The panel shows about 15 psig at the booster discharge and -4 psig at its suction while the triplex pump is running. Start here: the booster is developing about 19 psi, because pump head is the pressure rise across the pump, not the discharge gauge reading alone. The low discharge reading comes primarily from suction pressure falling below atmospheric under flow.
Read the two gauges together
Calculate the boost from suction to discharge before comparing the operating point with the pump curve:
Developed pressure = discharge pressure - suction pressure
For the recorded condition:
15 psig - (-4 psig) = 19 psi
A curve value of 15 m of water corresponds to about 21 psi of developed pressure. The observed 19 psi rise is therefore close to the stated curve expectation near 115 m3/h and approximately 1245 RPM. A direct comparison of 15 psig discharge with 21 psi curve head incorrectly treats the suction gauge as zero.
Gauge pressure may be negative while absolute pressure remains positive. Using the stated atmospheric reference of approximately 14.5 psia, a suction reading of -4 psig is approximately 10.5 psia. A discharge reading of 15 psig is approximately 29.5 psia; the difference remains 19 psi.
For a rigorous comparison, use total developed head:
H = (Pdischarge - Psuction)/(rho × g) + (vdischarge² - vsuction²)/(2 × g) + (zdischarge - zsuction)
If both gauges are close to the pump and the pipe diameters and gauge elevations are similar, the velocity and elevation corrections may be small. Calculate them when the taps differ materially in elevation or pipe size. Apply the pumped liquid density rather than assuming water if its specific gravity differs.
Match each symptom to its cause
| Observed symptom | What it indicates | First check |
|---|---|---|
| 15 psig discharge and -4 psig suction under flow | The booster develops about 19 psi; suction-system loss is subtracting from the discharge gauge pressure. | Calculate discharge minus suction before changing the pump. |
| Deadheaded booster runs within specification | The pump can produce its shutoff head, but the test says little about suction losses at 115 m3/h. | Repeat measurements at the actual operating flow. |
| Another booster pump gives the same result | The shared suction piping, screen, filter, tank head, instrumentation, or downstream demand is more likely than an individual pump defect. | Measure pressure across each common suction component. |
| Suction is near 1.5 m or 5 ft of static head at no flow but falls below zero while running | Flow-dependent friction is consuming the available gravity head. | Inspect the inlet line and screen or filter for excessive resistance. |
| Pressure improves at deadhead but drops when the triplex runs | The operating point has moved from zero flow to the required flow, while inlet losses have risen. | Plot the running flow and developed head on the curve. |
| Speed increase raises noise or produces unstable pressure | Higher flow has increased inlet loss and may be reducing cavitation margin. | Stop increasing speed and check suction absolute pressure and NPSH. |
Understand what the triplex pump changes
The triplex pump does not directly erase booster head. It draws flow from the booster discharge. That demand moves the booster away from deadhead and onto its running point, where its curve supplies less head than at zero flow.
The flow also passes through the booster suction line, screen, filter, fittings, and tank connection. Their losses rise as flow rises. With only about 1.5 m or 5 ft of static tank head available, those losses can pull the booster suction from a positive no-flow value to -4 psig. The developed head can remain near the curve while the discharge gauge falls by the amount lost at the suction.
The triplex is a pulsating load. A gauge may display an average, a damped value, or pressure peaks depending on its response. That is not the first fault to chase: the simultaneous readings already identify a large suction depression. Check pulsation only after validating gauge accuracy, tap location, and the steady or averaged flow.
Deadheading changes both sides of the test. With no flow, friction through the inlet system approaches zero, suction pressure recovers toward static tank head, and the booster moves to its shutoff-head point. A good deadhead result proves the pump can build zero-flow head; it does not prove that the inlet can supply 115 m3/h.
Prove the operating point
Confirm both gauges use the same pressure reference and engineering units. Check their zero readings with the system depressurized, and use pressure instruments with suitable range and accuracy.
Place the suction measurement as close as practical to the booster inlet and the discharge measurement close to its outlet. Record tap elevation and pipe diameter if they differ.
Run the booster and triplex together at the required condition. Record booster speed, actual flow, suction pressure, discharge pressure, tank liquid level, and the condition of the screen or filter at the same time.
Calculate developed differential pressure. At the recorded readings, use 19 psi, not 15 psi, for the first curve comparison.
Convert pressure rise to liquid head using the actual density, then add velocity and elevation terms where needed. Plot that result at the measured flow and speed.
Measure pressure immediately upstream and downstream of the filter or screen. A large component differential identifies where the available static head is being consumed.
Check the pump manufacturer’s required suction conditions and NPSH requirement at the measured flow and speed. Calculate available NPSH from absolute suction conditions, liquid vapor pressure, velocity head, and installation elevation.
If the actual flow is not measured, the assumption that booster flow equals the triplex rate may be wrong during bypass flow, leakage, recirculation, accumulator charging, or transient operation. Establish the flow balance before using 115 m3/h as the curve coordinate.
Remove suction restriction first
Increase the inlet line area and use a much larger screen where the existing components cannot pass the required flow from the available gravity head. The target is not merely a clean-looking filter. The complete tank-to-booster path must deliver the required flow without consuming enough head to pull the pump inlet below its acceptable suction condition.
Clean and inspect the screen or filter. Look for a collapsed element, obstructed tank outlet, partially closed valve, fouled fitting, or debris at the entrance.
Measure the differential pressure across the screen or filter at operating flow. Replace or enlarge the component when its loss consumes an excessive share of the available 5 ft static head.
Review suction-pipe diameter, length, reducers, bends, valves, and entrance geometry. Reduce avoidable restrictions and eliminate air pockets or leak paths on portions operating below atmospheric pressure.
Raise the tank when the process layout permits. Added elevation increases static suction head directly; a few metres can materially change the suction gauge reading.
Retest at the same tank level, booster speed, and triplex flow. Compare suction pressure, filter differential, developed head, and discharge pressure with the baseline.
Replacing the booster again wastes time when a second pump produces the same condition. Adjusting only the discharge side also misses the measured loss upstream of the pump.
Increase speed only after restoring suction
Running at 1600 RPM was identified as a route to approximately 25 psi discharge after the inlet line and screen can supply the required flow. Treat that as a system condition, not an isolated speed command. If you accelerate the existing restricted arrangement, the screen and piping loss will rise, suction pressure can fall further, and cavitation may begin.
Use the affinity laws only as a first estimate for the same pump and impeller:
Flow is proportional to speed.Head is proportional to speed squared.Power is proportional to speed cubed.
The final operating point comes from the intersection of the speed-corrected pump curve and the system curve. Before commanding 1600 RPM, check the published pump curve, allowable speed, driver power, motor current, coupling limits, and required NPSH. Read those values from the applicable pump and driver documentation; the installation data does not supply their limits.
For a 25-30 psig booster discharge while suction remains at -4 psig, the pump must develop approximately 29-34 psi before velocity and elevation corrections. Restoring suction to 0 psig reduces that requirement to approximately 25-30 psi. This is why recovering suction head can be as important as selecting more pump head.
Verify the correction under load
Operate at the normal tank level and required triplex flow. Do not use a deadhead test as the acceptance condition.
Record stabilized suction and discharge readings simultaneously. Calculate developed head and confirm that the point falls on the appropriate speed curve within instrument and curve tolerances.
Verify that suction absolute pressure and calculated available NPSH meet the pump manufacturer’s requirement for the operating point. Listen for gravel-like noise and watch for unstable pressure, vibration, or loss of flow.
Confirm the filter or screen differential remains acceptable when clean and through its intended service interval. A clean-only pass can hide a design with no fouling allowance.
Confirm the triplex receives
25-30 PSIat the location that matters, not merely at the booster flange. Include pressure loss between the booster and triplex inlet.Repeat at the lowest normal tank level and highest required flow. Those conditions provide the smallest static suction head and the greatest inlet loss.
Avoid the recurring mistakes
Do not compare discharge gauge pressure directly with a total-developed-head curve. Subtract suction pressure and correct for velocity, elevation, and liquid density where applicable.
Do not use deadhead performance to approve a suction system. Flow-dependent restriction disappears during that test.
Do not increase speed before checking inlet loss and NPSH. More speed can turn a pressure complaint into cavitation or driver overload.
Do not assume the nominal triplex capacity is the measured booster flow. Account for every branch, bypass, leak, and recirculation path.
Do not diagnose from gauges read at different times. The triplex load and tank level can change both values.
Do not overlook a suction-side air leak. A line operating below atmospheric pressure can admit air without leaking liquid outward.
FAQ
Can I compare 15 psig discharge directly with a 21 psi pump curve?
No. With -4 psig at the suction, the booster develops 15 - (-4) = 19 psi; compare that differential, with any required velocity and elevation corrections, against the curve.
Does the triplex pump reduce booster discharge pressure?
It creates the flow demand that moves the booster off deadhead and increases losses through the inlet pipe, screen, and filter. The measured suction depression then subtracts directly from the discharge gauge pressure even while developed pump head remains near the curve.
Can I raise the booster speed to 1600 RPM now?
First enlarge or correct the restricted inlet path, measure the screen differential, and verify available NPSH, driver load, and the manufacturer’s speed limit. Stop if suction pressure continues falling, pressure becomes unstable, or cavitation symptoms appear; escalate to the pump manufacturer’s official support channel when the measured operating point still misses the published curve or required limits cannot be identified.