the middle?

So the jet is divided and turned back on both sides symmetrically. Turning the water through nearly one hundred and eighty degrees extracts close to the maximum momentum available, and splitting it keeps the axial forces balanced so the bearings are not pushed sideways.

Is a Turgo just a cheaper Pelton?

No — it solves a different problem. Because the jet enters one side of the runner and exits the other, the spent water does not have to get out of the way of the incoming jet. That lets a Turgo accept a larger jet for a given runner diameter, so it handles more flow, or the same flow at a higher speed, than a Pelton of the same size.

Why is a cross-flow turbine good at low flow?

Because it can be built with a divided inlet, typically in a one-third and two-thirds split. Running one section, the other, or both gives three sensible operating widths, so the machine can be matched to whatever the stream is doing without running one large nozzle badly throttled.

Which is most efficient?

At its design point a well-made Pelton is usually the highest of the three, with the Turgo close behind and the cross-flow lower. That ranking matters less than it appears, because a stream spends most of the year away from any single design point, and the flatter part-flow curve often delivers more energy across a year than the higher peak does.

Sources

Named organisations whose published material underpins this article. Where no link is given, the source is named rather than linked.

Editorial Team

Research, drafting and review

Articles are drafted from primary engineering and physics references with AI-assisted tools, then reviewed and fact-checked line by line by a human editor before publication. We publish explanations, not recommendations: no products, no pricing, no country-specific rules, and no invented author personas.

Last reviewed 21 September 2026. How we research and review

Wind & Hydropower

What Is the Betz Limit?

No open-flow rotor can capture more than 59.3% of the wind's power. The derivation in full, what the limit does not say, and where it does not apply.

Updated 21 September 20267 min read