Fire Water Tie-In: Hydraulic Proof, Not Distance Alone

Brian Holt7 min read
Other ManufacturerProcess ControlTechnical Reference
Licensed PE Working through this on a live machine? A Maine-licensed engineer can take it from here — included with IMD hardware, by the hour for everything else. Book an engineer

The usual quick fixes do not settle this design. A proposed 14-inch main, a larger pressure-switch setting, or an adequate tank volume addresses only one part of the duty. Prove the complete path from the existing tank and suction header to the compressor station at 2,000 gpm and 150 psig, then configure pump starting from measured static pressure.

Reject the quick fixes first

Quick fix Why it fails Required check
Accept the proposed 14-inch pipe from flow alone Nominal size does not establish inside diameter, velocity, friction loss, or surge response. Calculate the installed pipe, fittings, valves, elevation profile, and hydraulically remote endpoint.
Add 20 m of static head to the controller setpoint The grade difference may not equal the elevation difference between the pressure switch and the remote pressure reference. Friction also varies with flow. Survey the switch, minimum tank water level, pump centerline, and station reference elevations.
Use a remote pressure transmitter as a direct replacement A transmitter adds power, wiring, signal processing, failure-state, and controller-acceptance questions. Obtain acceptance from the fire-pump controller manufacturer and authority having jurisdiction before changing the primary start circuit.
Rely on adequate tank volume Stored volume does not prove suction-header capacity, available pump head, or remote residual pressure. Check the full system at the governing simultaneous demand.
Raise jockey-pump pressure The jockey pump maintains standby pressure; it does not establish the fire-pump duty at 2,000 gpm. Coordinate jockey and fire-pump settings after measuring system static pressure and leakage behavior.

Identify the real hydraulic constraint

The 900 m separation is not an automatic rejection. The controlling question is whether the shared source and new pumping arrangement can deliver the required flow and residual pressure without unacceptable suction conditions or interference from other plant demands.

Use the energy equation across the whole path. At the design flow, the pump must overcome the required station pressure, elevation rise, suction and discharge friction, minor losses, and applicable velocity-head differences. Using water at approximately 1,000 kg/m³, a 20 m rise represents:

Delta P = rho × g × Delta z = 1,000 × 9.81 × 20 = 196,200 Pa ≈ 28.4 psi

If the true hydraulic endpoints differ by 20 m, the lower endpoint needs approximately 150 + 28.4 = 178.4 psig before adding pipe, valve, fitting, and equipment losses. Treat that as a screening value. Replace the station grade and tank grade with surveyed elevations of the actual pressure references.

Do not calculate main loss from a nominal 14-inch label. Obtain actual inside diameter, material or roughness basis, total developed length, fitting quantities, valve data, branch flows, and the station endpoint definition. Model the 900 m route at 2,000 gpm and at any larger governing combined demand.

Prove the shared source and suction header

  1. Confirm the required fire scenario, demand duration, 2,000 gpm flow, 150 psig residual-pressure location, and any simultaneous existing-plant demand with the authority having jurisdiction.
  2. Use the minimum operating tank level, not the nominal or full level. Verify that usable volume covers the approved scenario while preserving any required unusable volume.
  3. Survey tank water levels, pump suction and discharge elevations, the entire main profile, and the station test point.
  4. Model losses through the existing suction header, tie-in, isolation valves, strainers or other installed components, proposed main, fittings, and station distribution piping.
  5. Plot the resulting system curve against the selected fire-pump curve. Check the operating point at the required demand and at other required pump test points taken from the approved design basis.
  6. Calculate net positive suction head available at minimum tank level and maximum required flow. Compare it with the pump manufacturer’s requirement and check suction pressure throughout the shared-header operating cases.

Stop here if the existing suction header cannot feed the new pump without degrading an existing fire-water duty. A larger remote main cannot correct a restricted common suction path.

Set pressure sensing from actual reference points

Relate every pressure setting to its measurement elevation. For flow from a local switch toward the higher station:

P_remote = P_switch - rho × g × (z_remote - z_switch) - Delta P_loss(Q)

At no flow, main friction is approximately zero, so the static relationship is primarily elevation-dependent. If the station pressure reference is exactly 20 m above the switch, a remote threshold corresponds to a local pressure approximately 28.4 psi higher. Do not apply that correction until the switch elevation and required remote threshold are fixed.

Do not add design-flow friction loss to the automatic-start setpoint as though that loss exists during standby. The switch normally sees static or low-flow conditions before the fire pump starts. Establish the start sequence by measuring normal static pressure, jockey-pump cut-in and cut-out behavior, credible leakage pressure decay, and the pressure that represents a real demand.

A station transmitter can provide valuable indication, alarming, and test data. Using it as the primary automatic-start input is a separate design decision. Confirm that the controller accepts the signal arrangement, define behavior for loss of instrument power or signal, supervise the circuit, protect its routing, and obtain approval before replacing the controller’s pressure-sensing method. Never bypass the listed controller logic to get the system running.

Commission the complete duty at the station

  1. Calibrate the local pressure instruments, station instrument, and flow-measurement device against traceable references.
  2. Record tank level and static pressures at the pump and station before flowing water. Compare their difference with the surveyed elevation.
  3. Flow the station system to 2,000 gpm. Verify at least 150 psig at the exact point defined by the fire scenario.
  4. Record pump suction pressure, discharge pressure, flow, station pressure, tank level, and driver operating data at the same time.
  5. Compare measured differential pressure with the hydraulic model. A larger loss points to an incorrect pipe diameter, valve position, obstruction, unmodeled fitting, instrument error, or unaccounted branch flow.
  6. Create controlled pressure decay and confirm jockey-pump and fire-pump starting in the approved sequence. Test each automatic input and the defined response to an instrument or signal failure.
  7. Repeat the governing shared-header case with applicable existing demand. Confirm that neither system falls below its approved duty.

Do not accept a churn-pressure test as proof of station performance. It verifies a no-flow condition and cannot expose the loss in the 900 m main or common suction header.

Prevent the recurring field failures

  • Lock or supervise required valve positions according to the approved fire-water design.
  • Mark every pressure value with its location, elevation, flow condition, and instrument identifier.
  • Retain the pump curve, system curve, elevation survey, controller settings, and commissioning readings together.
  • Investigate disagreement between local and remote pressure before altering setpoints.
  • Recalculate the system after any pipe reroute, valve change, added consumer, tank operating-level change, or suction-header modification.
  • Use an approved impairment plan while the tie-in affects the existing plant fire-water source.

FAQ

What happens if the 14-inch fire-water main is accepted without a hydraulic calculation?

The system may deliver 2,000 gpm but miss 150 psig at the station because nominal diameter alone does not account for actual bore, 900 m of pipe, fittings, valves, elevation, or common-header losses.

What happens if I add 28.4 psi directly to the fire-pump start setting?

That correction is valid only when the pressure switch is exactly 20 m below the remote reference and the comparison is made at negligible flow. Survey both elevations and coordinate the resulting setting with the jockey-pump sequence.

What happens if the remote transmitter signal fails?

The response depends on the approved controller architecture and its defined failure state. Test loss of signal and instrument power during commissioning; do not use the transmitter as the primary start input without controller-manufacturer and authority approval.

What happens if the tank volume is adequate but the suction header is undersized?

The new pump can suffer low suction pressure or inadequate net positive suction head while the station loses residual pressure. Model and flow-test the shared header at the governing combined demand.

What happens if the station cannot hold 150 psig at 2,000 gpm?

Stop acceptance and isolate the cause with simultaneous suction, discharge, flow, tank-level, and remote-pressure readings. Stop work and contact the fire-pump controller or pump manufacturer through official support when measured behavior conflicts with approved data, and escalate unresolved compliance questions to the authority having jurisdiction.

Back to blog