How Sheave Diameter Sets Speed and Torque

Why this matters

A sheave change reads like a speed adjustment and behaves like a load change. The sheave itself is honest: it trades speed for torque at constant power, and nothing is created or destroyed at the shaft. The trouble is on the other end. A centrifugal fan or pump absorbs power with the cube of its speed, so a modest change in diameter arrives at the motor multiplied by roughly three, and a tech who opened an adjustable sheave a couple of turns to chase an airflow number can hand a comfortably loaded motor a load it cannot carry through a hot afternoon. The tach reads better, the customer is happy, and the overload trips in July.

Isolate before you change a sheave setting

Lock out and tag the motor at its disconnect. A wheel holds rotational energy after the power is off, which is stored mechanical energy under 29 CFR 1910.147: verify zero rotation by eye and block the wheel or close and secure the damper before reaching inside the guard. Adjusting an adjustable-pitch sheave means loosening a setscrew on a flange that carries thread and spring load; back it off with the flange supported and your hand out of the closing path between the flanges. Any running current reading is taken with a clamp meter at an already-accessible conductor, because opening an energized enclosure to meter is energized work under 29 CFR 1910.333(a)(1), which requires a written justification and arc-flash protection per NFPA 70E. Take a running tach reading from outside a closed guard and stand out of the plane of rotation, since a failed belt or sheave fragment departs along that plane. The guard goes back on before restart under 29 CFR 1910.219 in general industry or 29 CFR 1926.300(b) on a construction site.

What the sheave actually does

Speed comes out of the pitch diameters and nothing else:

driven speed = motor speed x driver pitch diameter / driven pitch diameter

Torque goes the other way by the same factor, because the net pull in the belt is the same value at both ends of the drive and only the radius it acts on differs. Power in equals power out minus drive losses, which for a correctly tensioned and aligned V-belt drive at its design point are commonly published in the mid-90s percent range and fall off outside it. So the sheave is a trade, not a source. It cannot give the machine more power than the motor is producing, and it cannot reduce the power a load demands.

Two definitions worth nailing down, because half the field errors here are definitional. Pitch diameter is the diameter at which the belt's tensile cords sit, which is not the sheave's outside diameter and not the groove bottom. Effective diameter is where the belt is actually riding today, which on a worn groove is smaller than the pitch diameter. Every equation above uses pitch diameter and every real machine runs on effective diameter, and the gap between them is a separate article.

The amplification: why a small diameter change is a large load change

On a centrifugal fan or pump operating along a fixed system curve, flow scales with speed, pressure with the square of speed, and absorbed shaft power with the cube. Combine that with speed scaling directly with driver diameter and you get the rule that matters:

On a centrifugal load, absorbed power scales with the cube of the driver's pitch diameter.

The condition in that sentence is load-bearing. It holds only for a centrifugal machine, only along a fixed system curve, and only where the pressure the machine works against is genuinely proportional to the square of flow. A system with a large fixed component that does not scale, a filter bank near the end of its life, or a damper that moved between the two readings all break it, and a positive-displacement load never obeyed it in the first place.

The convenient shorthand, that a one percent diameter change is a three percent power change, is a first-order approximation. At a 12 percent diameter change it already predicts 35 percent where the true figure is 40 percent, and the gap widens from there. Use it for a feel and do the cube for a decision.

Enlarging the driver is not the same as shrinking the driven

Both reach the same speed, and they land differently on the belt.

The net pull row is the one worth internalising. Net pull is torque divided by the driver's effective radius. Enlarging the driver raises torque with the cube but divides it by a larger radius, so the belt's burden rises only with the square. Shrink the driven instead and the driver radius never changes, so the belt takes the full cube. The motor is loaded identically either way; the belt is not.

Worked example: two turns on an adjustable sheave

A belt-driven supply fan. Motor nameplate 3.0 hp at 1750 rpm, driven sheave pitch diameter 9.0 in, center distance 15.0 in. The adjustable driver sheave measures 3.4 in pitch diameter as found, and a clamp reading at the accessible conductor puts the motor at about 78 percent of its nameplate power. Fan speed is 1750 x 3.4 / 9.0 = 661 rpm.

A tech closes the adjustable flanges to raise the belt in the groove, and the driver now measures 3.8 in pitch diameter.

Speed. 1750 x 3.8 / 9.0 = 739 rpm, an 11.8 percent increase, which is exactly the 3.8 / 3.4 diameter ratio.

Motor load. On a fixed system curve, 1.1176 cubed = 1.40, so absorbed power rises about 40 percent. The motor was at 78 percent of nameplate; 78 x 1.40 = 109 percent. It is now over nameplate power, and the only thing between that and insulation damage is the service factor. A common general-purpose service factor is 1.15, and it applies at rated voltage and frequency at the rated ambient temperature, commonly 40 C, with continuous operation inside the service-factor band shortening insulation life rather than being free headroom. On a rooftop in summer, at the end of a long feeder with sagging voltage, none of those conditions hold.

Shaft power and torque. Before: 0.78 x 3.0 = 2.34 hp, and motor torque is 63,025 x 2.34 / 1750 = 84 lb-in. After: 2.34 x 1.40 = 3.28 hp, and torque is 63,025 x 3.28 / 1750 = 118 lb-in.

Net pull. Before, at a 1.7 in driver radius: 84 / 1.7 = 49.6 lb. After, at a 1.9 in radius: 118 / 1.9 = 62.1 lb. The belt's burden rose 25 percent, which is the square of the 11.8 percent speed change, while the motor's rose 40 percent. Had the same speed been reached by shrinking the driven sheave instead, the driver radius would have stayed at 1.7 in and net pull would have risen the full 40 percent, to 69.4 lb, about 10 percent more pull on the same belt for the same result.

What got better. The diameters converged from a 5.6 in difference to 5.2 in, so wrap on the small sheave went from 180 - 2 x arcsin(5.6 / 30.0) = 158.5 degrees to 160.0 degrees. Belt speed went from pi x 3.4 x 1750 / 12 = 1,558 ft/min to 1,741 ft/min, both inside the range where published belt ratings are healthy, so the drive itself is fine. Nothing about the belt is going to warn anyone.

Same change, different load. Put that identical sheave change on a positive-displacement machine, a rotary screw compressor or a piston pump, and the driven torque is roughly constant with speed rather than rising with its square. Power then scales directly with speed: 78 x 1.118 = 87 percent of nameplate. Same 11.8 percent speed increase, same sheave, and the motor stays comfortably inside its rating. The sheave did not behave differently. The load did.

How the fan case fails. Not immediately. It runs, it moves more air, and the motor sits above nameplate on mild days when the ambient keeps the winding cool. The failure is a summer overload trip that gets read as a nuisance trip, then a contactor replacement, then a motor. Anyone diagnosing it a year later has no record that a sheave was ever moved.

The limits that bind before the arithmetic does

• Motor shaft minimum sheave diameter. Motor makers and NEMA publish minimum sheave pitch diameters by horsepower and speed, sized to keep shaft bending stress inside limits. A sheave below that number is a shaft problem, not a belt problem, and no tension setting rescues it.
• Belt section minimum diameter. Every section has a minimum recommended sheave diameter; below it the belt is bent tighter than its construction tolerates and fatigues from the inside out.
• Adjustable sheave travel. Makers publish a minimum effective pitch diameter and a maximum number of turns open, because past that the belt rides on the groove bottom and the wedge is gone. Count the turns and record them.
• Belt speed. Published horsepower ratings per belt rise through a broad mid range and fall off at both ends. Read the rating table at the belt speed you are actually going to run, not at the one the drive was selected for.

How to verify you got this right

• Take the motor current before you change anything, with a clamp meter at an accessible conductor, and write it down with the ambient temperature. Without a before reading you cannot say what the change did.
• Measure both pitch diameters after the change, do not assume the adjustment moved what you thought it moved. Count the turns and record the number, because that is the only way the next tech can put it back.
• Recompute expected power with the cube on a centrifugal load or the first power on a constant-torque load, and compare it against nameplate before you close the guard.
• Take the current reading again at the hottest condition you can reach, not at commissioning temperature. A motor that sits at 100 percent of nameplate power on a cool morning is not at 100 percent in August.

Read current as an indicator rather than a measurement of power: it tracks load but not proportionally, since power factor and efficiency both move with loading and the relationship flattens near and above full load. Where the decision is close, get the nameplate figures and the actual voltage rather than trusting amps alone.

References

• 29 CFR 1910.147 for mechanical isolation and stored rotational energy; 29 CFR 1910.333(a)(1) and NFPA 70E for energized-work justification and protection
• 29 CFR 1910.219 (general industry) and 29 CFR 1926.300(b) (construction) for guarding of belts, pulleys and sheaves
• Motor manufacturer and NEMA data for minimum sheave pitch diameter by horsepower and speed, and for service factor conditions
• Belt and sheave manufacturer engineering data for minimum sheave diameter by section, horsepower ratings by belt speed, and adjustable-sheave travel limits
• See related: How to Work Out a Drive Ratio From What Is on the Machine; What a Worn Sheave Groove Does to a New Belt; How a Variable Speed Drive Changes the Mechanical Picture