Torque Multiplication

Part of Gear Making

How gear trains multiply torque to move heavy loads, and how to calculate the gear reduction needed for specific applications.

Why This Matters

Torque is the rotational force that actually does work in a machine — it’s what tightens a bolt, grinds grain, lifts a load, or forms metal under a press. The human body can produce about 1-2 ft-lbs of sustained torque at a hand crank, and a horse provides roughly 30 ft-lbs at walking pace. To crack rock, press oil from seeds, or operate a heavy blacksmith’s trip hammer, you need hundreds or thousands of ft-lbs.

Gears provide the answer through torque multiplication: by reducing rotational speed, a gear train proportionally increases torque. A 100:1 gear reduction turns 2 ft-lbs of human crank effort into 200 ft-lbs of output torque — enough to slowly drive a metal-forming press or grind tough material. This principle is the foundation of every mechanical advantage system.

Understanding torque multiplication also means understanding its limits and trade-offs. You cannot generate power from nothing; torque multiplication always trades speed for force. A clear grasp of these trade-offs lets you design machines that use available power sources effectively and match machinery to the realistic capabilities of human, animal, and water power.

The Physics of Torque Multiplication

Torque (T) equals force (F) multiplied by radius (r): T = F × r. A 100-lb force applied 2 feet from a pivot creates 200 ft-lbs of torque.

When two gears mesh, the force at the contact point is the same for both gears (Newton’s third law). But the radii differ, so the torques differ:

T_output / T_input = r_output / r_input = N_output / N_input = Gear Ratio

For a simple speed reduction (large gear driven by small pinion):

• Output torque = Input torque × (N_driven / N_driver) × Efficiency
• If N_driven/N_driver = 5 (5:1 reduction), output torque = 5 × input torque (minus losses)

Conservation of energy means power stays constant (minus friction):

• Power = Torque × Angular Velocity
• T_in × ω_in = T_out × ω_out (approximately, ignoring losses)
• Therefore T_out = T_in × (ω_in / ω_out) = T_in × Gear Ratio

This is not magic — the output shaft turns more slowly, so the power balance works out. You cannot have high torque AND high speed at the output without a more powerful input.

Designing for Required Torque

Step 1: Determine required output torque

For grinding grain: typical millstone requires 50-200 ft-lbs depending on stone weight and material. Start with the grain mill manufacturer’s data, or estimate based on stone weight (heavier stones require less additional driving torque since friction increases with normal force gradually, but more torque for acceleration).

For a hydraulic press: required torque at the screw = Force × Lead / (2π × Efficiency). A 10-ton press with a 0.5-inch lead screw at 50% efficiency needs T = 20,000 × 0.5 / (6.28 × 0.5) = 3,185 in-lbs ≈ 265 ft-lbs.

For pumping water: Power = Flow rate × Pressure × Volume. Torque = Power / Angular velocity.

Step 2: Calculate available input torque

Water wheel at 15 rpm producing 0.5 HP: T = Power / ω = (0.5 × 33,000 ft-lbs/min) / (2π × 15) = 16,500 / 94.2 = 175 ft-lbs.

Human at crank: sustained 1-1.5 ft-lbs at 60-80 rpm; brief maximum 3-5 ft-lbs at 20-30 rpm.

Horse on whim (circular capstan): 30-50 ft-lbs at 3-5 rpm depending on radius and gait.

Step 3: Calculate required gear ratio

Ratio = Required Output Torque / (Available Input Torque × Efficiency)

For water wheel (175 ft-lbs) driving mill requiring 120 ft-lbs: Ratio = 120 / (175 × 0.85) = 0.81 — actually a speed increase of 1.25:1 is needed, since the water wheel provides more torque than needed. This means a 1:1.25 step-up in torque terms (the water wheel has surplus torque but maybe too low speed for the millstone).

Practical Torque Multiplication Devices

Compound gear trains: Each stage multiplies torque by its reduction ratio. A three-stage train with 4:1, 4:1, and 4:1 ratios gives 64:1 total torque multiplication. At each stage, the shaft torque increases while speed decreases.

Critical design point: The highest-torque shaft (the output) carries the most force through bearings, keyways, and gear teeth. Size these components for the output torque, not the input. A 64:1 reduction from a 2 ft-lb input produces 128 ft-lbs output — the output shaft and its gear must handle 128 ft-lbs, which requires substantially larger components than the input stage.

Worm gear: A single worm and worm wheel can achieve 20:1 to 60:1 in one stage, ideal when high torque multiplication is needed in compact space. The trade-off: efficiency drops to 50-80% depending on worm lead angle. With a self-locking worm (small lead angle), the drive is non-reversible — useful for winches and presses that need to hold load without running backwards.

Lever + gear combination: For intermittent, very high torque (like a trip hammer or coin press), a lever multiplied by a gear reduction from a flywheel is common. The flywheel stores kinetic energy during the free part of rotation and releases it at impact, effectively multiplying instantaneous torque far beyond steady-state capability.

Torque vs. Power: Common Confusions

People often conflate torque with power. They’re distinct:

• Torque is rotational force — the tendency to cause rotation
• Power is the rate of doing work — torque multiplied by speed

A strong man applying 10 ft-lbs of torque to a stuck nut at 0 rpm produces zero power (no movement). A small electric motor spinning at 3,000 rpm with only 0.5 ft-lbs of torque produces 3,000 × 0.5 / 5,252 = 0.29 HP.

High torque at low speed can accomplish work that requires sustained force but little movement: pressing, lifting, breaking free a stuck mechanism. Low torque at high speed moves things quickly but with limited force.

When designing gear trains, always verify both: does the output have enough torque for the load, and does the input have enough power to sustain operation? A high-ratio gear train can produce enormous torque from a modest power source, but only at very low speed. The work still takes the same time; gearing just changes the force-speed trade-off.

Efficiency stacking: In a multi-stage gear train, efficiency losses compound. Three stages at 97% efficiency each give 97% × 97% × 97% = 91% overall. Five stages at 95% each give 77% overall. This is why minimizing the number of gear stages is good design practice — but never at the cost of using excessively large gears that can’t be fabricated.