Belt Drive Design: How to Calculate Pulley Ratios, Belt Length and Tension

How Pulley Ratios Determine Speed and Torque Output

Establish your speed ratio before specifying any other component, because every downstream calculation, including belt length, tension, and shaft load, depends on it. The velocity ratio equals the driver pulley diameter divided by the driven pulley diameter. A 100 mm driver paired with a 200 mm driven pulley produces a 0.5:1 ratio, halving output speed and doubling output torque relative to the input.

Torque scales inversely with speed across the belt drive. When the driven pulley rotates slower than the driver, the mechanical advantage increases proportionally, assuming constant power and negligible losses. A drive running at 1,500 rpm input with a 3:1 reduction delivers 500 rpm output, with torque three times the input value before accounting for friction losses, which typically range from 2 to 5% in well-maintained V-belt systems.

Selecting a large diameter ratio across a single stage risks excessive belt wrap angle on the smaller pulley, reducing grip and accelerating wear. Keeping the diameter ratio below 6:1 per stage is standard practice for industrial V-belt drives; multi-stage arrangements handle higher combined reductions without this constraint.

Calculating Belt Length for Open and Crossed Configurations

Getting belt length wrong by even a small margin forces a redesign: too short overstresses bearings, too long causes slip and vibration under load.

For an open belt configuration, where both pulleys rotate in the same direction, pitch length is:

L = 2C + (π/2)(D + d) + (D − d)² / 4C

Here, C is the centre distance, D the larger pulley diameter, and d the smaller. The squared term corrects for angular offset caused by unequal pulley sizes.

A crossed belt configuration, used when opposite rotation is required, replaces the subtraction in the correction term with addition, because both wrap angles increase when the belt crosses:

L = 2C + (π/2)(D + d) + (D + d)² / 4C

Crossed belts generate higher contact stress at the crossover point, making them better suited to lower-speed drives.

Because commercial belts come in discrete sizes, calculate the theoretical length first, then select the nearest size from ISO 22: 2013 and adjust centre distance accordingly.

Tension Requirements: Tight Side, Slack Side, and Effective Pull

Too little belt tension causes slip under load; too much accelerates bearing wear and shortens belt life. The tight side carries effective pull plus slack side tension; the slack side carries only the residual pre-tension preventing belt lift. Effective pull F_e equals the difference: T₁ – T₂ = F_e, calculated directly from power and speed as F_e = P / v.

The Euler-Eytelwein relationship, T₁ / T₂ = e^μθ, gives the maximum tight-to-slack ratio before slip occurs, where μ is the belt-pulley friction coefficient and θ is wrap angle in radians on the smaller pulley. A small wrap angle, common when pulley diameters differ or centre distance is short, forces T₁ higher for the same F_e. Fitting an idler pulley on the slack side increases wrap angle and eases this constraint without raising tight-side tension.

Selecting Belt Type and Cross-Section for Load Conditions

Cross-section size determines whether a belt transmits load efficiently or fails through slip, heat buildup, or fatigue. A section too small runs hot and stretches; one too large adds unnecessary shaft load.

Classical V-belt sections (A, B, C, D) suit general-purpose drives from fractional kilowatt up to around 500 kW. Narrow sections (SPZ, SPA, SPB, SPC) carry higher power per unit width, useful when space is constrained or centre distances are short. Synchronous belts eliminate slip entirely, making them the correct choice where precise speed ratio matters, such as camshaft drives or indexing conveyors. Tooth shear limits peak torque, and they require no lubrication.

Service factor adjusts rated belt capacity to real operating conditions. Multiply motor power by the service factor before selecting a cross-section from manufacturer load tables. Shock loading or frequent starts typically require a factor of 1.3 to 1.5. Gates design tables and Optibelt calculation tools both provide service factor charts covering driver type and daily operating hours. Selecting from nameplate power alone leaves the drive undersized for peak demand.

Shaft Load, Bearing Forces, and Safety Factors

Bearing selection follows tension calculation because total radial shaft load equals the vector sum of T1 and T2. Where both tensions act in roughly the same direction, this approaches T1 + T2, always exceeding the effective pull Fe alone.

Centre distance affects shaft load directly. Short centre distances increase wrap angle on the smaller pulley, allowing lower T1 for a given Fe and reducing bearing forces. Longer centre distances reduce wrap angle, raising T1 and bearing loads. Designers fixing centre distance for packaging reasons must verify the resulting shaft load against bearing dynamic load ratings.

ISO 22: 2013 recommends service factors between 1.0 and 2.0. Smooth, constant-torque loads sit at the lower end; reversing or high-inertia drives apply 1.5 or above. Multiply calculated design power by this factor before entering any belt selection chart.

Bearing life estimation uses corrected radial load against the basic dynamic load rating C in the L10 formula. Doubling the radial load on a ball bearing cuts rated life by a factor of eight. Specify bearings with adequate load margins rather than targeting minimum rated life at nominal conditions.

Common Calculation Errors and How to Avoid Them

Verify unit consistency before running any formula. Mixing millimetres and metres in the same expression is the most frequent source of incorrect belt length and tension results. Confirm that diameter, centre distance, and velocity share the same base unit.

Using nominal pulley diameter instead of pitch diameter introduces compounding errors. The belt pitch line runs above the groove base, so effective diameter always exceeds groove diameter. For V-belts, the difference is typically 2–4 mm per side, affecting velocity ratio and Fe calculations.

Neglecting wrap angle correction on small pulleys overstates available friction capacity. The Euler-Eytelwein relationship requires θ in radians. Converting incorrectly, or reading the smaller wrap angle, produces a T1/T2 ratio the belt cannot sustain before slipping.

Safety factor selection deserves the same scrutiny as the main tension formula. Shock-loaded applications require a service factor applied to design power before selecting cross-section, just as working load limits in lifting equipment account for dynamic forces. Match your service factor to the load classification in the manufacturer’s charts, never default to 1.0.

Carry at least four significant figures through pitch length, tension ratio, and shaft load calculations before rounding the final output.