A Stirling Engine Beta is a closed-cycle external combustion engine in which a displacer and a power piston operate inside the same cylinder, with the displacer shuttling the working gas between a hot end and a cold end through a regenerator. Robert Stirling patented the original air engine in 1816, and the beta arrangement became the form Philips refined for its 1950s portable generator program. The displacer leads the power piston by roughly 90°, producing pressure swings that drive the flywheel. Modern beta units run on propane, biomass, or solar heat and reach 25-35% thermal efficiency in cogeneration applications.
Stirling Engine Beta Interactive Calculator
Vary phase angle, dead volume, regenerator recovery, and charge pressure to see the beta Stirling timing and relative output change.
Equation Used
This calculator normalizes the beta Stirling mechanism around the article baseline: 90 deg displacer lead, 90% regenerator recovery, 30% dead-volume limit, and 30 bar charge pressure. The sine phase term rewards the correct 90 deg relationship, the pressure term follows the article note that higher charge pressure raises specific power roughly linearly, and the dead-volume factor penalizes values above 30%.
- Output is a normalized teaching index, not a full Schmidt-cycle power prediction.
- Charge pressure raises specific power approximately linearly over the practical range shown.
- The displacer leads the power piston; 90 deg phase gives the maximum phase factor.
- Dead volume above 30% is penalized because the article notes pressure ratio and power collapse.
The Stirling Engine Beta in Action
A beta Stirling shares one cylinder between two pistons. The displacer sits at the hot end, has loose clearance to the bore, and does almost no sealing work — its only job is to shove the working gas (usually helium or hydrogen, sometimes air) back and forth between the heater head and the cooler. The power piston sits below it on the cold side, fits the bore tightly, and converts the resulting pressure swings into rotation through a crankshaft. Both pistons drive the same crank but are phased about 90° apart, with the displacer leading. That phase angle is what makes the cycle work — set it wrong and the engine either stalls or runs backward.
Gas heated at the hot end expands and pushes the power piston down. The displacer then sweeps that gas through the regenerator matrix — a fine wire mesh or stacked screens that stores heat from the gas as it passes — and into the cold space. Cooled gas contracts, the power piston rises on flywheel inertia, and the displacer pushes the gas back through the regenerator, picking up the stored heat on its way to the hot end. The regenerator is the single component that separates a real Stirling from a toy. Without it you waste 60-70% of the heat input. With a properly designed mesh — typical porosity 65-75%, wire diameter 40-80 µm — the regenerator can recycle 90%+ of the heat that crosses it each stroke.
Tolerances matter more than most builders expect. Displacer-to-bore clearance typically runs 0.3-0.6 mm on a 60 mm bore — too tight and thermal growth seizes the displacer at the hot end, too loose and you bypass the regenerator entirely. Dead volume (the gas that never gets swept) should sit below 30% of total swept volume; above that the Schmidt analysis pressure ratio collapses and indicated power drops by half. Common failure modes are heater-head cracking from thermal cycling on Inconel 625 or 310SS heads, regenerator clogging from oil migration past the power piston, and crankshaft phase drift from a slipped displacer rod connection.
Key Components
- Displacer piston: Loose-fitting piston at the hot end of the cylinder that shuttles working gas between hot and cold spaces. Typical clearance 0.3-0.6 mm on a 60 mm bore. It must be lightweight (often hollow stainless or ceramic) because it reverses direction every stroke and a heavy displacer wastes power on its own inertia.
- Power piston: Tightly sealed piston at the cold end that extracts work from the pressure swings. Typically uses a graphite-filled PTFE ring or precision lapped fit with 5-15 µm clearance. Any blow-by here directly cuts indicated power and contaminates the regenerator with seal debris.
- Regenerator: Porous matrix between hot and cold spaces that stores and releases heat each stroke. Stainless wire mesh is standard with 65-75% porosity and 40-80 µm wire. A well-designed regenerator recovers 90%+ of the heat passing through it; a poor one drops cycle efficiency by 20 percentage points.
- Heater head: External heat exchanger at the hot end, often Inconel 625 or 310 stainless tubes for propane firing, copper for solar dish use. Must survive 700-800°C continuous with daily thermal cycling. Wall thinning from creep is the typical end-of-life mode after 20,000+ hours.
- Cooler: Water- or air-cooled jacket around the cold space that rejects waste heat. Cold-end temperature directly limits Carnot efficiency, so dropping cooling water from 60°C to 30°C can lift output by 8-10%.
- Crankshaft and rhombic drive: Mechanism that maintains the 90° phase between displacer and power piston. Philips popularised the rhombic drive specifically for beta engines because it cancels lateral forces on the displacer rod and keeps gas seals working at the cold end. A standard slider-crank works but loads the rod seal harder.
- Working gas charge: Pressurised helium (typical 30-150 bar) or hydrogen for high-output units, air for educational models. Higher charge pressure raises specific power roughly linearly until heat exchanger limits cap it. Hydrogen leaks aggressively and embrittles steel — design the seal package accordingly.
Where the Stirling Engine Beta Is Used
Beta Stirlings show up wherever the heat source is awkward, dirty, or remote — anywhere you would rather burn the fuel outside the engine than inside it. They tolerate biomass, solar concentration, propane, and waste heat without redesign because the working gas never touches combustion products. That makes them the natural fit for off-grid power, marine auxiliary, residential cogeneration, and any application where vibration and noise must stay low. Their weakness is start time (10-15 minutes to full power) and specific cost per kW, so they rarely compete in markets where a diesel genset starts in 5 seconds and costs a third as much.
- Submarine propulsion: The Swedish Kockums Gotland-class submarines run two 75 kW Stirling AIP units that let them stay submerged for weeks on liquid oxygen and diesel fuel.
- Residential cogeneration: The Microgen / Inspirit Energy free-piston beta unit produces 1 kW electrical and 6 kW thermal from a natural gas burner, fitted into UK domestic combi boilers.
- Solar dish power: The Stirling Energy Systems / Sandia SunCatcher used a 25 kW Kockums-derived four-cylinder Stirling at the focus of a parabolic dish, hitting 31% solar-to-grid efficiency in 2008 testing at Sandia National Labs.
- Off-grid telecom power: Qnergy PowerGen 1200 and 5650 free-piston beta Stirlings run unattended on natural gas or propane at remote pipeline and cellular sites in northern Canada and Alaska.
- Cryocoolers (reverse cycle): Sunpower CryoTel beta-configuration free-piston Stirlings cool infrared sensors and HTS magnets to 60-80 K aboard satellites and military thermal imagers.
- Biomass CHP: ÖkoFEN's pellematic Smart_e wood-pellet boiler uses a Microgen beta Stirling to generate 600 W electrical alongside 9 kW thermal for European homes.
- Educational and hobby: The Böhm HB11 and HB13 beta engines, machined in Germany, are the reference benchtop demonstrators used in mechanical engineering courses worldwide.
The Formula Behind the Stirling Engine Beta
The Schmidt analysis gives the closed-form indicated power of a beta Stirling assuming sinusoidal volume variations and isothermal hot and cold spaces. It is the first calculation any designer runs because it tells you whether your bore, stroke, charge pressure, and speed combination has any hope of producing the target output. At the low end of typical operating speeds (around 300 RPM) you get smooth running but specific power well below capability. At the nominal design point (roughly 1500 RPM for a 1-5 kW engine) Schmidt power scales linearly with speed. Push past 2500-3000 RPM and the heat exchangers can't move heat fast enough — actual power falls below Schmidt prediction by 30-50% because the gas never reaches the hot wall temperature each cycle.
Variables
| Symbol | Meaning | Unit (SI) | Unit (Imperial) |
|---|---|---|---|
| Pi | Indicated power output | W | ft·lbf/s |
| pmean | Mean cycle pressure of working gas | Pa | psi |
| Vswept | Power piston swept volume | m³ | in³ |
| f | Engine frequency (cycles per second) | Hz | Hz |
| τ | Temperature ratio Thot / Tcold (absolute) | dimensionless | dimensionless |
| δ | Schmidt pressure-amplitude factor (function of —, swept-volume ratio and dead volume) | dimensionless | dimensionless |
| θ | Phase angle between displacer and power piston | rad or ° | rad or ° |
| Xd | Reduced dead volume (dead volume / swept volume, temperature-weighted) | dimensionless | dimensionless |
Worked Example: Stirling Engine Beta in a 3 kW biomass-fired beta Stirling for a Patagonian sheep station
A sheep station 180 km south of Coyhaique in Chilean Patagonia is replacing a worn-out 6 kW diesel genset with a wood-fired beta Stirling that burns lenga firewood the station already cuts for heating. You are sizing a single-cylinder beta unit with a 90 mm bore, 60 mm stroke power piston, charged with helium at 50 bar mean pressure, hot-end 700°C, cold-end 40°C, displacer phase 90°, dead-volume ratio Xd = 0.7, and a target nominal speed of 1200 RPM. You need to know the indicated power at the low (600 RPM), nominal (1200 RPM), and high (2000 RPM) ends of the planned operating range to confirm the burner and heat-exchanger sizing.
Given
- Bore = 90 mm
- Stroke = 60 mm
- pmean = 50 bar (5 × 10⁶ Pa)
- Thot = 973 K (700°C)
- Tcold = 313 K (40°C)
- θ = 90 °
- Xd = 0.7 dimensionless
- Speed range = 600 / 1200 / 2000 RPM
Solution
Step 1 — compute swept volume and temperature ratio:
τ = 973 / 313 = 3.11
Step 2 — estimate the Schmidt pressure-amplitude factor δ. For τ = 3.11, swept-volume ratio ≈ 1.0 and Xd = 0.7, standard Schmidt charts give δ ≈ 0.32. The denominator (τ + 1 + 2Xd) = 3.11 + 1 + 1.4 = 5.51.
KSchmidt = (5 × 106) × (3.82 × 10−4) × (2.11) × (0.32) × 1 / 5.51
KSchmidt ≈ 234 J/cycle
Step 3 — nominal indicated power at 1200 RPM (f = 20 Hz):
After mechanical losses (rhombic drive friction, seal drag, cooling fan) you typically lose 30-35% of indicated power, so brake output lands near 3.0-3.3 kW — exactly the design target. Step 4 — low end at 600 RPM (f = 10 Hz):
At 600 RPM the engine runs almost silently and the burner can be turned down to 12 kW thermal input. This is the mode you'd use overnight when the station only needs lighting and refrigeration. Step 5 — high end at 2000 RPM (f = 33.3 Hz), theoretical Schmidt:
In practice you will not see anywhere near 7.8 kW. Above roughly 1600 RPM the heater-head tubes can't transfer heat fast enough to keep the working gas at 973 K through the expansion phase, so τ effectively drops to 2.6-2.7 and δ collapses. Real brake power tops out around 3.8-4.0 kW and the engine runs hot and rough — the heater head glows visibly brighter because residence time on the cold side falls below what the cooler can handle.
Result
At the nominal 1200 RPM design point the Schmidt analysis predicts 4. 7 kW indicated and roughly 3.0-3.3 kW at the shaft, which lines up with the 3 kW genset target. At 600 RPM you get about 1.5 kW brake — quiet, fuel-sipping, ideal for overnight loads — while pushing to 2000 RPM only buys you 3.8-4.0 kW because the heater head becomes the bottleneck, not the cylinder. If your measured brake power comes in 25%+ below this prediction, the three failure modes to chase first are: (1) phase angle drift from a loose displacer-rod yoke shifting θ from 90° to 75-80°, which directly cuts sin(θ) and pulls 6-8% off output; (2) regenerator flow leakage through an oversized displacer-bore clearance above 0.7 mm, which lets gas bypass the matrix and drops effective τ; and (3) helium charge loss past a degraded crankcase O-ring, since dropping pmean from 50 bar to 35 bar cuts indicated power proportionally — almost a third gone before you notice the gauge reading.
Stirling Engine Beta vs Alternatives
The beta is one of three Stirling layouts (alpha, beta, gamma). Picking between them and against a small diesel genset comes down to packaging, efficiency, and how dirty your fuel is. Here is how a beta stacks up against the alpha configuration and a comparable diesel.
| Property | Beta Stirling | Alpha Stirling | Small Diesel Genset |
|---|---|---|---|
| Typical operating speed | 500-3000 RPM | 500-2000 RPM | 1500-3600 RPM |
| Thermal efficiency (shaft) | 25-35% | 25-32% | 30-40% |
| Specific cost (USD/kW, 1-10 kW class) | $3,000-6,000 | $3,500-7,000 | $400-900 |
| Cold-start time to rated power | 8-15 min | 8-15 min | 5-30 sec |
| Maintenance interval (hours between service) | 5,000-8,000 | 4,000-7,000 | 250-500 |
| Fuel flexibility | Any external heat (wood, gas, solar, waste) | Any external heat | Diesel/biodiesel only |
| Typical service life (hours) | 40,000-80,000 | 30,000-60,000 | 10,000-20,000 |
| Noise at 1 m | 55-65 dBA | 60-70 dBA | 85-100 dBA |
| Mechanical complexity | Single cylinder, two pistons, rhombic or yoke drive | Two cylinders, two cranks, hot connecting duct | Single cylinder, valves, injection, cooling |
Frequently Asked Questions About Stirling Engine Beta
Because the cycle is symmetric in time. The displacer leads the power piston by 90° in one direction of rotation and lags by 90° in the other. Both directions produce a valid thermodynamic cycle as long as one end is hot and the other cold — there is no valve timing or ignition timing that picks a preferred direction the way a diesel has.
Builders sometimes mark the flywheel with an arrow and forget which way it goes after a rebuild. If yours runs backward, just stop it and spin it the other way. If it runs equally poorly in both directions, your phase angle has drifted away from 90° in both senses, which usually points to a slipped displacer-yoke pin.
Hydrogen has roughly 30% better heat transfer and lower viscous loss than helium at the same pressure, so the same engine produces 15-25% more power on hydrogen. Philips, Stirling Thermal Motors, and Kockums all chose hydrogen for that reason in production hardware.
The cost is sealing. Hydrogen permeates through standard elastomers, embrittles ferritic steels, and presents a flammability risk if a heater head cracks. For a remote off-grid installation with no service crew, helium is the right answer — slightly less power, far fewer surprises. Pressurise to 60-80 bar instead of 50 to recover most of the lost specific power.
Schmidt assumes isothermal hot and cold spaces, perfect regeneration, and no flow loss. Real engines lose efficiency in roughly this order: regenerator imperfection (5-8 points), heat exchanger pinch (3-5 points), flow friction through the regenerator and connecting ducts (2-4 points), and mechanical friction (2-3 points). Add those up and a 10-point gap from Schmidt to measured is normal.
If your gap is bigger than 12 points, the regenerator is the prime suspect. Pull it and inspect — oil migration from the power piston coats the wire mesh and kills both heat capacity and flow area. A black, sticky regenerator after 500 hours means your power piston seal is shedding lubricant into the working space.
The rhombic drive, patented by Philips engineer R. J. Meijer in the 1950s, uses two counter-rotating crankshafts geared together and a yoke linkage to drive both the displacer rod and the power piston coaxially. It cancels lateral side loads on the displacer rod, which means the rod-to-cylinder seal sees only axial motion and lasts an order of magnitude longer.
A slider-crank works fine for an air-charged demo engine running at 1 bar. For a pressurised helium engine at 50 bar, side load on the displacer rod will eat through a PTFE rod seal in under 200 hours. If you do not want to machine a rhombic, a single-crank yoke (Ross yoke) gives most of the benefit with about half the parts.
No, and this trips up a lot of first-time builders. Power scales with volume (length cubed) but heat transfer scales with surface area (length squared). A geometrically scaled engine becomes heat-transfer-limited as it gets larger — the heater head can't push enough watts into the gas, so τ collapses and efficiency drops.
Real scale-ups add finned or tubular heater heads with 5-20× the surface area of a simple cup, pressurise the working gas to 30-100 bar to raise specific power without bigger volume, and drop charge gas to helium or hydrogen for better convective coefficients. Skip those steps and your scaled-up engine produces less power than the original benchtop unit.
Almost always cold-end heat rejection. As the cooler jacket warms up, Tcold rises from 30°C to 70°C, which drops τ from 3.2 to 2.7 and pulls indicated power down by 25-30%. Schmidt is brutal about this — every 10 K rise on the cold side costs you real output.
Check coolant flow rate and inlet temperature. A blocked thermostat, undersized radiator, or marginal fan are the typical culprits. Rule of thumb: size your cooler for 2.5× the rated brake power in heat rejection, not 1.5× — a Stirling rejects much more low-grade heat than a diesel of the same output.
0.3-0.5 mm radial clearance is the working window for a steel-on-steel pairing running 600-700°C hot end. Below 0.3 mm and thermal expansion of the displacer at full operating temperature will close the gap and seize — most builders learn this the hard way after an engine that ran fine cold locks up after 10 minutes. Above 0.6 mm and you start bypassing the regenerator, dropping efficiency 3-5 points per extra 0.1 mm.
If you machine a ceramic or graphite displacer the working window tightens to 0.15-0.25 mm because thermal growth is much smaller. Always set clearance with the engine assembled cold and verify by spinning it by hand through one full revolution at room temperature with no resistance.
References & Further Reading
- Wikipedia contributors. Stirling engine. Wikipedia
Building or designing a mechanism like this?
Explore the precision-engineered motion control hardware used by mechanical engineers, makers, and product designers.
