From atmospheric pressure, saturation vapor pressure, suction lift, and pipe friction loss, compute the available NPSH (NPSHa) and the safety margin against required NPSH in real time. Designed for pump-cavitation education and quick checks.
Parameters
Atmospheric pressure P_atm
kPa
Saturation vapor pressure P_v
kPa
Suction lift h_s
m
Suction-pipe friction loss h_loss
m
Fluid is assumed to be water (rho = 1000 kg/m^3, g = 9.81 m/s^2). Required NPSH (NPSHr) depends on the actual pump; this tool assumes a representative value of 2.0 m.
Results
—
Available NPSH
—
Atmospheric head
—
Margin (NPSHa − NPSHr)
—
Safety
Suction-side schematic
Liquid surface, suction pipe, and pump position. Arrows show h_s (suction lift) and h_loss (friction loss) to visualize the meaning of the available NPSH.
Suction lift h_s vs available NPSH
X = h_s (m) / Y = NPSHa (m). Red band = cavitation region (NPSHa < NPSHr). Yellow dot = current operating point.
Theory & Key Formulas
The available NPSH is the absolute pressure head at the suction side minus the vapor-pressure head, geometric lift, and pipe friction loss. It measures the cavitation margin.
$$\mathrm{NPSH}_a \geq \mathrm{NPSH}_r + M,\qquad M = \mathrm{NPSH}_a - \mathrm{NPSH}_r$$
$P_\text{atm}$: atmospheric pressure [Pa], $P_v$: liquid saturation vapor pressure [Pa], $h_s$: suction lift [m] (positive when pump is above the surface), $h_\text{loss}$: suction-pipe friction loss [m], $\rho = 1000$ kg/m³ (water), $g = 9.81$ m/s², $M$ is the safety margin (this tool flags $M > 0.5$ m as Safe).
What is the NPSH Simulator?
🙋
My pump-selection assignment says I need to verify NPSHa exceeds NPSHr, but what exactly is NPSHa computing?
🎓
It is the headroom in meters of liquid column that is still left before evaporation starts at the pump suction. The formula is $\mathrm{NPSH}_a = (P_\text{atm}-P_v)/(\rho g) - h_s - h_\text{loss}$: take the atmospheric pressure head, subtract the vapor-pressure head, then subtract the height the pump is lifted above the surface and the friction loss. Positive means margin remains; near zero means cavitation begins.
🙋
I have heard hot water cavitates more easily — what drives that?
🎓
The vapor pressure $P_v$ rises exponentially with temperature. Water at 20 deg C has $P_v$ = 2.3 kPa, but at 80 deg C it jumps to about 47 kPa. That makes the $-P_v/(\rho g)$ term hit hard, lowering NPSHa. Slide $P_v$ from 2.3 to 47 kPa in this tool: NPSHa drops by roughly 4.5 m. That is why boiler feedwater pumps almost always run with flooded suction.
🙋
Why is suction friction loss h_loss such a big deal? The suction pipe is short, right?
🎓
True, suction pipes are usually 5-10 m, but friction grows with the square of velocity. So pushing the flow up by 50% gives more than 2x the friction. For a 50 mm pipe at 1 m^3/min, you can easily lose several meters of head. The standard rule is "size the suction one pipe larger than the discharge." Elbows, strainers, and reducers all add up, so this tool collapses everything into one h_loss term.
🙋
Is a 0.5 m margin really enough? It feels like more is safer.
🎓
Textbooks call 0.5 m the bare minimum. API 610 and ANSI/HI typically ask for around 1.0 m. Once you account for NPSHr test uncertainty (about ±10%), drift in operating point, and temperature excursions that raise $P_v$, 1 m is a wise design target. This tool flags 0.5 m as the boundary between Safe and Risk, but you should aim for at least 1.0 m on real designs.
Frequently Asked Questions
Because atmospheric pressure $P_\text{atm}$ falls with elevation. Sea-level 101.3 kPa drops to about 89.9 kPa at 1,000 m, 70 kPa at 3,000 m, and around 64 kPa at the top of Mount Fuji (3,776 m). The first NPSHa term $(P_\text{atm}-P_v)/(\rho g)$ shrinks by about 3.7 m at 3,000 m, and that loss flows directly into your design margin. Mountain plants and high-altitude hydropower intakes must apply this correction. Slide slPatm down to 70-80 kPa in this tool to feel a high-altitude condition.
No — flooded suction (h_s < 0) helps, but it is not a guarantee. With $P_v$ at 50 kPa (around 80 deg C water), the vapor head alone consumes about 5.1 m, and adding 3 m of friction means even a -2 m suction lift only delivers about 4.5 m of NPSHa. If the pump needs NPSHr = 4 m, the margin is just 0.5 m — already in the Risk zone. Flooded suction is a strong tool but always requires a numerical check.
Through standardized cavitation tests per ISO 9906 or ANSI/HI 1.6. The suction pressure is gradually reduced while head is monitored, and the suction-side NPSH at which a 3% head drop occurs is recorded as NPSHr (also called NPSH3). NPSHr depends strongly on flow rate and inducer presence, so the manufacturer publishes an NPSHr-vs-flow curve. This tool assumes a representative 2.0 m, but for real selection you must consult the actual NPSHr curve from the pump vendor.
The basic formula is the same, but $\rho$ and $P_v$ must be replaced. This tool fixes $\rho$ at 1000 kg/m^3 (water), so for diesel ($\rho \approx 850$ kg/m^3) the head-to-pressure conversion shifts (1 m of column equals about 8.34 kPa instead of 9.81 kPa). LNG and liquid ammonia are pumped near saturation, so $P_v$ is essentially the operating pressure and securing NPSHa is very hard. For refrigerant systems the design metric usually switches to "subcooling from saturation" rather than NPSHa. This tool is fine for learning, but plant-scale design needs property-aware tools (Aspen HYSYS, etc.).
Real-World Applications
Boiler feedwater pumps in thermal and nuclear plants: Boiler feedwater near 180 deg C has saturation vapor pressure approaching 1 MPa, so NPSHa from an open tank at atmospheric pressure is impossible. The standard fix is a deaerator placed many meters above the pump, plus a booster pump in front of the main feedwater pump — a two-stage architecture engineered specifically to provide the required NPSHa. Push $P_v$ to the upper slider limit (50 kPa, around 80 deg C water) in this tool and you will see how negative h_s has to be before the safety zone returns.
Municipal raw-water intake pumps: When pumping from a river or lake, both the variable water level and the long pipeline directly affect NPSHa. Drought lowers the water level (raising h_s) while summer warms the water (raising $P_v$) — a very unfortunate combination. Designers compute NPSHa at "lowest water level + highest temperature" and mandate at least 1.0 m of margin above NPSHr. Sweeping multiple conditions in this tool is a quick way to find the worst-case NPSHa and confirm the design is sized for it.
Light-hydrocarbon pumps in petroleum refining: For naphtha, LPG, and other high-vapor-pressure liquids, securing NPSHa is always on the design critical path. API 610 mandates NPSHa >= NPSHr + 0.6 m for gasoline service or + 1.0 m for lighter cuts. During plant layout, the column-bottom-to-pump-suction friction loss and the nozzle elevation must both be minimized; quick lookups of the kind this tool offers are useful for early-stage trade studies.
Building water-supply and chiller-loop pumps: These are room-temperature systems, but the friction loss h_loss between a rooftop cooling tower and a basement pump room can dominate NPSHa. Engineers run this calculation at "110% of design flow" and "fouled-strainer" conditions and check that NPSHa still meets the requirement. The same model is useful when evaluating whether a proposed pipe-downsizing retrofit is acceptable.
Common Misconceptions and Pitfalls
The most common misconception is to assume that "if the pump sits below the liquid surface, cavitation cannot happen." True, h_s < 0 raises NPSHa, but with hot or volatile liquids the vapor pressure dominates and even flooded suction may leave little margin. At 80 deg C water, $P_v \approx 47$ kPa, the vapor head alone consumes about 4.8 m. A pump with NPSHr = 3 m therefore needs more than 7.8 m of flooded head once you add h_loss. Slide $P_v$ to 47 kPa in this tool and watch the NPSHa curve drop dramatically.
Next is the belief that NPSHa is independent of flow rate. In reality, raising the flow increases the suction friction $h_\text{loss}$ as the square of velocity, so NPSHa decreases with flow. At the same time the pump's NPSHr increases with flow. The two curves move in opposite directions versus flow, so above a certain flow rate cavitation is inevitable. That is why "the maximum useful flow of a pump is set by cavitation," and you can reproduce the trend in this tool by treating h_loss as a flow-rate proxy.
Finally, do not assume that "NPSHr = 2 m applies to every pump." Real NPSHr ranges from below 0.5 m for small canned pumps, through 2-5 m for medium centrifugal pumps, up to over 10 m for large boiler feedwater pumps. Inducer-equipped low-NPSHr designs and steeply rising NPSHr-versus-flow curves are vendor-specific. Use this tool only for educational sensitivity studies of NPSHa, and consult the actual vendor NPSHr-vs-flow curve when selecting real hardware.