A Jonval turbine is an axial-flow reaction water turbine that drives a horizontal runner using water flowing downward through a ring of fixed guide vanes. The Lowell-area textile mills and the U.S. Mint at Philadelphia ran Jonval units in the mid-1800s. It exists to extract shaft power from low-to-medium head sites where a vertical waterwheel would be too slow and a Fourneyron radial turbine too complex. A well-built Jonval reaches 70-75% efficiency at heads of 2-12 m.
Jonval Turbine Interactive Calculator
Vary flow, head, efficiency, and shaft speed to estimate Jonval turbine shaft power and torque while the canvas shows the axial-flow runner response.
Equation Used
The Jonval turbine shaft power estimate uses the standard hydraulic power equation multiplied by overall efficiency. Flow Q and head H set the available water power; efficiency eta converts that water power to usable shaft power. Torque is then calculated from shaft power and rpm.
- Fresh water density is fixed at 1000 kg/m3.
- Gravity is fixed at 9.81 m/s2.
- Efficiency is overall turbine efficiency from water to shaft.
- Torque uses the entered shaft speed and the computed shaft power.
Inside the Jonval Turbine
The Jonval turbine takes a column of water, spins it through stationary guide vanes set at a fixed angle — usually 25° to 35° from the runner plane — then dumps that swirling water straight down through the curved blades of a horizontal runner. The runner sits below the guide vanes, not beside them. Water flows axially: in at the top, out at the bottom. That axial flow path is what separates it from Benoit Fourneyron's earlier radial-outflow turbine, and it's what makes the Jonval shorter, easier to seat in a tailrace pit, and friendlier to a mill millwright with hand tools.
The guide-vane-to-runner gap matters more than people realise. You want it tight — 3 to 6 mm on a 1 m diameter runner — because any extra space lets the swirl decay before it hits the blades, and you lose the tangential momentum the runner is trying to harvest. Open it up to 15 mm through wear and you can drop 8-10 efficiency points without seeing anything obviously wrong. The other tolerance that bites you is runner-blade exit angle. If the casting comes out 5° off the drawing, the water leaves with residual swirl, the draft tube cavitates, and you'll hear a low rumble through the headstock bearing.
The machine fails in three predictable ways. Wooden guide vanes (the cheap 19th-century build) waterlog and warp, throwing the inlet angle off. Cast-iron runners pit and lose blade profile after 20-30 years in silty water. And the foot-step bearing at the bottom of the vertical shaft — almost always lignum vitae against bronze in original installations — wears oval and lets the runner rub the discharge ring. You can hear that rub before you can measure it.
Key Components
- Fixed Guide Vanes (Distributor): A stationary ring of curved vanes above the runner that turns the descending water column into a swirling jet at a fixed angle, typically 25-35°. The vanes are usually cast iron or, in early builds, oak. Vane spacing must be uniform within ±2 mm or the runner sees a pulsing load and the bearing journals wear unevenly.
- Horizontal Runner: A flat disc carrying curved blades — 12 to 24 of them depending on size — that catches the swirling water and converts its momentum into shaft torque. Runner diameters in original mill installations ran 0.6 m to 2.5 m. Blade exit angle must hold within ±2° of design, otherwise residual swirl drops efficiency and stresses the draft tube.
- Vertical Shaft: Carries torque from the runner up to the mill machinery or generator. On large units the shaft was wrought iron, 100-150 mm diameter, supported top and bottom. The bottom foot-step bearing carries the entire runner weight plus axial water thrust — often 5-15 kN on a mid-sized mill turbine.
- Sluice Gate or Register: Throttles flow into the guide-vane ring to match load. Most Jonval installations used a cylindrical curtain gate that lifted vertically. Closing the gate below 40% open kills efficiency fast — the vanes were cut for one design flow and don't tolerate part-flow well.
- Draft Tube: A diverging tube below the runner that recovers kinetic energy from the exit water by slowing it before it reaches tailwater. A proper Jonval draft tube expands at 6-8° per side over a length of 3-4 runner diameters. Skip the draft tube and you leave 10-15% of available head sitting on the floor.
Where the Jonval Turbine Is Used
The Jonval ran the New England textile boom, the early Midwestern flour mills, and a fair share of European industrial sites between 1843 and roughly 1900. It got displaced by the Francis turbine — better part-flow efficiency, better casing — but Jonval units stayed in service in low-head installations for decades after, and a handful are still turning today as heritage micro-hydro plants. The reason it persisted is simple: at fixed flow and fixed head, with clean water, a Jonval is mechanically straightforward and cheap to build with foundry-and-millwright skills.
- Textile Manufacturing: The Boott Cotton Mills in Lowell, Massachusetts ran Jonval-style turbines from the 1840s to drive carding and weaving line shafts off the Pawtucket Canal.
- Government Mints: The U.S. Mint at Philadelphia installed a Jonval turbine in 1851 to power coining presses, replacing an aging breastshot waterwheel.
- Grain Milling: Watkins Mill in Lawson, Missouri (now a state historic site) used a Jonval-pattern turbine to drive flour stones and a wool carding line from the 1860s.
- Heritage Micro-Hydro: Restored Jonval units at sites like the Hanford Mills Museum in East Meredith, New York generate 20-40 kW for grid-tie demonstration and museum lighting.
- Sawmilling: 19th-century Pacific Northwest sawmills along the Willamette tributaries used Jonval turbines to power circular saws where head was 3-6 m and flow was steady year-round.
- Iron Works: Adirondack bloomery forges in upstate New York drove tilt hammers and bellows off Jonval turbines fed by dam-and-flume systems on small creeks.
The Formula Behind the Jonval Turbine
Shaft power available from a Jonval turbine is the standard hydraulic-power equation scaled by overall efficiency. What matters for a practitioner is how the result moves across the operating range. At the low end of typical Jonval head — around 2 m — you're mostly fighting fixed losses and efficiency sits closer to 60%. At the design point, 6-8 m head with full design flow, a clean Jonval hits 72-75%. Push past 12 m and you're outside the geometry the runner was cut for; cavitation starts at the blade trailing edge and efficiency falls back into the 60s. Sweet spot is right where the historical mill sites clustered: 4-9 m head with steady seasonal flow.
Variables
| Symbol | Meaning | Unit (SI) | Unit (Imperial) |
|---|---|---|---|
| P | Shaft power output | W | hp |
| η | Overall turbine efficiency (typically 0.60-0.75 for Jonval) | dimensionless | dimensionless |
| ρ | Water density | kg/m³ (≈1000) | lb/ft³ (≈62.4) |
| g | Gravitational acceleration | m/s² (9.81) | ft/s² (32.2) |
| Q | Volumetric flow rate through the runner | m³/s | ft³/s |
| H | Net hydraulic head across the turbine | m | ft |
Worked Example: Jonval Turbine in a restored Appalachian gristmill micro-hydro retrofit
You are sizing a Jonval-pattern micro-hydro turbine for a restored 1870s gristmill on Spivey Creek near Burnsville, North Carolina. The dam gives 5.5 m of net head at the turbine inlet, the creek delivers 0.45 m³/s of design flow during spring and autumn, and the owner wants to estimate shaft power for a grid-tie generator package.
Given
- H = 5.5 m
- Q = 0.45 m³/s
- ρ = 1000 kg/m³
- g = 9.81 m/s²
- ηnom = 0.72 dimensionless
Solution
Step 1 — calculate the gross hydraulic power available in the water column before turbine losses:
Step 2 — at the nominal Jonval efficiency for this head and flow, 72%, compute shaft power:
That's the design point. Steady, predictable, the runner sitting in its sweet spot for blade angle and exit swirl. With a 92%-efficient generator downstream you'd see roughly 16 kW at the terminals — enough to run the gristmill's restored line shaft and export 8-10 kW to the local grid.
Step 3 — at the low end of the typical operating range, 50% flow during late summer drought (Q = 0.225 m³/s), the Jonval's part-flow penalty kicks in because the guide vanes are fixed. Effective efficiency drops to roughly 0.55:
Not 8.7 kW like a linear scaling would suggest — the fixed-geometry distributor punishes you below 60% design flow. Step 4 — at the high end, spring freshet conditions push Q to 0.60 m³/s but H drops to 5.0 m because the tailrace floods and back-pressures the draft tube:
You gain power versus nominal but lose efficiency points to the elevated tailwater. Above this the runner starts to cavitate — you'll hear it as a gravelly hiss through the shaft housing.
Result
Nominal shaft power is 17. 5 kW at 5.5 m head and 0.45 m³/s flow. That sits the turbine right in the historical Jonval sweet spot and gives the gristmill comfortable headroom for the line shaft plus modest grid export. Across the operating range you'll see roughly 6.7 kW in late-summer low flow, 17.5 kW at design, and around 20 kW during spring freshet — note the low-flow number is well below a naive half-power estimate because fixed guide vanes don't throttle gracefully. If you measure noticeably less than 17 kW at design conditions, check three things in this order: (1) draft tube submergence — a tube exit too far above tailwater leaves head on the table and is the single most common Jonval retrofit error, (2) guide-vane gap erosion, since silt-driven wear opens the 3-6 mm gap into the 12-15 mm range and bleeds 8-10 efficiency points, and (3) generator coupling alignment, because a misaligned vertical shaft loads the foot-step bearing radially and steals 1-2 kW into bearing heat.
Jonval Turbine vs Alternatives
The Jonval competes against three main alternatives at low-to-medium head: the older Fourneyron radial-outflow turbine, the Francis mixed-flow turbine that eventually replaced it, and the Kaplan propeller turbine for very low head sites. Each has a different sweet spot and different failure mode. Here's how they line up on the dimensions practitioners actually search on.
| Property | Jonval Turbine | Francis Turbine | Kaplan Turbine |
|---|---|---|---|
| Typical head range | 2-12 m | 10-300 m | 1.5-20 m |
| Peak efficiency at design point | 72-75% | 92-95% | 90-93% |
| Part-flow efficiency at 50% Q | ~55% (poor) | ~80% (good) | ~88% (excellent, adjustable blades) |
| Mechanical complexity | Low — fixed vanes, simple runner | Medium — wicket gates, scroll case | High — adjustable runner blades, governor |
| Capital cost (relative) | 1.0× | 2.5-3× | 3.5-4.5× |
| Service life of runner in clean water | 30-60 years (cast iron) | 40-80 years (stainless) | 30-50 years (stainless) |
| Best application fit | Steady-flow low-head heritage and micro-hydro sites | Variable-flow medium-head commercial hydro | Very low head, highly variable flow rivers |
Frequently Asked Questions About Jonval Turbine
The guide vanes are fixed geometry. They were cut for one specific flow and one specific velocity triangle at the runner inlet. When flow drops, the actual approach angle no longer matches the runner blade leading edge, so water hits the blade backs instead of feeding cleanly into the curved face. That mismatch shows up as turbulence at the inlet and residual swirl at the outlet — both of which dump energy as heat instead of shaft torque.
Rule of thumb: below 60% of design flow, expect efficiency to fall faster than flow. This is the single biggest reason mill operators in the 1880s switched to Francis turbines, which use adjustable wicket gates to keep the inlet angle correct across the flow range.
If the site has steady year-round flow and the budget is tight, Jonval wins on capital cost and on visual/historical authenticity for an 1840s-1880s mill. If the flow varies more than 2:1 between seasons, the Francis pays back its higher cost within a few years through better part-flow efficiency.
Quick decision check: graph monthly average flow over a typical year. If 80% of the months sit within ±20% of design flow, Jonval is fine. If you see months at 40% and months at 150% of design, go Francis.
That rumble usually traces to runner blade exit angle gone off-spec, either from cavitation pitting on the trailing edges or from a runner that was re-cast at some point with a sloppy pattern. Residual swirl leaving the runner sets up a rotating pressure field in the draft tube that couples up the shaft as low-frequency vibration.
Diagnostic check: drop a borescope down the draft tube while the unit is shut down and look at the trailing edges of three or four blades. If they show pitting cavities deeper than 2-3 mm, the blade profile is no longer doing its job and you're paying for it in noise and lost efficiency.
A Jonval draft tube only recovers head if it's properly submerged and properly tapered. Measure two things: the water level at the tube exit relative to the tailrace, and the diameter ratio between exit and inlet. Exit must sit at least 200-300 mm below tailwater at minimum operating tailwater elevation, and the tube should diverge at 6-8° per side.
If the exit is at or above tailwater, you're running an open discharge, not a draft tube — you'll lose 10-15% of gross head straight to the tailrace. This is the most common mistake in heritage restorations where someone shortened the original tube during a concrete repair.
Three usual suspects, in order of frequency on field-measured units. First, head measurement — many operators measure gross head at the dam instead of net head at the turbine inlet, ignoring penstock friction losses that can eat 5-8% on a long or undersized supply line. Second, efficiency-curve mismatch — the 72-75% figure is at design flow, and most heritage sites run off-design most of the time. Third, mechanical losses in old line-shaft couplings and bearings can swallow 3-5% before the energy ever reaches the generator.
Get a clean reading: install a pressure tap immediately upstream of the guide vanes, measure flow with a current meter or salt-dilution gauge in the tailrace, and instrument the generator output. Then your 62% number tells you something useful instead of being an aggregate.
Not directly. The part-flow penalty in a Jonval is a hydraulic problem at the guide-vane-to-runner interface, not a generator-speed problem. Slowing the runner with a VFD changes the velocity triangles, but it doesn't fix the fixed-vane geometry mismatch when flow drops. You'll trade one inefficiency for another.
What does help: a two-speed or variable-speed setup combined with a sluice-gate flow controller that keeps the turbine running closer to design Q during low-flow months by actively storing water behind the dam and operating in batch mode. That's how several restored Adirondack heritage micro-hydro sites manage their seasonal flow variation today.
References & Further Reading
- Wikipedia contributors. Water turbine. Wikipedia
Building or designing a mechanism like this?
Explore the precision-engineered motion control hardware used by mechanical engineers, makers, and product designers.
