Control Valve Cv: Valve Travel Is Not Flow Percentage

The valve reports 50% travel, yet flow does not settle at 50% of the full-open rate. That behavior is normal unless both the valve’s inherent characteristic and the system pressure conditions make the relationship linear. Treat travel, valve coefficient, pressure drop, and flow as separate signals; measure each before changing tuning.

Why do the usual fixes fail?

Several quick fixes attack the displayed percentage instead of the hydraulic cause:

• Dividing full-open flow by valve travel: This assumes 50% travel produces 50% of maximum Cv. That is valid only for an ideal linear trim over its characterized range, not for equal-percentage trim or an unspecified valve.
• Retuning the controller: Tuning changes how the controller reacts to error. It cannot make an incorrect flow measurement, nonlinear trim curve, changing valve pressure drop, or oversized valve linear. Tuning does not fix wiring, scaling, or valve selection.
• Linearizing only the command signal: A 50% controller output does not prove 50% stem travel. Positioner calibration, linkage, deadband, stiction, and actuator limits sit between the command and the trim.
• Using the full-open rating at every position: The published maximum Cv describes full travel. Partial-travel capacity must come from the manufacturer’s travel-versus-Cv curve or the stated trim equation.
• Replacing the valve before checking pressure: Flow depends on both Cv and valve differential pressure. Piping losses, pump head, elevation, and downstream pressure can change as flow changes.

Look at the trend first. Compare command, measured position, upstream pressure, downstream pressure, and flow on the same time base before adjusting the loop.

What signals determine the wrong flow value?

The useful signal chain is controller output to actuator motion, motion to effective Cv, and effective Cv plus differential pressure to flow. A discrepancy must be located within that chain before changing controller gains.

What actually sets flow through the valve?

Cv is the flow of 60 °F water in US gallons per minute that produces a 1 psi pressure drop across the valve. For liquid service, the evidence gives the relationship:

Cv = Q × sqrt(Density / (PressureDrop × 62.4))

Rearranging it gives:

Q = Cv × sqrt(PressureDrop × 62.4 / Density)

The same relationship is often expressed using specific gravity as Q = Cv × sqrt((P1 − P2) / SG), with units matched to the coefficient definition. The equation shows two independent influences: flow is directly proportional to Cv when density and valve pressure drop remain fixed, but it varies with the square root of pressure drop when Cv remains fixed.

For the stated example, full-open Cv = 300 and measured flow is 800 m3/h. If partial travel really produced Cv = 150 and the fluid density and valve differential pressure stayed at their full-open values, flow would be half of 800, or 400 m3/h. If Cv = 75 under those same conditions, flow would be one quarter, or 200 m3/h. Those are conditional ratios, not general position-to-flow rules.

How does valve travel change Cv?

Travel, also called stroke, is the motion from the closed position to full travel. A reading of 50% travel means the stem, shaft, or closure member has moved halfway through its calibrated span; it does not identify the resulting capacity without the trim characteristic.

For an ideal linear characteristic, the evidence gives:

Cv(x) = Cvmax × x / 100

Under that model, a valve with Cvmax = 300 has Cv = 150 at 50% travel and Cv = 75 at 25% travel. Flow follows the same ratios only while differential pressure and fluid properties remain unchanged.

An equal-percentage characteristic behaves differently. Equal increments of travel produce equal percentage changes in capacity, so the curve rises slowly near the closed end and rapidly near full travel. One supplied form is:

Cv(x) = Cvmax × R^((x / T) − 1)

Here x is actual travel, T is full travel, and R is rangeability. At half travel, this equation gives Cv = Cvmax / sqrt(R), not automatically half of Cvmax. Read R and the usable travel limits from the valve data rather than assigning a value.

Actual equal-percentage globe-valve examples show why position alone is inadequate. A 1-inch valve was listed at Cv 0.79 at 10% opening and Cv 13.2 at 100%; another 4-inch valve was listed at Cv 4.9 at 10% and Cv 190 at 100%. Even valves described by the same broad characteristic can have different normalized curves. A butterfly valve can also provide poor control around 10% opening compared with a cage-guided globe valve with anti-cavitation trim.

Why does the installed flow curve differ from the trim curve?

The inherent valve characteristic relates travel to Cv under defined test conditions. The installed characteristic relates travel to flow after the valve is connected to the process. They match only when the differential pressure across the valve remains effectively constant.

As a valve opens, total flow rises and friction loss in the rest of the piping increases. More of the available pressure is then consumed by pipe, fitting, and equipment losses, leaving a different differential pressure across the valve. In a pumped system, the operating point also moves along the pump curve, so supply head can fall as flow increases. Static elevation can add or subtract pressure depending on the arrangement.

This interaction can make a linear trim produce a nonlinear installed flow curve. It can also make an equal-percentage trim produce a more nearly linear installed response because its increasing capacity compensates for the falling share of pressure drop across the valve. The deciding measurements are P1, P2, flow, and actual travel throughout the operating range.

How should the actual travel-to-flow relationship be calculated?

1. Confirm the fluid, operating state, and units. Record density or specific gravity at the condition used for the calculation.
2. Identify the exact valve and trim. Obtain the manufacturer’s travel-versus-Cv curve rather than using the full-open coefficient as a straight-line scale.
3. Calibrate the travel signal. Compare commanded position with measured stem or shaft position at closed, intermediate, and full travel.
4. Measure upstream and downstream pressure close enough to the valve to determine ΔP = P1 − P2. Collect these values at the same steady operating points as flow and position.
5. Read the effective Cv from the manufacturer curve at each measured travel. For a stated linear trim, calculate it with Cv(x) = Cvmax × x / 100; for stated equal-percentage trim, use the supplied curve or its documented equation and rangeability.
6. Calculate predicted flow with the liquid sizing relationship using compatible units. Apply the manufacturer’s sizing method if process conditions require corrections beyond the basic liquid equation.
7. Compare calculated flow with the transmitter reading. If travel is wrong, inspect the command-to-actuator path. If travel is right but effective capacity is wrong, inspect trim data and mechanical condition. If capacity is right but flow is wrong, investigate pressure and flow measurements or the hydraulic model.

How is the result verified before retuning?

Run a controlled travel test through the operating region, allowing the process to stabilize at each point. Trend controller output, actual travel, P1, P2, and flow. Repeat points while opening and closing; separation between the two paths indicates mechanical hysteresis, stiction, backlash, or positioner behavior rather than a simple capacity calculation error.

Calculate observed Cv from each stable flow and measured pressure drop, then compare it with the manufacturer curve at the same travel. Agreement in Cv but disagreement in flow percentage means the installed pressure-drop distribution is shaping the result. Disagreement between observed and published Cv directs attention to travel calibration, trim condition, valve configuration, or measurement quality.

Check low-flow resolution closely. An oversized valve can operate near the closed end, where small movements create large relative changes in flow and the actuator encounters seating effects or friction. The resulting cycling and poor low-flow control should be corrected through valve sizing, characteristic selection, or mechanical work before controller tuning.

When does valve sizing become the real correction?

Size the valve from required operating flow and the differential pressure actually available across the valve, then select a valve whose usable Cv curve covers the normal and limiting cases. Matching only the maximum required Cv can leave normal operation too close to the seat; choosing a much larger full-open coefficient reduces useful travel and makes small position changes hydraulically significant.

Valve style and trim must match the required controllability as well as maximum capacity. Compare the full operating envelope against the position-versus-Cv curve, including minimum controllable flow, normal flow, maximum flow, and the pressure drop at each condition. Use the measured installed characteristic to decide whether a linear or equal-percentage trim provides the desired process response.

FAQ