Braking a Motor Shaft: Eliminating Coast-Down Overtravel

A shaft that keeps rotating half a turn after the contactor drops out is doing arithmetic, not misbehaving. The rotating mass carries kinetic energy, the only torque opposing it is bearing and gear friction, and the coast angle is whatever those two numbers dictate. Nothing in the control scheme changes that. The number that matters is the reflected inertia at the motor shaft and the friction torque holding it back — everything else in the retrofit follows from those two.

Fixes That Fail on Coasting Loads

Four approaches get tried first on this symptom, and three of them leave the machine in the same place.

Dropping the run command earlier. If the overtravel is the same angle every cycle, an operator or a repositioned limit switch can anticipate it. That anticipation collapses the moment the coast angle varies — cold grease versus hot grease, loaded versus unloaded, chain tension after wear. When the overtravel is never the same twice, no amount of trigger tuning fixes it, and asking the operator to guess is unreasonable. A long coast-down time constant makes it worse: the machine drifts past the target, settles, and the cycle time is spent waiting.

Plugging the motor. Momentary reverse across-the-line contactor operation does stop a load hard, and it worked well on the old U-frame motors that had rotor mass and thermal margin to absorb it. T-frame motors killed the practice: the same plugging duty pushes rotor and stator temperatures past what the smaller frame tolerates, and the reverse torque transient loads couplings, chains and gear teeth in the direction they were never sized for. Reduced-voltage plugging through an autotransformer softens it, but by the time you have bought the autotransformer, a reversing starter, and a plugging switch or zero-speed timer to de-energize at standstill, you have spent more than a brake and added three failure modes.

Softening the stop in the drive. A ramp-to-stop only helps if the drive can actually absorb or dissipate the returned energy. Without a braking resistor or a regenerative front end, the drive rides the DC bus up, trips on overvoltage, or coasts anyway.

Counting on gearbox friction. Friction is what already produced the half turn. Adding a stage or a worm set changes the number, not the mechanism, and a self-locking worm ratio brings its own efficiency and backdriving arguments.

Where the Kinetic Energy Goes

This is energy, not logic. At the instant power is removed, the drivetrain holds

E_k = 0.5 * J_ref * w^2        [J], w in rad/s at the motor shaft
J_ref = J_motor + J_load / N^2  (N = total reduction ratio, motor:load)

That energy is spent against retarding torque over the coast angle, so

theta_coast = (J_ref * w^2) / (2 * T_retard)     [rad at the motor shaft]
theta at the load = theta_coast / N

A half turn at the output shaft with a 40:1 reduction is 20 motor revolutions of stored rotation — which is why the fix belongs on the machine, not in the pushbutton wiring. Invert the same relation to get the brake you need:

T_brake >= (J_ref * w^2) / (2 * theta_allowed) - T_retard
t_stop  =  (J_ref * w) / (T_brake + T_retard)     [s]

Torque referred through the gearbox is the reason brake placement dominates cost: a brake on the motor shaft sees roughly T_load / N, while a brake on the final shaft sees the full load torque and has to be sized, and paid for, accordingly.

Reading the Symptom Before Choosing Hardware

Observed behavior	Mechanism	What it rules in
Overtravel the same angle every cycle	Constant inertia, constant friction, repeatable coast	Early trigger point or DC injection is viable; friction brake still preferred for repeatability
Overtravel varies cycle to cycle	Load, temperature or lubricant viscosity changing T_retard	Mechanical friction brake — anticipation cannot track it
Shaft settles, then rotates backwards	Unbalanced or suspended load, spring return, or air backdraft on a fan backdriving the rotor	Fail-safe spring-set holding brake, or DC injection held on to lock the rotor
Shaft creeps while stopped	Zero holding torque once braking current is removed	Holding brake, not a dynamic-only braking scheme
Position repeats at the motor but wanders at the output	Chain, sprocket and coupling backlash between brake and load	Brake on the final shaft, or eliminate the chain
Stop torque adequate but motor runs hot after weeks	Braking energy dumped into the stator, starts/hour exceeded	Move the energy out of the motor — friction brake or braking resistor

Stopping Torque Versus Holding Torque

Two duties get confused and they need different hardware. Deceleration torque acts only while the shaft is turning: DC injection, dynamic braking into a resistor, and regenerative braking all produce it and all produce exactly zero at standstill. Holding torque acts at zero speed and comes from friction — a spring-set, electrically-released brake that clamps when its coil is de-energized. If the driven shaft can be backdriven by gravity, an unbalanced load, or pressure differential, the machine needs the friction brake regardless of how good the deceleration scheme is. Fail-safe direction matters: the brake must set on loss of power, not release.

C-Face Brake Retrofit Between Motor and Reducer

The cheapest mechanical answer on an existing installation is a C-face brake sandwiched between the motor and the gear reducer, because it works at motor-shaft torque.

1. Confirm the motor and reducer input are a matching C-face frame, and read the motor full-load torque from the nameplate speed and rating.
2. Size brake torque from T_brake >= (J_ref * w^2) / (2 * theta_allowed) - T_retard, then check it against motor full-load torque; specifying at or above full-load torque covers holding duty at the motor shaft.
3. Check the brake's thermal rating in joules per stop and stops per hour against your computed E_k and cycle rate. Friction brakes fade and wear on thermal overload well before they fail mechanically.
4. De-energize and lock out, then unbolt the motor from the reducer.
5. Bolt the brake housing to the reducer input face, then bolt the motor to the brake, aligning the brake hub on the motor shaft to the manufacturer's dimension.
6. Wire the coil and rectifier per the brake's diagram, releasing on the same command that energizes the motor and setting on stop and on any loss of control power.
7. Set the air gap to the published cold value, run twenty to thirty cycles, and re-check the gap after run-in.

This layout handles the great majority of coast-down complaints. What it does not cover is anything downstream of the brake: a broken chain or failed coupling still releases the load. Where that risk exists, put the brake on the final shaft and accept the larger, more expensive unit — and on the next machine, specify a final reduction gear with a bored hollow output that mounts directly on the driven shaft, which deletes the chain and sprockets and the failure they bring.

Clutch-Brake for High Cycle Rates

When the stop happens many times per hour, stopping the motor at all becomes the problem. A clutch-brake assembly mounted between the reduction gear and the driven shaft lets the motor run continuously: the clutch engages to index the shaft, then releases while the brake stops and holds it. The motor never sees a start or a stop, so starts-per-hour ceases to be a constraint. Pneumatically actuated clutch-brake units are standard practice on high-cycle indexing machinery such as alligator shears cutting re-bar, where the drive coasts and the tooling is stopped and held mechanically. The trade is an air supply, a larger footprint at the output shaft, and friction-face wear as a scheduled consumable.

DC Injection Braking and the Thermal Bill

If a VFD is already on the motor, DC injection braking is almost certainly available in firmware — look for the DC brake level, DC brake time, and the frequency or speed threshold at which injection begins, and set the level as a percentage of motor rated current with an eye on rotor heating. Any drive can also take a braking resistor, and a four-quadrant drive with an active front end returns the energy to the line instead of burning it. Where no drive exists, standalone DC injection units cover the gap. Two features separate a working unit from one that destroys itself:

• Zero-EMF detection. The unit must confirm motor back-EMF has decayed before it applies DC. Cheap units inject into a live winding and fail.
• Automatic shutoff at standstill. Injection must terminate on its own once the rotor stops, either on a timer or on a zero-speed sense.

Injection into the stator at rest also produces winding heat, which older installations used deliberately as anti-condensation heating in damp environments — useful on outdoor or washdown machines, and a thermal load you must still account for.

The thermal accounting is the part that gets skipped. DC braking makes the motor the heat sink for the entire kinetic energy of the moving mass, so treat every stop as thermally equivalent to a start. A motor rated for 10 starts per hour becomes a motor rated for 5 starts and 5 stops. Exceed that and the failure shows up as insulation degradation months later, not as a trip on the day of commissioning.

Numbers to Pull Before Ordering

Quantity	Why it decides the design	Where to read it
Motor full-load torque and base speed	Sets minimum C-face brake torque and w in the energy equation	Motor nameplate
Total reduction ratio N	Reflects load inertia and load torque to the motor shaft	Reducer nameplate or gear data sheet
Load inertia J_load (WK²)	Dominates E_k on large driven masses	Machine drawing, or computed from geometry and mass
Allowed overtravel at the output	Target angle in the brake-torque equation	Process requirement
Starts per hour rating	Halves when DC injection is used for stopping	Motor data sheet / duty rating
Brake energy per stop and stops per hour	Thermal limit on the friction brake	Brake manufacturer selection tables
Brake air gap, cold and wear limit	Release reliability and wear interval	Brake installation manual

Verification After the Retrofit

1. Mark the output shaft in degrees against a fixed reference and log overtravel for at least 25 consecutive cycles at the worst-case load. Spread, not average, is the acceptance criterion.
2. Time the stop from command drop to standstill and compare against t_stop = (J_ref * w) / (T_brake + T_retard). A measured time well above the calculation means the brake is releasing late, dragging, or under-torqued.
3. Cut control power mid-cycle and confirm the brake sets and holds the load with no creep.
4. Measure brake coil voltage and current at the coil, not at the panel, to confirm the rectifier output and that release is clean.
5. Run the machine at production cycle rate for a full shift, then check motor frame temperature and brake housing temperature. Rising temperatures across the shift means the duty cycle exceeds the thermal rating of whichever element is absorbing the energy.
6. Re-check and reset the air gap after run-in, and put it on the maintenance schedule.

When to Escalate

Stop and call the drive or brake manufacturer's application engineering when load inertia cannot be established from drawings, when computed brake energy per stop lands within roughly 20% of the brake's published thermal limit, or when the stop is part of a safety function and needs a rated stop category rather than a process brake. Bring the motor nameplate, the reduction ratio, the measured cycle rate, and your overtravel log — those four items are what a selection engineer needs, and without them any recommendation is a guess.

FAQ