The Lancaster Turbine is a 19th-century outward-flow reaction water turbine developed for New England textile and grist mills. Water enters a central distributor, passes through fixed guide vanes, and accelerates radially outward through a curved-bucket runner — the change in angular momentum spins the shaft. It was built to extract usable shaft power from low-to-medium-head mill races where a vertical waterwheel was too slow and inefficient. A well-built unit could deliver 60-75% efficiency at heads of 3-8 m, and dozens were installed across Massachusetts and Rhode Island mill towns through the 1860s-1880s.
Lancaster Turbine Interactive Calculator
Vary head and flow to see available hydraulic power in a Lancaster outward-flow reaction turbine.
Equation Used
The worked example uses hydraulic power from head and flow: P_h = rho g Q H. Here H is water head, Q is volumetric flow, rho is water density, and the result is the ideal water power available before turbine losses.
- Clean water with rho = 1000 kg/m^3.
- Gravity g = 9.81 m/s^2.
- Hydraulic power is before runner, bearing, and leakage losses.
- Steady flow through the turbine.
How the Lancaster Turbine Works
The Lancaster Turbine is a Fourneyron-type outward-flow turbine, meaning water moves from the centre of the runner outward to its rim — the opposite of the inward-flow Francis design that eventually replaced it. Water from the headrace drops through a vertical penstock into a central chamber, hits a ring of fixed guide vanes that angle the flow tangentially, and then enters the rotating runner buckets. The buckets curve the flow back toward radial, and the reaction force from that change in momentum drives the shaft. Head and flow rate together set the available power — at 5 m of head and 0.3 m³/s, you have roughly 14.7 kW of hydraulic power before losses.
The geometry is unforgiving. Guide-vane angle and runner-bucket entry angle have to match the flow velocity triangle at the design operating point, otherwise water slams the leading edge of the bucket instead of flowing onto it cleanly. Get the angles within 2-3° of design and you stay near peak efficiency. Drift outside that and efficiency falls off a cliff — a 10° mismatch can drop a 70% turbine to under 50%, and you hear it as a low rumble and feel it as shaft vibration. The other classic failure mode is the runner-to-casing radial gap. Too tight and silt grinds the runner edges. Too loose and water bypasses the buckets entirely; the unit just spins lazily under load.
Unlike an impulse wheel, the Lancaster runs fully submerged and operates on pressure difference across the runner, not on jet impact. That means cavitation is a real concern if tailwater drops or if the runner is mounted too high above the tailrace. Original installations specified the runner sit at or just below tailwater level, and when later operators raised the unit to ease maintenance access, they often destroyed the bucket trailing edges within a season.
Key Components
- Central Distributor / Penstock Inlet: Vertical chamber that delivers water from the headrace down to the runner centre. Cross-section sized so inlet velocity stays under 2.5 m/s — higher velocities cause turbulence that the guide vanes can't straighten out.
- Fixed Guide Vanes: Stationary curved blades arranged in a ring around the runner, angling the flow from radial to tangential. Typical guide-vane exit angle sits between 20° and 30° from tangent. The vane count is usually 1-2 fewer than the runner bucket count to avoid resonant pulsation.
- Runner with Curved Buckets: The rotating element. Water enters at the inner radius and exits at the outer radius, transferring angular momentum to the shaft. Bucket count typically 24-36. Inner-to-outer radius ratio around 0.7 — too small and the flow path is too short to extract energy, too large and bucket curvature gets aggressive and cavitation-prone.
- Vertical Shaft: Transmits torque up through the floor of the mill to bevel gearing or a flat-belt pulley. Forged wrought iron in original builds, 75-150 mm diameter depending on power rating. Runs in a footstep bearing at the bottom and a guide bearing at the floor level.
- Casing and Draft Tube: Encloses the runner and discharges water to the tailrace. The draft tube recovers some kinetic energy by converting velocity head back to pressure head — a properly flared draft tube adds 3-5% to overall efficiency.
- Headrace Gate: Sliding wooden or iron gate that throttles flow into the penstock. Crude flow control — a Lancaster turbine doesn't have variable guide vanes like a modern Kaplan, so part-load efficiency is poor.
Real-World Applications of the Lancaster Turbine
The Lancaster Turbine sat in the gap between low-RPM waterwheels and the high-efficiency Francis turbines that arrived later. It was a workhorse for any mill that needed steady shaft power from a modest head, and you'll still find rusted runners sitting in the basements of converted mill buildings across the northeastern US. Today the design is mostly historical, but micro-hydro restorers and heritage mill projects occasionally rebuild them.
- Textile Manufacturing: Cotton mills along the Blackstone River in Rhode Island used Lancaster Turbines to drive line shafts powering carding and spinning machines through the 1870s.
- Grain Milling: Stone-grinding gristmills in central Massachusetts adopted Lancaster runners to replace overshot wheels when seasonal flow dropped — a turbine could keep working at lower head than a wheel could.
- Heritage Restoration: The Slater Mill historic site in Pawtucket has restored hydraulic prime movers from this era as part of its working museum exhibits.
- Sawmilling: New Hampshire sawmills running circular blades at 400-600 RPM used vertical-shaft turbines like the Lancaster with bevel gearing to step up to blade speed.
- Paper Manufacturing: Early paper mills in the Berkshires drove beater rolls and Fourdrinier machines from Lancaster-type outward-flow turbines until central electric power arrived around 1910.
- Micro-Hydro Restoration: Off-grid heritage property owners occasionally rebuild Lancaster runners for sites with 4-6 m head where the original mill foundation and tailrace are still intact.
The Formula Behind the Lancaster Turbine
What you actually want to know is how much shaft power you can pull out of a given head and flow rate. The hydraulic power equation gives you the theoretical input — multiply by turbine efficiency to get shaft output. At the low end of the typical Lancaster operating range (around 2 m head), available power per m³/s drops to under 20 kW and the unit struggles to overcome bearing friction at part load. At the high end (8-10 m head) cavitation risk climbs and you start getting closer to where a Francis turbine just does the job better. The sweet spot for a Lancaster sits around 4-6 m head with 0.2-0.5 m³/s flow.
Variables
| Symbol | Meaning | Unit (SI) | Unit (Imperial) |
|---|---|---|---|
| Pshaft | Shaft power delivered by the turbine | W | hp |
| η | Overall turbine efficiency (typically 0.60-0.75 for a Lancaster) | dimensionless | dimensionless |
| ρ | Water density | kg/m³ | lb/ft³ |
| g | Gravitational acceleration | m/s² | ft/s² |
| H | Net head (headrace level minus tailrace level, less penstock losses) | m | ft |
| Q | Volumetric flow rate through the turbine | m³/s | ft³/s |
Worked Example: Lancaster Turbine in a heritage gristmill restoration in Vermont
You are sizing a rebuilt Lancaster Turbine for a heritage gristmill restoration on a tributary of the Winooski River in central Vermont. The dam gives 4.5 m of net head at the runner, the stream delivers a steady 0.25 m³/s during the milling season, and the owner wants enough shaft power to drive a single 1.2 m granite millstone at 110 RPM through a bevel gear and lay shaft.
Given
- H = 4.5 m
- Q = 0.25 m³/s
- ρ = 1000 kg/m³
- g = 9.81 m/s²
- η = 0.68 dimensionless (assumed for a well-rebuilt Lancaster)
Solution
Step 1 — compute the hydraulic power available before turbine losses, at the nominal design flow:
Step 2 — apply the assumed efficiency to get nominal shaft power:
That is comfortably above the 4-5 hp a single 1.2 m millstone draws at full grind, so you have headroom for the bevel gearing losses and the lay shaft drag.
Step 3 — check the low end of the operating range. In a dry late summer, flow drops to roughly 0.10 m³/s and head drops to about 3.8 m as the headrace level falls:
Note the efficiency itself drops to roughly 0.60 because the Lancaster has no variable guide vanes — at part flow the velocity triangles no longer match the bucket angles. 3 hp is borderline for actual milling. The stone will turn but you'll feel it bog down on hard wheat.
Step 4 — check the high end. During spring runoff, flow can hit 0.45 m³/s and head climbs to 5.0 m as the headpond fills:
Efficiency drops again because flow exceeds the runner's design point — water spills past the buckets rather than transferring momentum cleanly. You'll also need to throttle the headrace gate, because 19 hp through a wood-toothed bevel gear sized for 10 hp will shred teeth inside an afternoon.
Result
Nominal shaft power is 7. 5 kW (10.1 hp) at 4.5 m head and 0.25 m³/s — plenty for the millstone with margin for drivetrain losses. Across the operating range you get roughly 2.2 kW in late-summer drought, 7.5 kW at design point, and 14.3 kW at spring flood — so the sweet spot sits squarely in the design-flow window, and the owner needs the headrace gate to manage both extremes. If you measure shaft power on a brake test and find significantly less than 7.5 kW at design conditions, the most likely causes are: (1) guide-vane angle off by more than 3° from the runner bucket entry angle, which you'll hear as low-frequency rumble, (2) the runner mounted above tailwater level causing cavitation pitting on the bucket trailing edges, or (3) excessive radial gap between runner rim and casing, typically over 2 mm, letting flow bypass the buckets — easy to confirm with a feeler gauge through the inspection port.
Choosing the Lancaster Turbine: Pros and Cons
The Lancaster sits between waterwheels and modern Francis turbines, and it has well-known weaknesses on either side. Here's how it compares on the dimensions that actually matter when you're choosing a prime mover for a low-head site.
| Property | Lancaster Turbine | Francis Turbine | Overshot Waterwheel |
|---|---|---|---|
| Peak efficiency | 60-75% | 90-94% | 60-70% |
| Part-load efficiency (at 40% flow) | Poor — drops to ~50% | Good — adjustable wicket gates hold ~80% | Fair — naturally tolerant of flow variation |
| Useful head range | 2-10 m | 10-300 m (low-head variants down to 2 m) | 3-12 m |
| Typical shaft RPM | 80-300 RPM | 100-1000 RPM | 4-12 RPM |
| Footprint and civil works | Compact, vertical shaft, fits in mill basement | Compact, requires spiral casing and draft tube | Large, requires headrace flume and exposed wheel pit |
| Cavitation risk | Moderate — runner must sit at tailwater level | Moderate to high at low setting | None — operates in air |
| Build complexity / cost today | High — bespoke runner casting, no off-shelf parts | Moderate — commercial micro-hydro suppliers exist | Low to moderate — wood and steel construction |
| Best application fit | Heritage restoration with 4-6 m head | New micro-hydro install at any head | Rural off-grid site valuing simplicity over efficiency |
Frequently Asked Questions About Lancaster Turbine
Classic symptom of a runner-to-casing gap that's opened up. At no load there's almost no torque demand and the small amount of flow actually crossing the buckets is enough to spin the assembly. Apply load and the buckets need to extract real momentum from the flow, but most of the water is bypassing through the gap and discharging into the tailrace without ever touching a bucket.
Pull an inspection plate and check the radial clearance with a feeler gauge — anything over 2 mm on a runner under 1 m diameter is a problem. Original Lancaster builds ran clearances of 0.5-1.0 mm. The fix is either a re-machined casing wear ring or weld-and-grind buildup on the runner rim.
If the site has historical significance and the original turbine pit, penstock, and tailrace are intact, rebuild the Lancaster — heritage funding bodies will often pay a premium for authenticity, and the original civil works are sized for the runner geometry. If you just want shaft power and the heritage angle is incidental, a crossflow (Banki/Ossberger) is the better engineering choice. Crossflow tolerates flow variation far better, costs a third as much to procure, and sits at 80-85% efficiency across a wide operating band where a Lancaster tops out around 70% at design point only.
Two likely causes, and both relate to thermal behaviour as the assembly warms and seats. The first is footstep bearing wear — the bottom thrust bearing in original Lancaster builds was lignum vitae or bronze, and when it wears, the runner drops a few millimetres, putting the bucket trailing edges into the casing floor on each revolution. You'll feel the thump through the shaft itself.
The second is vane-runner blade-pass resonance. If the guide vane count and runner bucket count share a common factor (say 24 vanes and 36 buckets, both divisible by 12), you get pressure pulsations every 1/12 of a revolution that can excite the shaft's natural frequency once the bearings warm and damping changes. Original designs used coprime counts (e.g. 23 vanes, 36 buckets) to avoid this. If you see matched counts, that's your problem.
Three places eat efficiency on a real Lancaster install. First, draft tube geometry — if the original flared wooden draft tube was replaced with a straight steel pipe during a 20th-century rebuild, you've lost the velocity-head recovery, which is worth 3-5%. Second, penstock friction — original masonry penstocks often have surface roughness equivalent to 5-10 mm, and at design flow that costs another few percent of net head you assumed you had. Third, guide-vane condition — corroded or chipped guide vanes don't deliver the design exit angle, and the velocity-triangle mismatch costs 5-10% on its own.
Measure your actual net head at the runner inlet with a manometer rather than trusting the headrace-to-tailrace static drop. You'll often find 0.5-1 m has vanished into penstock losses you didn't account for.
You need step-up gearing for any synchronous AC generator. A Lancaster runs at 80-300 RPM depending on size and head — a 4-pole 60 Hz synchronous generator needs 1800 RPM, so you're looking at a 6:1 to 20:1 step-up. The traditional approach was a bevel gear at the top of the vertical shaft into a horizontal lay shaft, then a flat-belt drive to the generator pulley.
A modern alternative is a permanent-magnet alternator wound for low RPM — units rated for direct drive at 200-400 RPM exist for the wind turbine market and adapt well to a Lancaster shaft. That eliminates the gear losses (worth 3-5%) and removes a wear point.
Tighter than people expect. A Lancaster runner is typically 400-800 kg of cast iron or bronze running at 100-250 RPM. Static balance to within 50 g·m is the floor — anything looser and you'll feel vibration through the mill floor and accelerate guide-bearing wear. For runners over 600 kg, two-plane dynamic balance is worth the cost; static balance alone leaves a couple force imbalance that shows up as a wobble that gets worse with RPM.
If you're rebuilding an original runner with corroded buckets, weld repair changes the mass distribution. Always rebalance after any bucket repair, and never just patch one bucket — repair them in opposing pairs to keep the imbalance manageable.
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.
