Clutch & Brake

Clutches and brakes are the on-off and stop-hold elements of every power transmission line. A clutch couples a driving shaft to a driven shaft so torque can be engaged and released on command; a brake couples a rotating member to a fixed frame so kinetic energy can be dissipated, stopped, or held. Mechanically the two are near twins, sharing the same friction surfaces, actuators, and torque equation, which is why they are catalogued, sized, and often packaged together as a single clutch-brake unit.

This guide treats clutch and brake as one engineering family. It covers the working principle, the major actuation types from electromagnetic friction to magnetic particle and oil-shear, the friction materials that set the temperature ceiling, the torque-sizing math, the spec-sheet parameters that drive selection, and the decision sequence procurement engineers can reuse as an RFQ template.

Cutaway of an Ogura VCEH electromagnetic actuated clutch showing the steel housing, mounting bolts, central bore hub, and the copper field coil exposed in cross-section

Photo: Oguraclutch, CC BY-SA 3.0, via Wikimedia Commons

This guide is written for industrial purchasing engineers and design engineers. It covers 6 chapters spanning what a clutch and brake is, actuation types, sensing and engagement principles, friction materials and standards, key specification parameters, and the selection decision sequence, with 7 selection FAQs and manufacturer comparisons. Parameters and designations reference public engineering standards including DIN 15435 and FEM 1.001 for hoist holding brakes, DIN VDE 0580 for electromagnetic devices, EN 14492-2 for power-driven winches and hoists, and IEC 61911 (the international equivalent of DIN VDE 0580) and AGMA service-factor practice for related drivetrain sizing.

Chapter 1 / 06

What is a Clutch and Brake

A clutch is a mechanical device that engages and disengages the transmission of torque between two coaxial rotating members, typically a driving shaft connected to a continuously running motor and a driven shaft connected to a load. A brake is the same device with one member grounded to a stationary frame, so instead of transferring torque it absorbs kinetic energy and converts it to heat, bringing a load to rest or holding it in place. Because both rely on the same physics, the same friction couple, the same actuator, and the same torque relationship T equals mu times N times r times the number of friction surfaces, the two are treated as a single product family by every major manufacturer.

The reason a clutch and a brake so often appear together is the way industrial machines are powered. A single AC motor typically runs without interruption to avoid the heavy starting current and thermal stress of repeated restarts. A clutch then connects that running motor to the work shaft when motion is wanted, and a brake stops the work shaft when motion must end, all while the motor keeps spinning. Packaging the two into one clutch-brake unit, where the clutch engages as the brake releases and vice versa, gives clean cyclic start-stop control on presses, packaging machines, conveyors, and indexing tables.

A typical electromagnetic friction unit is built from four functional parts: the coil or field, wound from copper or aluminum magnet wire inside a carbon-steel shell; the rotor or input member, which carries one friction face; the armature or output member, a steel plate that is pulled into contact when the field energizes; and the hub that locates the assembly on the shaft. Friction material is bonded or mounted flush with the steel faces to set the torque and wear characteristics, although some mobile units run steel on steel and rely on magnetic attraction alone.

The engineering history of friction couplings runs alongside the machine age itself. Cone and band clutches appeared on early line-shaft mills; the disc clutch and the dry plate friction clutch matured with the automobile in the early twentieth century. Industrial electromagnetic clutches and brakes, energized by a DC coil rather than a foot pedal or lever, were commercialized through the mid twentieth century by makers such as Warner Electric, Ogura, and Miki Pulley, and remain the workhorse of automated machinery. Magnetic particle devices for proportional tension control and oil-shear units for high-cycle wet operation followed as process control demands grew.

Four engineering metrics dominate clutch and brake quality across all these types: static and dynamic torque, response time, thermal capacity expressed as energy per engagement and allowable cycle rate, and service life measured in millions of cycles or hours of slip. These four together set the total cost of ownership. A unit that is cheap to buy but undersized in thermal capacity will overheat its friction facing, fade in torque, and wear out prematurely, so the lowest purchase price rarely yields the lowest lifecycle cost.

Chapter 2 / 06

Actuation and Operating Types

Clutches and brakes are first classified by how they are actuated, because the actuation method determines wiring, fail behavior, and controllability. The four mainstream actuation methods are electric (electromagnetic), pneumatic, hydraulic, and mechanical. A second, safety-critical classification splits brakes into power-off and power-on logic. The table below summarizes the operating types you will meet on a typical RFQ.

Operating TypeActuationEngaged StateTypical Use
Power-off (spring-applied) brakeElectric coil release, spring applyDe-energizedServo holding, cranes, hoists, vertical axes
Power-on electromagnetic clutchElectric coil applyEnergizedIndexing, conveyors, packaging start-stop
Pneumatic clutch-brakeAir apply or air releaseAir pressurePower presses, metal forming, shears
Hydraulic clutch-brakeOil pressureHydraulic pressureHeavy mobile, marine, mill drives
Wrap-spring clutchMechanical (coiled spring grip)Spring wrapSingle-revolution feeders, low-cost indexing
Tooth (positive) clutchElectric or mechanicalTooth engagementHigh-torque, no-slip, synchronized at low rpm

Power-off, spring-applied brakes are the safety default. Compression springs clamp the friction rotor when the coil is de-energized; energizing the coil pulls the armature back against the springs and frees the rotor to turn. Because loss of power, a broken wire, or a coil failure all result in the brake clamping, these units hold servo motor axes, robot joints, elevators, cranes, and hoists. They are the brakes referenced by DIN 15435, FEM 1.001, and EN 14492-2, which mandate fail-safe spring application, a manual release lever, and microswitch monitoring of armature position for lifting duty.

Power-on units require the coil to be energized to produce torque and free-wheel when de-energized. They are well suited to cyclic on-command starting and stopping of indexing tables, feed rolls, and conveyors, where free rotation in the rest state is desirable and the load cannot run away. They must never be relied on to hold a suspended load during a power outage.

Pneumatic and hydraulic clutch-brake units dominate heavy duty. Pneumatic units typically operate at about 5.5 to 6.0 bar (80 to 87 psi) and reach torque ratings from roughly 20 Nm to well over 100,000 Nm; press combination units from Goizper, for example, span clutch torque from about 65 Nm to 150,000 Nm. Hydraulic actuation packs even higher force into a small envelope for mill, marine, and mobile drives. Wrap-spring and tooth clutches round out the family: wrap-spring units give cheap, precise single-revolution indexing, while tooth or jaw clutches transmit large torque with zero slip but must be engaged only at rest or below about 20 rpm.

Chapter 3 / 06

Engagement Principles and Technologies

Beneath the actuation method sits the physical principle that develops torque. Four engagement technologies cover almost all industrial clutches and brakes: dry friction (single-face and multi-disc), magnetic particle, hysteresis, and oil-shear. Each has a distinct torque-versus-speed character, controllability, and thermal envelope, so there is no universal technology. The table below compares them on the metrics that matter at selection.

TechnologyTorque vs. SlipControllabilityWearTypical Use
Dry friction (single-face / multi-disc)Drops with speedOn / offFacing wearsStart-stop, holding, presses
Magnetic particleFlat vs. slipProportional to currentPowder, sealedWeb tension control, test stands
HysteresisFlat vs. slipProportional to currentNo contact wearPrecision tension, low torque
Oil-shearViscous, smoothOn / off, high cycleVery lowHigh-cycle stop, cyclic packaging

Dry friction electromagnetic is the most common principle. When the DC coil energizes, magnetic flux pulls the steel armature into the rotor, and friction between the mating faces transmits torque. Output torque is nearly linear with coil voltage and current in a DC unit, which lets a partial voltage trim torque, although operating below rated voltage reduces holding capacity. Torque falls roughly 8 percent for every 20 degrees Celsius rise in coil temperature, so a hot coil holds less than a cold one. A double-flux rotor adds about 30 to 50 percent torque over a single-flux design, and a triple-flux design adds 40 to 90 percent, by routing the magnetic circuit through more friction faces.

A critical practical effect of dry friction units is burnishing. Although faces are machined and lapped flat, microscopic peaks remain, so a new unit contacts only on those peaks and can deliver as little as 50 percent of its rated static torque out of the box. Cycling the unit 20 to more than 100 times at reduced inertia or speed wears the peaks down, seats the contact area, and brings torque up to specification. Burnishing must be planned into commissioning whenever application torque is close to the catalog rating.

Magnetic particle devices fill the gap between input and output rotors with fine magnetizable powder. Energizing the coil chains the particles into a torque-carrying bridge whose strength is nearly proportional to coil current and, crucially, independent of slip speed. That makes them ideal for web tension control on winders and unwinders, where torque must stay constant as roll diameter and line speed change, and for dynamometer and test-stand loading. They can run in continuous controlled slip, and the sealed powder produces no wear debris, though the torque-current curve shows hysteresis, so the same current yields slightly different torque on rising versus falling current.

Hysteresis units develop torque through magnetic hysteresis in a rotor passing through a field, with no mechanical contact across the working gap. Torque is again proportional to current and flat versus slip speed, but with even smoother output, no wear particles, and long life, which suits precision low-torque tensioning and test work. Oil-shear clutch-brake units run the friction plates immersed in a film of transmission fluid that both transmits torque by viscous shear and carries heat away, giving extremely high cycle rates, low wear, and smooth, quiet engagement on packaging and high-speed indexing machinery.

Chapter 4 / 06

Friction Materials and Standards

For every dry friction clutch and brake, the friction material sets the coefficient of friction, the allowable surface pressure, the temperature ceiling, and the wear life. A typical friction couple pairs a steel mating face against a high-friction lining that is molded, woven, sintered, or solid. The lining family chosen is the single biggest lever on thermal capacity, so it must be matched to the energy per engagement and the cycle rate, not just to the static torque.

Organic and molded linings use resin-bonded fibers and fillers. They are quiet, cheap, and have a moderate, stable coefficient of friction, but they fade and degrade well below 250 degrees Celsius, limiting them to light and moderate duty. Sintered metal linings are powder-metal friction pads fused under heat and pressure; they tolerate far higher interface temperatures, carry high surface pressure, and resist galling, which is why press and crane brakes favor them. Reported dynamic friction coefficients for sintered materials run from about 0.06 to 0.14 across roughly 2,000 to 11,000 ft/min sliding speed and about 50 to 680 psi of face pressure. Ceramic and carbon linings push the temperature ceiling highest, with some ceramic facings rated to operate without fading near 540 degrees Celsius (about 1,000 degrees Fahrenheit), at the cost of an abrupt, high static-to-dynamic friction ratio that makes engagement harsher.

The defining limitation of all dry friction units is heat. Every engagement converts kinetic energy plus slip energy into heat at the interface; energy per engagement multiplied by cycles per minute gives the thermal power the unit must dissipate by convection and conduction. Exceeding that limit raises the facing above its rated temperature, which both lowers the coefficient of friction (torque fade) and accelerates wear. When required dissipation exceeds the air-cooled limit, the answer is an oil-shear or water-cooled unit, not simply a larger dry brake.

The table below summarizes friction-lining families against their working envelope. Treat it as an initial screen; before specifying, obtain the manufacturer friction-and-wear data for the exact facing, mating-face finish, and duty cycle.

Lining FamilyDynamic CoF (typ.)Temp CeilingBest For
Organic / molded0.30 to 0.45Below 250 °CLight to moderate, quiet duty
Woven0.35 to 0.45~300 °CConformable, band and cone units
Sintered metal0.06 to 0.14 (high speed)Up to ~600 °CPresses, cranes, high pressure
Ceramic / carbonHigh, abrupt~540 °CHigh heat, high energy per cycle

On the standards side, lifting and material-handling brakes are the most heavily regulated. DIN 15435 and the FEM 1.001 rules require a holding-brake torque of at least 1.6 to 1.8 times the static load torque referred to the brake shaft, rising to 2.5 times for personnel hoists. EN 14492-2 mandates a manual release lever on power-driven hoist winches, and crane-rated brakes add armature-position microswitch monitoring so the control system knows the brake state. DIN VDE 0580 governs the electrical and thermal design of electromagnetic devices and coils. Enclosure protection for industrial spring-applied brakes is commonly available up to IP54, with higher ratings for washdown or outdoor service.

Chapter 5 / 06

Key Specification Parameters

Reading a clutch or brake datasheet is a core procurement skill. A single unit may list a dozen or more parameters, but only a handful drive the selection decision: static torque, dynamic torque, response (engagement and release) time, coil voltage and power, thermal energy per engagement and allowable cycle rate, air gap and wear allowance, and inertia. Each is decoded below.

Static torque is the torque the unit holds without slipping at zero or very low speed; it is the figure used to size holding brakes and clamps. Dynamic torque is the torque transmitted while the faces are slipping at speed, and it is always lower than the static rating, often markedly so. Sizing a stopping or starting duty from static torque alone is a classic error. Catalog torque also assumes a burnished unit at rated voltage and a cold coil; remember the roughly 8 percent torque loss per 20 degrees Celsius coil rise and the up-to-50 percent shortfall before burnishing.

Response time has two halves. Engagement time is the reaction delay plus torque rise after the coil switches, and for electromagnetic friction units it typically falls between 1/200 of a second and about 1 second. Disengagement time runs from switching the stator until torque has fallen to roughly 10 percent of rated torque. A spring-applied brake stops a load in about 0.1 to 3 seconds depending on inertia and speed. The magnetic air gap, coil inductance, and the release energy path all shift these numbers, and brief over-excitation at about three times nominal voltage can shorten pickup by roughly a third.

Coil voltage and power are the electrical interface. DC electromagnetic units are standard at 24 VDC, with 12, 45, and 90 VDC also common; AC units add a rectifier. Coil watts set the heat the field itself adds and influence response. Thermal capacity is specified two ways: maximum energy per single engagement (joules), and maximum continuous cycle rate or thermal horsepower for repetitive duty. Both must be checked, because a unit may pass a single hard stop yet overheat under rapid cycling.

The remaining parameters are mechanical. Air gap is the running clearance between armature and rotor; as the facing wears the gap widens, response slows, and fixed-armature designs eventually stop engaging, so adjustable or auto-adjust gaps extend service life. Inertia of the rotating clutch or brake members adds to the load the system must accelerate and decelerate, and matters on high-cycle machines. The list below groups the parameters by selection role.

  • Torque sizing: static torque, dynamic torque, service factor, voltage and temperature de-rating.
  • Dynamics: engagement time, release time, allowable speed (rpm), reflected inertia.
  • Thermal: energy per engagement (J), cycles per minute, thermal horsepower, friction-facing temperature ceiling.
  • Electrical: coil voltage (12 / 24 / 45 / 90 VDC), coil power (W), over-excitation and fast-release wiring.
  • Mechanical and environmental: bore and mounting, air gap and wear allowance, enclosure (IP), manual release, condition monitoring.
Chapter 6 / 06

Selection Decision Factors

To turn the preceding chapters into a specific model, follow the decision sequence below. Most selection mistakes come not from one wrong number but from deciding the type before the duty is understood, or from sizing on static torque when the duty is dynamic. These eight steps work as a fixed RFQ template.

  1. Function and fail behavior: First decide clutch, brake, or combined clutch-brake unit, then choose power-off (fail-safe) versus power-on logic. Any load that must be held during a power loss, a vertical axis, hoist, or servo holding duty, requires a spring-applied power-off brake.
  2. Torque sizing: Compute dynamic torque from reflected inertia and the required start or stop time, add steady load torque, then apply a service factor (commonly 1.5 to 2.5 times running load torque). For lifting duty, apply DIN 15435 and FEM 1.001 multipliers (1.6 to 1.8, or 2.5 for personnel). Verify the unit is sized on static torque for holding and dynamic torque for stopping.
  3. Engagement technology: On-off start-stop favors dry friction electromagnetic; proportional web tension favors magnetic particle or hysteresis; very high cycle rates favor oil-shear; heavy presses favor pneumatic or hydraulic clutch-brake units.
  4. Thermal duty: Calculate energy per engagement and multiply by cycles per minute, then compare against the unit thermal horsepower curve. If air-cooled capacity is exceeded, step to oil-shear or water-cooled, not merely a larger dry unit.
  5. Speed, inertia, and response: Confirm maximum operating rpm, reflected inertia, and the required engagement and release times, and specify over-excitation or fast-release wiring where cycle time is tight.
  6. Electrical interface: Coil voltage (24 VDC is the default; 12, 45, 90 VDC and AC-plus-rectifier are options), coil power budget, and the control method (simple switch, current-controlled supply for particle units, or drive-integrated brake control for servos).
  7. Mechanical and environmental fit: Bore, shaft and flange mounting, air gap and wear allowance, enclosure rating (up to IP54 typical, higher for washdown or outdoor), and required manual release and condition monitoring for lifting duty.
  8. Certification and total cost of ownership: Confirm the standards the application demands (DIN 15435, FEM 1.001, EN 14492-2, DIN VDE 0580, machine-safety approvals for presses), then weigh purchase price against burnishing labor, facing replacement interval, downtime, and spare-part lead time.

One last commonly overlooked dimension is serviceability and run-in. A dry friction unit needs burnishing to reach rated torque, periodic air-gap adjustment as the facing wears, and eventual facing or rotor replacement, so local spare-part stock and field service availability determine repair response over a five to ten year production life. Established suppliers, including Warner Electric, Ogura Industrial, Miki Pulley, KEB, INTORQ (Lenze), Magtrol, Placid Industries, Ortlinghaus, and Goizper, maintain distribution and replacement parts in major markets, which makes them safer choices for long-running lines. Always verify the exact series, voltage, and certification against the live datasheet before placing an order.

FAQ

What is the difference between a clutch and a brake?

A clutch couples two rotating members so torque transfers from a driving shaft to a driven shaft, while a brake couples one rotating member to a stationary frame so it dissipates kinetic energy and stops or holds a load. Mechanically they are near-identical: both use friction surfaces, an actuator, and the same torque equation. The only structural difference is that one half of a brake is grounded to the housing instead of turning. Many products combine both, a clutch-brake unit, so a single shaft can be started and stopped independently of a continuously running motor. Selection logic, torque sizing, and friction material choices carry over directly between the two.

What is the difference between a power-off (fail-safe) and a power-on brake?

A power-off brake, also called spring-applied, fail-safe, or safety brake, is engaged by compression springs in its de-energized state and is released only when the coil is energized to pull the armature back against the springs. If power, a wire, or the coil fails, the brake clamps automatically, which is why it holds vertical axes, servo motors, cranes, and hoists. A power-on brake is the inverse: the coil must be energized to apply braking torque, and it free-wheels when de-energized. Power-on units suit cyclic stop-on-command duty such as indexing tables and conveyors, but they cannot be trusted to hold a load during a power outage.

How do I size the torque rating for a clutch or brake?

Start from the dynamic torque needed to accelerate or decelerate the reflected inertia within the required time, then add the steady load torque, then apply a service factor. A common rule is to specify static torque at 1.5 to 2.5 times the running load torque for general machinery. Crane and hoist holding brakes follow DIN 15435 and FEM 1.001, which require at least 1.6 to 1.8 times the static load torque referred to the brake shaft, rising to 2.5 times for personnel hoists. Remember that a new friction unit may deliver only about 50 percent of its catalog static torque until it is burnished, because contact is initially confined to surface peaks until 20 to more than 100 break-in cycles seat the faces, and that dynamic torque at speed is lower than the static rating.

What is engagement time and how fast can these devices respond?

Engagement time is the reaction delay plus the torque rise time after the coil is switched, and for electromagnetic friction units it typically falls between 1/200 of a second and about 1 second. Disengagement time is measured from switching the stator until torque has fallen to roughly 10 percent of rated torque. A spring-applied brake can stop a load in about 0.1 to 3 seconds depending on inertia and speed. Response depends strongly on the magnetic air gap, coil inductance, and how energy is dumped at release. Over-excitation, briefly driving the coil at about three times nominal voltage, can cut pickup time by roughly one third.

When should I use a magnetic particle or hysteresis clutch instead of a friction clutch?

Choose magnetic particle or hysteresis units when you need controllable, repeatable torque that is independent of slip speed, which is exactly what web tension control on winders, unwinders, and test stands demands. In a magnetic particle device the transmitted torque is nearly proportional to coil current and stays constant regardless of slip speed, so it can run in continuous controlled slip. Hysteresis units add the benefit of no mechanical contact in the magnetic gap, giving smooth torque, long life, and no wear particles. Friction clutches are the better, cheaper choice for simple on or off start-stop duty where proportional torque is not required.

How do I dissipate the heat generated during slipping or repeated cycling?

Every engagement converts kinetic and slip energy into heat at the friction interface, and energy per cycle multiplied by cycles per minute gives the thermal power the unit must shed. Dry friction units are limited by the friction material temperature ceiling: organic facings degrade well below 250 degrees Celsius, while sintered metal and ceramic facings tolerate higher temperatures, with some ceramic facings rated near 540 degrees Celsius. When dissipation exceeds the air-cooled limit, move to oil-shear or water-cooled clutch-brake units, which run the friction plates in fluid to carry heat away, or de-rate the cycle rate. Confirm both the per-engagement energy and the thermal horsepower against the manufacturer curves.

Which manufacturers and series are common for industrial clutches and brakes?

For electromagnetic friction clutches and brakes, Warner Electric (CB clutch-brake series, 12, 24, or 90 VDC, roughly 2.8 to 565 Nm), Ogura Industrial (RNB power-off brakes about 2 to 200 Nm, MCNB spring-applied brakes in 24, 45, or 90 VDC), and Miki Pulley are widely specified. For servo and hoist fail-safe brakes, KEB, INTORQ (Lenze), and SG Transmission are common. For tension control, Magtrol HCF hysteresis clutches, Ogura, and Placid Industries magnetic particle units lead. For heavy press clutch-brake units, Ortlinghaus (406 and 420 series) and Goizper (5.8 series, clutch torque up to 150,000 Nm) dominate. Always verify the exact model, voltage, and certification against the live datasheet before ordering.

Ask SpecForge AI