The Prouty-Nooje automatic brake is a centrifugally-engaged friction brake that triggers when a rotating shaft exceeds a preset speed. Spring-loaded shoes mounted on the rotating member fly outward against a fixed drum once centrifugal force overcomes spring preload, clamping the shaft and dissipating kinetic energy as heat. It exists to stop runaway rotation in hoists, lifts, and winches without operator input. A properly tuned unit on a 500 kg freight lift will engage within 0.3 s of overspeed and arrest the load inside one drum revolution.
Prouty-Nooje Automatic Brake Interactive Calculator
Vary shoe mass, radius, shaft speed, spring preload, and lining friction to see centrifugal engagement force and braking torque.
Equation Used
The shoe centrifugal force rises with mass, radius, and the square of rpm. When Fc exceeds the spring preload Fs, the surplus is treated as normal force at the drum, producing friction torque T for two shoes.
- Two identical pivoted shoes are used.
- Shoe CG radius is approximated as the effective friction radius.
- Radial shoe leverage is treated as 1:1.
- Brake torque is zero until centrifugal force exceeds spring preload.
How the Prouty-nooje Automatic Brake Actually Works
The mechanism is dead simple in concept and unforgiving in execution. A pair of pivoted shoes — usually two, sometimes four — sits on a carrier that rotates with the protected shaft. Each shoe is held inward by a calibrated spring. While the shaft turns at normal speed, centrifugal force on the shoe mass is below the spring preload, and the shoes ride clear of a stationary drum machined into the housing. The instant rotation exceeds the trip speed, centrifugal force wins, the shoes swing outward, and friction lining contacts the drum bore. From that moment the brake behaves like any other friction brake — torque equals normal force times coefficient of friction times effective radius — and kinetic energy converts to heat in the drum wall.
Why the centrifugal engagement instead of an electromagnet or hydraulic actuator? Because it needs no power, no signal, no controller, and no operator. It is purely a mechanical fail-safe brake mechanism, and that is exactly why Prouty and Nooje patented it for elevator hoist gear in the late 1800s. The spring-loaded shoe brake responds to the one variable that always indicates runaway — angular velocity — and it does so in milliseconds.
Get the spring rate wrong and you get one of two failure modes. Springs too soft, the shoes drag at normal RPM, the lining glazes, and the drum bore polishes — by the time the shaft actually overspeeds, friction coefficient has dropped from 0.35 to under 0.15 and the brake cannot generate rated torque. Springs too stiff, trip speed climbs above the safe envelope, and the load is already accelerating dangerously by the time the shoes engage. The spring preload tolerance on a typical hoist unit is ±3% of nominal — not 5, not 10. Shoe-to-drum clearance must sit between 0.4 mm and 0.8 mm cold; tighter than that and thermal growth will close the gap during a long descent and self-engage the brake.
Key Components
- Rotating shoe carrier (spider): Bolts directly to the protected shaft and carries the pivoted shoes. Concentricity to the shaft must hold within 0.05 mm TIR or one shoe will engage before the other and twist the drum. Cast iron or forged steel on production units.
- Friction shoes: Pivoted lever arms with a bonded friction lining on the outer face. Lining thickness is typically 8-12 mm with a wear limit at 3 mm remaining. Lining coefficient sits at 0.35-0.40 when fresh, dropping to 0.25 if glazed.
- Calibrated retraction springs: Set the trip speed. On a 1500 RPM hoist motor a typical preload is 80-120 N per shoe with a tolerance of ±3%. Spring relaxation over time is the single most common reason an old unit trips low.
- Stationary drum: Bore that the shoes engage against. Cast iron, ground to a surface finish of Ra 1.6 µm or better. Bore diameter tolerance is H8. Drum mass acts as the heat sink — undersized drums crack from thermal shock during a hard stop.
- Pivot pins and bushings: Locate each shoe on the carrier and let it swing outward. Bushing radial play above 0.15 mm causes shoe chatter and irregular engagement torque, which you'll hear as a buzzing growl during a stop.
Industries That Rely on the Prouty-nooje Automatic Brake
You find the Prouty-Nooje and its direct descendants anywhere a rotating system can run away if a primary control fails. The original target was elevator hoist machinery, but the same centrifugal engagement principle protects mine hoists, theatrical rigging winches, drilling drawworks, and even amusement-ride drive shafts. Wherever an overspeed governor brake is needed and electrical signalling is undesirable, the centrifugal automatic brake earns its place.
- Vertical transportation: Backup overspeed brake on geared elevator machines such as the original Otis traction units that adopted Prouty-style hardware in the 1890s
- Mining hoists: Runaway protection on Koepe and drum hoists where loss of mains power must not allow the cage to descend uncontrolled
- Theatrical rigging: Counterweight runaway brakes on JR Clancy stage rigging systems, engaging if a fly-line load releases unexpectedly
- Oil and gas drilling: Auxiliary brake on cable-tool drilling drawworks similar to Bucyrus-Erie 22W rigs, catching the drum if the main band brake glazes
- Amusement rides: Drive-shaft overspeed brakes on observation wheels and slow-moving dark rides, holding the wheel if the main motor brake releases unintentionally
- Industrial winches: Centrifugal lowering brake inside chain-block hoists like Yale VL series, controlling descent rate of a manually paid-out load
The Formula Behind the Prouty-nooje Automatic Brake
The core sizing question is simple — at what shaft RPM do the shoes leave their seats and engage the drum. That trip speed is set by the balance between centrifugal force on the shoe centre of mass and the retraction spring preload. At the low end of a typical operating range, you might design for a 30% overspeed trip on a slow drum hoist running 60 RPM nominal, where a soft spring and heavy shoe combination dominates. At the high end, a motor-shaft application at 1800 RPM nominal with a 25% trip margin needs a much stiffer spring, and shoe mass must drop to keep the centrifugal force manageable. The sweet spot for most industrial hoist installations is a trip speed 20-40% above rated — tight enough to catch a real fault, loose enough to ignore normal startup transients.
Variables
| Symbol | Meaning | Unit (SI) | Unit (Imperial) |
|---|---|---|---|
| ωtrip | Angular velocity at which shoes leave their seats and engage the drum | rad/s | rad/s |
| Fspring | Retraction spring preload force per shoe at the seated position | N | lbf |
| mshoe | Mass of one friction shoe including lining | kg | lb |
| rcg | Radial distance from shaft centre to shoe centre of gravity in the seated position | m | in |
Worked Example: Prouty-nooje Automatic Brake in a grain elevator bucket-leg head shaft
Specifying the Prouty-Nooje overspeed brake on the head shaft of a 30 m grain elevator bucket leg in a feed mill, similar in scale to a Sukup or GSI commercial leg. The head shaft runs at 80 RPM nominal driving a 600 mm pulley with loaded buckets. You need the brake to trip if a drive belt parts and the loaded down-side allows the shaft to reverse and accelerate. Each shoe weighs 1.4 kg with its centre of gravity 95 mm from the shaft centreline. You want trip at roughly 30% overspeed.
Given
- Nnom = 80 RPM
- Trip margin = 30 %
- mshoe = 1.4 kg
- rcg = 0.095 m
- Number of shoes = 2 —
Solution
Step 1 — convert nominal RPM to angular velocity and apply the 30% trip margin to find the target trip speed:
Step 2 — solve the trip equation for the required spring preload at nominal trip:
That 15.8 N is the seated preload each retraction spring must deliver. A standard compression spring at 25 mm free length with 0.65 N/mm rate, preloaded to 24 mm working length, lands right on it.
Step 3 — at the low end of the typical trip-margin range (15% overspeed, ωtrip = 9.64 rad/s) the required preload drops:
A 15% margin trips at roughly 92 RPM. That is dangerously close to nominal — any normal startup surge or belt-slip recovery will trigger a false engagement, and you will spend your week resetting the brake and replacing glazed lining. Fine for a lab demo, wrong for a production grain leg.
Step 4 — at the high end of the typical range (50% overspeed, ωtrip = 12.57 rad/s):
A 50% margin trips at 120 RPM. By then a fully loaded down-side of bucket cups is already shedding grain back into the boot pit and the shaft is wound up with stored kinetic energy you now have to dump as heat in one drum revolution. Drum surface temperature on engagement jumps roughly with the square of trip speed — going from 30% to 50% margin nearly doubles the heat load on the drum bore.
Result
Nominal preload comes out to 15. 8 N per shoe for a 30% overspeed trip at 104 RPM. In practice that means the brake sits dormant during every normal start and stop, ignores the small shaft surges from buckets entering the boot, and only fires when something has genuinely failed. Compare across the operating range: 15% margin (12.4 N preload, trip at 92 RPM) trips on every nuisance event and burns lining; 50% margin (21.0 N preload, trip at 120 RPM) catches faults so late the drum sees double the heat load. The 25-35% margin band is the sweet spot for industrial hoist and conveyor work. If your installed brake trips too early or too late versus this prediction, check three things first: (1) spring relaxation — old springs lose 5-15% preload over a decade, dropping trip speed proportionally; (2) shoe-pivot bushing wear letting rcg drift outward by 1-2 mm, which raises centrifugal force at any given RPM and lowers trip speed; (3) lining absorption of moisture or oil mist, which adds shoe mass and again lowers the trip point.
Choosing the Prouty-nooje Automatic Brake: Pros and Cons
The Prouty-Nooje is one of three common ways to catch a runaway shaft. Each has a clear place. Compare them on the dimensions that actually matter to a hoist or winch designer.
| Property | Prouty-Nooje centrifugal brake | Spring-applied electromagnetic disc brake | Centrifugal flyweight governor + external brake |
|---|---|---|---|
| Engagement trigger | Shaft overspeed only, fully mechanical | Loss of electrical power | Overspeed sensed mechanically, brake applied separately |
| Response time from trigger to full torque | 50-200 ms | 30-80 ms | 200-500 ms (governor lag plus brake stroke) |
| Holding torque per kg of mechanism | Moderate, 15-40 N·m/kg | High, 40-80 N·m/kg | Low, 8-20 N·m/kg |
| Power required to stay disengaged | None | Continuous coil current, 30-200 W | None for governor; coil power for brake |
| Trip speed accuracy | ±5-8% of setpoint | Not applicable (power-loss trip) | ±2-4% with quality governor |
| Service life before lining replacement | 20-50 hard engagements typical | 1-5 million cycles in light slipping duty | Brake-dependent, governor essentially lifetime |
| Typical installed cost (industrial 5 kW class) | $400-900 | $600-1500 | $1200-2500 |
| Best application fit | Backup overspeed protection on hoists, legs, winches | Primary motor brake on cranes and elevators | Code-mandated elevator car safeties |
Frequently Asked Questions About Prouty-nooje Automatic Brake
You are almost certainly looking at shoe-pivot binding combined with a marginal spring preload. During startup the shaft accelerates fast enough that pivot stiction releases all at once and the shoes flick outward at an effective RPM well below the calculated trip. Once spinning steadily the shoes settle back, and a slow ramp to overspeed never delivers the same impulse, so the brake fails the bench test.
Free the pivots — pull each shoe, polish the pin to 0.4 µm Ra, and check bushing clearance is between 0.05 and 0.10 mm. If pivot drag is acceptable and the false trip persists, your spring preload is sitting within 5% of the centrifugal force at running speed; bump preload up by one shim washer (typically 1-2 N) and re-test.
If the lift carries people or falls under ASME A17.1 or EN 81, you do not have a choice — the code mandates a rope-gripper or rail-clamp car safety triggered by a separate overspeed governor, and a Prouty-Nooje on the hoist drum does not satisfy that requirement. The car safety acts on the rails, not the shaft, so it still works if the rope itself parts.
For freight-only or material-only lifts outside passenger codes, the Prouty-Nooje on the hoist shaft is a legitimate, cheaper option. It catches drive-train runaway but cannot catch a parted rope. Many designers fit both — the centrifugal brake handles drive-side faults, the rail-grip catches a rope failure.
Three causes account for almost every case. First, spring relaxation — coil springs lose preload over years of compression, and a 10% drop in Fspring drops trip speed by about 5% per the square-root relationship in the formula. Pull and load-test the springs against a fresh datasheet curve.
Second, shoe-mass increase from absorbed oil, water, or rust on the carrier. A 5% mass gain shifts trip speed down about 2.5%. Weigh each shoe and compare to drawing.
Third — and the one most people miss — lining swelling pushes rcg outward. New asbestos-free linings can grow 0.3-0.6 mm radially over the first year of service. That is enough to drop trip speed by 3-5% on a small unit.
Yes, and on long-drum hoists this is common practice — but only if you stagger the trip speeds. Set them identical and manufacturing variation guarantees one engages first, takes the full thermal hit, and glazes before the second contributes. The glazed unit then trips later or not at all on the next event.
The standard approach is to set the primary unit to 25% overspeed and the secondary to 35%. The primary handles routine overspeed events alone, and the secondary only joins in during a true runaway. You get redundancy without thermal pile-on.
You are seeing the classic static-versus-dynamic friction gap on a brake that was sized using static coefficient values. Lining manufacturers quote µstatic at around 0.40 but µdynamic during a hot stop drops to 0.20-0.25, and torque scales linearly with µ. Size the brake on dynamic coefficient, never static.
The other contributor is drum thermal expansion. As the drum bore heats during the stop it grows radially faster than the rotating shoe carrier, opening the contact pressure and bleeding torque. On stops longer than 2 seconds this matters. Specify a drum with at least 8 mm wall thickness and you keep transient bore growth under 0.05 mm.
Yes. Build a small bench fixture that drives the shoe carrier alone with a variable-speed motor and a strobe tachometer. Watch for shoe lift-off through a transparent guard or detect it electrically with a thin foil contact strip on the drum bore. Compare the measured lift-off RPM against your calculated ωtrip.
Acceptance band is typically ±5% of nominal trip speed. Outside that band, do not just adjust spring preload to mask the error — find the cause first. A trip speed that drifts in service tells you something physical changed (springs, mass, geometry), and shimming it back into spec without diagnosing the root cause means it will drift again.
Building or designing a mechanism like this?
Explore the precision-engineered motion control hardware used by mechanical engineers, makers, and product designers.
