The Bartrum and Powell rotary engine is a 19th-century single-vane rotary steam engine that converts boiler steam pressure directly into shaft rotation. The defining component is the spring-loaded abutment, a sliding block that rides against an eccentric rotor and divides the working chamber into a high-pressure admission side and a low-pressure exhaust side. It exists to eliminate reciprocating mass — no piston, no crosshead, no crank — so the engine runs smoothly at high speed in a small footprint. In practice it delivered roughly 5 to 25 indicated horsepower at 200 to 600 RPM in workshop and small-launch service.
Bartrum and Powell Rotary Engine Interactive Calculator
Vary eccentricity, bore, speed, and end clearance to see abutment stroke, tracking frequency, and sealing risk in the rotary steam engine.
Equation Used
The abutment is driven in and out by the eccentric rotor. Its total stroke is twice the eccentricity, and it completes one tracking cycle per shaft revolution. End-clearance percentage compares the selected axial clearance with the 0.05 mm leakage warning limit stated in the article.
- Abutment follows the eccentric rotor once per shaft revolution.
- Abutment total in-out travel equals twice the rotor offset.
- Axial leakage risk is referenced to the article limit of 0.05 mm clearance.
- This calculator focuses on geometry and sealing, not full steam thermodynamics.
How the Bartrum and Powell Rotary Engine Actually Works
Steam enters through an admission port just upstream of the abutment, fills the crescent-shaped space between the eccentric rotor and the cylinder bore, and pushes the rotor around. The rotor is mounted off-centre relative to the bore, so the gap between rotor surface and bore wall opens as it rotates past the inlet and closes as it approaches the exhaust port. The abutment — a flat block held against the rotor by a spring or steam pressure behind it — seals the gap between admission and exhaust at all times. Without that seal the engine is just a leaky annulus and produces almost no torque.
The geometry is unforgiving. If the rotor-to-bore eccentricity is set wrong by even 0.5 mm on a 150 mm bore, the swept volume per revolution drops noticeably and the abutment either bottoms out or lifts off the rotor at top dead centre. Vane tip leakage past the abutment is the dominant loss mechanism — get the spring force wrong and you either burn the abutment face from dry contact or blow steam straight from inlet to exhaust. Common failure modes are abutment face wear (showing up as a falling indicated mean effective pressure week over week), spring fatigue leaving the abutment chattering at high RPM, and rotor-end leakage past the side covers if the axial clearance opens beyond about 0.05 mm.
The design is best understood as a rotary expander where one moving partition does the job that piston rings and a slide valve do in a reciprocating engine. That trades the reciprocating-mass problem for a sealing problem, and the sealing problem is the reason rotary steam engines never displaced the crank-and-piston layout in main service.
Key Components
- Eccentric Rotor: A solid cylindrical rotor mounted off-centre on the output shaft, typically with eccentricity of 8 to 15 mm in a 150 mm bore. Its outer surface forms one wall of the working chamber and must be ground concentric to the shaft within 0.02 mm to keep the abutment tracking cleanly.
- Spring-Loaded Abutment: A radial sliding block, usually bronze-faced, held against the rotor by a coil spring or steam pressure on its back face. It seals the chamber between admission and exhaust ports and reciprocates once per revolution as the eccentricity drives it in and out by twice the offset distance.
- Cylinder Bore and Side Covers: The fixed cylindrical chamber housing the rotor, closed at each end by side covers that carry the shaft bearings. Axial clearance between rotor and side covers is held to roughly 0.03 to 0.05 mm — any wider and steam blows past axially regardless of how well the abutment seals.
- Admission Port: An inlet drilling positioned just past the abutment in the direction of rotation, sized to pass the full mass flow at design RPM without choking. Port timing is fixed by geometry — there is no separate valve gear, which is part of the engine's appeal.
- Exhaust Port: An outlet drilling immediately upstream of the abutment so the chamber sweeps to nearly zero volume before the next admission begins. Poor exhaust port placement leaves residual steam in the chamber and drops indicated power directly.
- Output Shaft and Bearings: A through-shaft running in plain or roller bearings in each side cover. Because the rotor is eccentric, the shaft sees a once-per-revolution radial load equal to the net pressure force on the rotor — typically 2 to 8 kN in workshop-sized engines.
Industries That Rely on the Bartrum and Powell Rotary Engine
Bartrum and Powell rotary engines and their close cousins served wherever a small, smooth, compact prime mover was wanted and the operator could tolerate the steam economy penalty that came with the sealing losses. They were never going to drive a mill or a ship, but they fit nicely into niches where a reciprocating engine's bulk and vibration were the bigger problem. You'll find them today in heritage collections rather than working installations, but the underlying layout — eccentric rotor with a sliding abutment — survived into refrigeration compressors and pneumatic motors where the steam-tightness problem doesn't apply.
- Heritage Steam: Working demonstration at the Anson Engine Museum in Poynton, Cheshire, where late-Victorian rotary steam engines run on compressed air for visitor open days.
- Small Steam Launches: Single-vane rotary engines fitted to compact Thames steam launches in the 1880s where vibration-free running mattered for passenger comfort.
- Workshop Drive: Direct-coupled drive for small lathes and bench drilling machines in late-19th-century jobbing shops, replacing belt drives from a central mill engine on light duties.
- Marine Auxiliaries: Auxiliary fan and pump drives on small steam vessels where reciprocating-engine vibration was unacceptable near accommodation spaces.
- Education and Demonstration: Teaching engines at the Powerhouse Museum in Sydney and similar institutions illustrating early rotary expander geometry to engineering students.
- Derived Modern Use: The same eccentric-rotor-and-vane geometry survives in sliding-vane air compressors used by Hydrovane and Mattei in industrial compressed-air service.
The Formula Behind the Bartrum and Powell Rotary Engine
What you actually want to know about a rotary engine like this is its indicated power across the realistic operating range — not just at the nameplate point. At the low end of the typical band, around 150 RPM, the engine runs smoothly but produces little useful work and the abutment spring has plenty of time to follow the rotor. At the nominal design point, around 300 RPM for a workshop-sized unit, you hit the sweet spot where mean effective pressure is still high and mechanical losses haven't yet taken over. Push past 600 RPM and the abutment starts skipping off the rotor face — the spring can't accelerate the block fast enough to follow the eccentricity — and indicated power falls below the geometric prediction.
Variables
| Symbol | Meaning | Unit (SI) | Unit (Imperial) |
|---|---|---|---|
| IHP | Indicated power developed in the working chamber | W | hp |
| Pm | Mean effective pressure across one rotor revolution | Pa | psi |
| Vs | Swept volume per revolution (twice the eccentric crescent area times rotor length) | m3 | in3 |
| N | Rotor speed | rev/min | RPM |
Worked Example: Bartrum and Powell Rotary Engine in a recommissioned 1884 Bartrum and Powell rotary engine
You are confirming indicated power across three shaft speeds on a recommissioned 1884 Bartrum and Powell single-vane rotary engine being returned to demonstration running at the Anson Engine Museum in Poynton, Cheshire, where the engine is bench-mounted and supplied with saturated steam at 5 bar gauge from the museum's package boiler. The rotor is 150 mm diameter, 200 mm long, with 12 mm eccentricity. Mean effective pressure measured from an indicator card averages 3.0 bar across the working stroke.
Given
- Dbore = 174 mm
- Drotor = 150 mm
- e = 12 mm
- L = 200 mm
- Pm = 3.0 bar (300,000 Pa)
- Nnom = 300 RPM
Solution
Step 1 — compute swept volume per revolution. The crescent area between rotor and bore equals the difference of the two circle areas, and the abutment sweeps this volume twice per rotor revolution as the eccentric high spot passes both ports:
Step 2 — at the nominal 300 RPM design point, compute indicated power:
That's the figure you'd quote on the engine's brass plate, and it matches what 1880s Bartrum and Powell engines of this size typically delivered in workshop service.
Step 3 — at the low end of the typical operating range, 150 RPM, the engine produces half the indicated power because IHP scales linearly with speed for a fixed mean effective pressure:
At this speed the engine is barely loafing — you can hear individual abutment events as a soft click, and there's plenty of margin in spring force to keep the abutment glued to the rotor. Useful for demonstration but uneconomic on steam consumption per horsepower.
Step 4 — at the high end, 600 RPM, the geometric prediction is:
In practice you won't see 4.91 hp on the indicator card at 600 RPM. Vane tip leakage rises sharply once abutment chatter begins, mean effective pressure drops to perhaps 2.2 bar instead of 3.0, and real output flattens to around 3.6 hp. Above roughly 700 RPM the abutment lifts off completely on each cycle and indicated power collapses.
Result
Nominal indicated power at 300 RPM is 1,830 W, or about 2. 45 hp — enough to drive a small workshop lathe through a flat belt with steam to spare. Across the operating range you see roughly 1.23 hp at 150 RPM, 2.45 hp at the 300 RPM sweet spot, and a theoretical 4.91 hp at 600 RPM that in practice tops out near 3.6 hp once vane chatter sets in. If your indicator card shows less than the predicted nominal, check three things in order: (1) abutment face wear opening the rotor-to-block clearance beyond 0.05 mm, which drops mean effective pressure directly; (2) side cover axial clearance — anything past 0.05 mm lets steam leak around the rotor ends regardless of how well the abutment seals; and (3) admission port timing relative to the abutment, since a port edge that's been re-machined off-position by even 2 mm shifts the effective cutoff and cuts swept-volume utilisation.
Bartrum and Powell Rotary Engine vs Alternatives
The rotary layout solved one real problem — reciprocating mass and the vibration that comes with it — and created another, namely steady-state steam tightness. Compared to the conventional crank engines of its day and the modern derivatives that survived, the Bartrum and Powell sits in a narrow niche on engineering merit.
| Property | Bartrum and Powell Rotary | Single-Cylinder Reciprocating Steam Engine | Sliding-Vane Air Compressor (modern derivative) |
|---|---|---|---|
| Typical operating speed | 150–600 RPM | 60–300 RPM | 1,000–3,000 RPM |
| Indicated thermal efficiency | 8–12% | 12–18% | Not applicable (driven, not driving) |
| Vibration at the foundation | Very low — no reciprocating mass | High — reciprocating piston and crosshead | Very low |
| Sealing reliability over 1,000 hours | Poor — abutment and end-face wear dominate | Excellent — piston rings are mature technology | Good — oil-flooded vane tips |
| Footprint per indicated horsepower | Small — about 60% of a reciprocating engine | Large | Smallest |
| Capital cost (period equivalent) | Higher — precision bore and abutment fitting | Lower — well-established machining | Mass-produced today |
| Best application fit | Small smooth-running auxiliaries, demonstration | Mill drive, locomotive, marine main engine | Industrial compressed air |
Frequently Asked Questions About Bartrum and Powell Rotary Engine
This is almost always thermal growth opening the side-cover axial clearance. The cast iron rotor and the cover plates expand at slightly different rates as steam temperature soaks through the casing, and a clearance set cold at 0.04 mm can open to 0.08–0.10 mm at running temperature. Steam then bypasses the abutment seal axially around the rotor ends, mean effective pressure drops, and indicated power follows.
The fix is to set cold clearance based on a hot-running measurement — run the engine for 45 minutes, shut down, immediately measure end float, and shim the cold setting to land at 0.03–0.04 mm hot.
Single-vane wins on simplicity and authenticity for heritage work — one moving partition, one spring, one set of port timings. It gives you a torque pulse once per revolution which is fine for flywheel-buffered loads but rough for direct drives.
Multi-vane (4 or 6 vanes) smooths the torque output and roughly doubles the indicated power for the same bore and length, but each vane is another sealing interface and another wear path. For a demonstration engine that runs maybe 50 hours a year, single-vane is the right call. For anything seeing daily duty, multi-vane is more forgiving of wear — at the cost of historical accuracy.
That spike is the admission port opening too late relative to the abutment passing top dead centre. Steam admission is supposed to begin the instant the abutment clears the inlet edge so the expanding chamber fills smoothly. If the port edge is a couple of millimetres late, the chamber starts expanding into a near-vacuum, then the port suddenly opens and dumps high-pressure steam in — that's your spike.
Check the angular position of the admission port edge relative to the abutment slot. On a 174 mm bore, every 1 mm of arc at the bore equals roughly 0.66° of rotor angle, which is enough timing error to show clearly on the card.
Stiffer springs make the chatter worse, not better, past a certain point. The abutment has to follow a sinusoidal radial displacement once per revolution, and the peak inward acceleration scales with N2. A stiffer spring raises the natural frequency of the abutment-spring system but also raises the peak contact force on the rotor, accelerating bronze face wear.
The correct fix is to reduce abutment mass — drilling the back face or switching to a lighter bronze grade — rather than stiffening the spring. Some late designs added steam pressure behind the abutment (a balance port from the inlet) so the abutment is pressure-loaded, not just spring-loaded, which scales naturally with operating conditions.
Roughly a 35% shortfall like that almost always points to mean effective pressure being lower than assumed, not to a swept-volume error. The two most common culprits at this magnitude are wet steam (water carryover from the boiler dropping the effective expansion ratio) and a partially blocked admission port restricting mass flow at speed.
Quick diagnostic: take an indicator card. If the admission line is rounded and the peak pressure sits well below boiler pressure, you're throttled at the inlet — clean the port and check the steam stop valve isn't half shut. If the admission line is sharp but the expansion curve sags early, you have wet steam and need to check the boiler water level and any superheat coil.
Yes, and most museum runners do exactly this — the Anson Engine Museum and several others run their rotary engines on shop air at 4–6 bar. The engine doesn't care whether the working fluid is steam or air at the same pressure, but two things change.
First, there's no condensation lubrication, so you must inject a fine oil mist into the air supply or the abutment face will gall within hours. Second, air leakage past the abutment is more obvious — you'll hear it where steam would silently condense — which actually makes air a useful diagnostic fluid for finding sealing problems before a steam test.
References & Further Reading
- Wikipedia contributors. Rotary engine. Wikipedia
Building or designing a mechanism like this?
Explore the precision-engineered motion control hardware used by mechanical engineers, makers, and product designers.
