A 12-meter belt conveyor for a bottling line was fitted with a 2.2 kW gear motor. The catalog calculation said 1.5 kW was enough. On startup, the belt didn’t move. The motor hummed, the overload relay tripped, and the bottles stayed put. The installer checked the belt — it wasn’t jammed. The conveyor hadn’t run in 6 months, and the belt had stuck to the bed. The 1.5 kW calculation was right for running friction. It ignored breakaway friction, which was 4x higher.

The three torques that size a conveyor

A conveyor drive needs to cover three separate load cases:

1. Breakaway torque — the torque to start the belt moving from rest. After a shutdown, the belt sticks to the bed, the product weight settles into the belt, and the bearings are at rest. The breakaway friction coefficient is 2-4x the running coefficient. For a rubber belt on a steel bed, running friction is about 0.15; breakaway can be 0.5-0.6. The conveyor that sat for 6 months had a breakaway friction around 0.5.

2. Running torque — the torque to keep it moving at speed. This is the catalog number. Friction is lower once moving, but the drive still carries the full load.

3. Acceleration torque — the extra torque to accelerate the belt and product from zero to speed. For a 12 m belt carrying 50 bottles/m at 0.5 m/s, the total inertia (belt, product, pulleys, reducer, motor) accelerates to speed in about 1 second. That needs roughly 30-50% extra torque over running torque.

The numbers on the bottling conveyor

Belt: 12 m × 0.5 m, rubber on steel bed, total moving mass about 300 kg including product. Running friction coefficient 0.15.

Running force = m × g × μ = 300 × 9.81 × 0.15 = 441 N. At 0.5 m/s, power = 441 × 0.5 = 220 W. The catalog 1.5 kW had 6x margin on running torque. The designer was right — for running.

Breakaway force = 300 × 9.81 × 0.5 = 1472 N. At the same 0.5 m/s (assuming the motor starts at speed), power = 736 W. Still under 1.5 kW. So why did the 2.2 kW motor trip?

Because the breakaway force applied at the drive pulley radius with a 20:1 gearbox, at motor start, draws current. The motor’s starting torque is about 2x rated, so it should start… unless the belt had fully stuck. The belt had adhered to the bed (rubber to steel after months of sitting with product weight on it). The true breakaway coefficient was closer to 0.8 — the belt and bed had bonded slightly. Breakaway force = 300 × 9.81 × 0.8 = 2354 N. Power at 0.5 m/s = 1177 W. Starting torque at 2x rated on a 2.2 kW motor gives about 2x the rated 14 Nm = 28 Nm at the motor. Through a 20:1 reducer, that’s 560 Nm at the pulley. The pulley radius is 0.1 m, so the force at the belt is 5600 N. That should move 2354 N… The motor hummed and tripped — meaning the overload was set to motor FLA (not 2x) and the ramp-up didn’t have enough time. The belt did move eventually, but the motor current was at 180% FLA for the first 5 seconds and the thermal relay tripped.

What actually fixed it

Three changes, cheapest first.

1. A longer ramp time. The VFD (the conveyor had one) was set to a 1-second ramp. Increasing the ramp to 5 seconds let the motor deliver 1.5x rated torque steadily instead of spiking to 2x and tripping. The conveyor started smoothly. This alone fixed the immediate problem.

2. Belt lift-off at startup. The belt had stuck to the bed. A routine: before a cold start after a long shutdown, lift the belt by hand at the mid-span (or use a belt lifter) to break the adhesion. This brought the breakaway coefficient back to 0.5 instead of 0.8.

3. The real sizing rule for the next conveyor: size the drive for breakaway torque, not running torque. The sizing formula: T_drive = F_breakaway × r_pulley, where F_breakaway = m × g × μ_breakaway (use 0.5-0.6 for a stuck belt, 0.3-0.4 for a clean, recently-run belt). Then add the acceleration torque and pick the motor so its starting torque covers the sum with a 1.5x margin. The next conveyor used a 3.0 kW motor — the extra 0.8 kW cost $120 and eliminated the entire class of startup problems.

The duty cycle factor

Conveyors that start and stop frequently (every 2 minutes or less) are harder on the motor than continuous-running ones. Each start heats the motor. The motor’s service factor covers some of this, but a conveyor with 100 starts/hour needs a motor rated for the breakaway torque at every start, plus a VFD to control the ramp. The bottling conveyor started every 90 seconds. The catalog-sized 1.5 kW motor would have overheated even if the breakaway torque hadn’t stuck the belt. The 3.0 kW motor runs cool at the same duty.

The belt tension check

One more variable: belt tension. A conveyor belt that’s over-tensioned adds friction at every idler and at the drive pulley. The catalog friction coefficient assumes correct tension — typically 1-2% of belt breaking strength. The bottling conveyor’s belt was tensioned to 4% (someone cranked it “to be safe”). That added 30% to the running friction and doubled the bearing loads on the pulleys. After re-tensioning to 1.5%, the running current dropped 25%.

Conveyor drives are sized on breakaway torque, not running torque. The catalog 1.5 kW was right for a running belt and wrong for a belt that had stuck to the bed for 6 months. Use μ_breakaway of 0.5-0.6, add acceleration torque, put in a 5-second ramp, and check the belt tension. The motor that tripped wasn’t undersized for running — it was undersized for starting.