This simulator models a twin-pump station with independent suction branches, branch check valves and flowmeters feeding a common discharge header, showing how two pumps operating in parallel share a common header head while their individual flows add at the combining junction — and why adding a second pump usually increases total flow by less than double.
• A real-time 3D cutaway workbench (common suction manifold, Pump A and motor, Pump B and motor, individual branch nonreturn check valves, two branch flowmeters, combining discharge header, and a station flow/power recorder) with home view, focus-selected-part, show full enclosure, exploded view, auto-rotate, expand and hide-labels scene tools. • A dedicated 'Combined pump and system curves' chart showing both branch curves and the combined system curve. • Experiment controls: Pump A speed, Pump B speed, Pump A enabled and Pump B enabled checkboxes, system static head, common system resistance, branch A resistance and branch B resistance sliders, plus pause/resume, single-step and 1 s-step buttons, four playback speeds, and dedicated 'Stop pump B' / 'Enable pump B' actions. • A Curves & measurements analysis tab with two further live charts (branch A flow, branch B flow and total flow; header head), the underlying per-branch and combined equations, and snapshot readouts (branch A/B flow, combined flow, header head, delivering pump A/B head, branch A/B loss, estimated combined shaft power, branch A/B check-valve state). • An Experiments tab with four guided fixtures (matched pair, one pump stopped, mismatched speeds, restricted branch B) and a Model verification bench with a timestamped event log and copyable trial report. • A Learn & assess tab with four guided lessons, a knowledge-check quiz with reset, and a written model-scope statement linking to a KSB pump-selection reference.
Each pump branch must overcome the same downstream header head plus its own branch resistance loss: Hi(qi) = 40(ni/1800)² − 0.006qi², solved for qi against Hheader + branch loss. The combining junction then simply sums the branch flows, Qtotal = qA + qB, while the header head itself is set by the common system curve Hheader = Hstatic + Ksystem·Qtotal². Because system head rises with total flow, adding a second matched pump typically increases total delivered flow by less than a factor of two — the matched-pair experiment demonstrates this directly on the combined curve chart.
If a branch's pump is stopped, disabled, or simply too weak (lower speed) to overcome the header pressure at the solved operating point, the simulator's ideal check valve for that branch closes and reports zero forward flow — isolating it rather than allowing reverse flow to circulate backward through it. The mismatched-speeds and one-pump-stopped experiments both show this isolation behavior, and the branch's displayed pump head reads as zero in that state, representing an isolated branch rather than a measured casing pressure.
This is a steady incompressible model with synthetic monotonic pump curves, ideal nonreturn valves and no common-suction limitation or staging-transient behavior. Shaft power assumes a fixed illustrative 75% efficiency for delivering branches only, excluding idle electrical losses for stopped or isolated pumps.
Branch flows add together at a common header head — each pump's individual flow contribution is found where its own head curve meets the shared downstream header pressure, and those flows are then summed at the combining junction.
Because system resistance rises with the square of total flow, the operating point moves up the (fixed) system curve as combined flow increases. The header head required at higher total flow is greater, which reduces each pump's individual contribution compared with running alone — so total flow typically increases by less than a factor of two.
Its branch check valve is modeled as closing, isolating that branch with zero forward flow rather than allowing reverse flow to pass backward through the stopped or overpowered pump. The simulator reports zero displayed head for an isolated branch, since it represents isolation rather than a measured casing pressure.
Yes — even with identical pump speeds, a branch with higher resistance delivers less flow than a branch with lower resistance at the same header head, as shown in the restricted-branch-B experiment.