Irrigation Pump Curves: Find the Operating Point as Zones Change
An irrigation pump delivers the flow and head where its pump curve intersects the active system curve. Opening another zone changes the network resistance; it does not instruct a fixed-speed pump to deliver the sum of the zones' previous individual flows. This guide calculates that movement for an original hypothetical network, then checks the consequences for zone delivery, efficiency and control. The result is a worksheet for evaluating changing zone combinations with the designer and pump supplier.
An irrigation pump installation beside a tailwater recovery pond in Arkansas. Photo: Jeff Vanuga, USDA NRCS. Public domain.
Start with the curve for the installed pump configuration
A pump performance curve relates total head to flow at a specified rotational speed and configuration. Obtain the model, impeller diameter or stage arrangement, speed and test conditions that match the installed equipment. A curve for a different impeller or speed can give a plausible-looking but incorrect operating point. Keep the efficiency, shaft-power and required net positive suction head curves with the head-flow plot.
NDSU Extension's irrigation pump guide explains how to read flow, head and efficiency together and recommends retaining the correct pump curve. The maximum-flow and maximum-head labels describe different locations on that curve. They are not a guaranteed simultaneous duty. The irrigation pump sizing worksheet establishes the target duty; this article checks where a selected pump actually operates as the network changes.
For the worked example, use the invented relationship Hp = 50 − 0.01Q², with Hp in metres and Q in m³/h. Treat it only as a teaching approximation over the flows calculated below. It is not an IrriNex product curve, a full test report or permission to extrapolate to shutoff or runout. Its numerical coefficient depends on the chosen units.
Define what the system curve includes
A system curve gives the head needed to move a particular flow through a defined arrangement. Specify the water source, destination elevations, active branches, valve positions and component condition. In this example an open source supplies identical, unregulated sprinkler branches at the same elevation. The static rise is 10 m. Source pressure, additional elevation differences and pressure-regulating devices are absent from this simplified model.
Let Q be total pump flow and q the flow in an individual branch, both in m³/h. The shared suction and delivery path requires 0.003Q² metres of head. Each branch, including its pipe and pressure-dependent outlet requirement, needs 0.012q² metres above its elevation. These are invented combined resistance relationships; the branch term includes the outlet requirement, so it must not be added again. Use the pressure-budget guide to define real measurement boundaries and avoid counting a component twice.
With a single branch open, q = Q and Hs = 10 + 0.015Q². With both identical branches open, symmetry gives q = Q ÷ 2, so Hs = 10 + 0.003Q² + 0.012(Q ÷ 2)² = 10 + 0.006Q². The shared pipe still carries the total flow. Do not halve every loss merely because another branch opens. Different branch elevations or resistances require a network calculation instead of equal flow division.
Solve the intersection for each zone combination
Set pump head equal to system head. For one branch, 50 − 0.01Q² = 10 + 0.015Q² gives 40 = 0.025Q². The positive solution is Q = 40 m³/h and H = 34 m. Substitution checks both sides: the pump supplies 50 − 16 = 34 m, and the network requires 10 + 24 = 34 m. The flow is a calculated result, not a preselected zone demand.
For two branches, 50 − 0.01Q² = 10 + 0.006Q² gives 40 = 0.016Q². The result is Q = 50 m³/h and H = 25 m, with 25 m³/h in each branch. Total flow rises, but each branch receives less than the 40 m³/h it received alone. The new total is not 80 m³/h. The operating point has moved along the unchanged fixed-speed pump curve to meet a different system curve.
| Condition | System relationship, head in m | Total flow, m³/h | Pump head, m | Flow per active branch, m³/h |
|---|---|---|---|---|
| One branch, reference source level | 10 + 0.015Q² | 40 | 34 | 40 |
| Two identical branches, reference source level | 10 + 0.006Q² | 50 | 25 | 25 |
| One branch, source level lower by 6 m | 16 + 0.015Q² | 36.88 | 36.40 | 36.88 |
Now compare those results with the crop-delivery design. If each sprinkler branch requires its single-branch delivery to achieve the intended application pattern, the two-branch result is not acceptable simply because the pump still runs. Check pressure at the sprinklers and the resulting distribution. Pressure-compensating emitters and regulating valves have their own operating ranges; do not apply this unregulated quadratic branch model to them without an appropriate hydraulic model.
Separate a mathematical intersection from an acceptable duty
A curve crossing establishes a possible steady hydraulic balance under the model's assumptions. It does not certify the duty. Compare every proposed operating point with the manufacturer's permitted flow range, preferred operating region, maximum pressure, shaft-power requirement and motor capability. There is no universal acceptable percentage around the best efficiency point that can replace the selected equipment's limits.
Check the suction side at each duty as well. Available net positive suction head must satisfy the pump's required value plus the allowance specified for the application. Pumping water level, intake submergence, suction losses, temperature and local atmospheric conditions affect this check. A discharge point that appears on the curve can still be unsuitable if the source or intake cannot support it. Do not infer cavitation margin from a discharge gauge alone.
Steady curves also do not describe the pressure transients while valves move. The operating plan needs an approved route through startup, sector changes and shutdown, with independent protection active. The pump and valve interlock worksheet defines those permissions and failure responses. A satisfactory final intersection does not make every intermediate valve position permissible.
Recalculate when source level or resistance changes
The final table row assumes that the pumping source surface falls by 6 m while all other modeled properties remain unchanged. The static rise becomes 16 m. For one branch, 50 − 0.01Q² = 16 + 0.015Q² gives Q² = 1360, so Q is approximately 36.88 m³/h and pump head is 36.40 m. The pump develops more head at the new point while delivering less water.
This differs from adding 6 m to the previous head while keeping the previous flow. That calculation would describe a new required duty at an imposed flow, not the actual intersection of this unchanged pump and network. Real well drawdown may itself vary with flow and time. If it does, replace the fixed source-level assumption with the measured or designed source response rather than treating it as a constant.
Filter loading, throttling and pipe changes can also alter resistance. Use the actual component relationship and recalculate the intersection; do not simply add a constant loss measured at one flow to every point on the curve. A regulator that changes position to maintain downstream pressure may produce a different or piecewise system relationship. Record the valve's control mode and minimum operating differential before interpreting a measured curve shift.
Read efficiency and power at the new point
Efficiency is a separate curve reading. For an arithmetic illustration, assume pump-only efficiencies of 72% at the single-branch point and 70% at the two-branch point. These invented values are not derived from the head-flow equation and are not claimed product performance. In a real review, read the actual efficiency at the relevant speed and impeller configuration, then account for how long each duty occurs.
For water with density 1000 kg/m³ and gravity 9.81 m/s², shaft power in kW is 0.002725QH ÷ η, when Q is in m³/h and η is pump efficiency as a fraction. The single-branch case gives 0.002725 × 40 × 34 ÷ 0.72 = 5.15 kW after rounding. The two-branch case gives 0.002725 × 50 × 25 ÷ 0.70 = 4.87 kW. The second case has lower assumed pump efficiency but also a lower hydraulic power requirement.
These are shaft-power estimates, not motor nameplate sizes or measured electrical input. Motor and drive losses require separate data, and the motor must cover the approved operating range. The two cases also deliver different branch flow and pressure, so their power comparison does not prove an equivalent-service energy saving. For a schedule comparison, preserve the required irrigation outcome and use measured input energy, delivered volume and operating duration for each admissible duty.
Move the pump curve when speed changes
At unchanged impeller geometry, the affinity relationships can estimate corresponding pump-curve points under suitable similarity assumptions. They do not guarantee that installed flow changes in direct proportion to speed when the system includes static head. The U.S. Department of Energy pumping sourcebook explains why variable-speed operation needs particular care in systems with significant static head.
As another teaching calculation, reduce the speed ratio to r = 0.90 and approximate the new curve by Hp,r(Q) = r²Hp(Q ÷ r) = 40.5 − 0.01Q². Intersect it with the unchanged single-branch system: 40.5 − 0.01Q² = 10 + 0.015Q². This gives Q² = 1220, Q approximately 34.93 m³/h and H = 28.30 m. Simply multiplying the old 40 m³/h by 0.90 would give 36 m³/h, which is not this new intersection.
The calculation changes the pump curve while retaining the network's 10 m static rise. It does not prescribe an approved drive setting or assume constant efficiency. Confirm the actual reduced-speed curves, minimum permitted speed, motor cooling, suction conditions and control stability with the supplier. Retain the irrigation outlet requirements: a lower-power operating point is not useful if the sprinklers cannot provide the required coverage.
Verify the model with a repeatable field record
For each permitted combination, record the pump configuration and speed, active branches, valve positions, pumping source level, stabilized flow and pressure measurements. Use the irrigation flow-meter installation guide to check measurement quality. Compare pump differential total head with the curve, accounting for suction and discharge pressure, instrument elevations and velocity-head differences where relevant. A single discharge pressure reading is not automatically the pump's total head.
| Review item | Record | Decision supported |
|---|---|---|
| Pump configuration | Model, impeller or stages, speed and matching curve revision | Whether the correct performance curve is being used |
| Network state | Active zones, valve mode, source level and component condition | Which system relationship applies |
| Hydraulic result | Stabilized total flow, branch flow where needed and consistent head measurements | Whether the modeled intersection matches field behavior |
| Operating limits | Permitted range, suction assessment, power and critical outlet pressure | Whether the measured duty is acceptable for equipment and irrigation |
| Schedule evidence | Hours, delivered volume and measured input energy for each accepted combination | Whether alternative schedules provide the required service efficiently |
Investigate a discrepancy before changing the pump. Incorrect speed, a partly closed valve, an unexpected bypass, a damaged impeller or a changing source can produce different observations. Check the measurement boundary and instruments before assigning a cause. Repeating the same permitted state after stabilization helps distinguish a reproducible hydraulic difference from a transient or a reading error.
For an IrriNex component review, provide the pump curves, marked network drawing, zone combinations and completed comparison record. Ask for the selected pipe, valve and filter data needed to refine the system model. Approve a combination only after checking both the pump's operating limits and the water delivered at the active outlets; total pump flow alone cannot establish irrigation performance.
