Watch a spinning impeller with cavitation bubbles forming at the low-pressure eye when NPSHa<NPSHr and collapsing in the high-pressure rim. Visualize how erosion, noise, and head breakdown arise.
Presets
Fluid & Pipe Conditions
Fluid Temperature T (°C)
°C
Static Head Hs (m)
m
Pipe Loss hf (m) @ design point
m
Pump Specifications
NPSHr @ design point (m)
m
Operating Flow Q (m³/h)
m³/h
Speed N (rpm)
rpm
Results (live)
—
NPSHa (m)
—
NPSHr (m)
—
Margin ΔH (m)
—
Cavitation status
—
Head ratio H/H₀
—
Pvap (kPa)
—
Inlet pressure (kPa)
—
Noise / erosion
Impeller & Cavitation Visualization
Blue = low-pressure impeller eye where bubbles form; they collapse (white flash) at the high-pressure rim, eroding the blade. Cavitation occurs only when NPSHa<NPSHr.
Blade Pressure vs Vapor Pressure
Where the blade-surface static pressure (blue), minimal at the inlet, drops below the vapor-pressure line (red dashed), cavitation occurs.
Head vs NPSH Curve (breakdown)
As NPSHa approaches NPSHr, head drops 3% (knee point); below NPSHr it breaks down steeply (cliff). ● marks the current operating point.
When local static pressure at the eye falls below vapor pressure, bubbles form; head breaks down once $NPSH_a < NPSH_r$.
What is Centrifugal Pump Cavitation (NPSH)?
🙋
In the animation, blue bubbles appear near the center of the impeller and flash white as they vanish at the edge. Is that cavitation?
🎓
Yes, that's the whole life cycle of cavitation. The impeller eye (inlet) is where velocity peaks and local static pressure is lowest. When NPSHa<NPSHr, the pressure there drops below the saturated vapor pressure, so the liquid "boils" at room temperature and forms bubbles. They ride along the blade out to the high-pressure rim and collapse the instant pressure recovers. That white flash is the implosion.
🙋
The bubbles just pop — why does that wreck the impeller?
🎓
When a bubble collapses, a localized micro-jet and shock wave of several hundred MPa hammers the metal surface. One hit is harmless, but hundreds per second at the same spot fatigue and chip the metal — that's erosion. In plants, neglecting it for a few months has been known to punch holes through impellers. Press the "Full cavitation" preset and you'll see the collapses concentrate at the rim.
🙋
In the bottom-right "Head vs NPSH" chart, there's a cliff where the ● drops suddenly. What is that?
🎓
That's head breakdown. As NPSHa approaches NPSHr, head first drops by 3% — the "knee point." In fact NPSHr is defined as exactly that 3% head-drop point. Below it, the blade passages choke with vapor and the flow collapses, so head falls off a cliff and the pump can no longer push water. Raise the temperature slider to drive the ● toward the cliff.
🙋
In the middle "Blade pressure" chart, when the blue line dips below the red dashed line (Pvap) it shades red. Is that linked to the bubbles?
🎓
Exactly. The blue line is the static-pressure profile from suction through the impeller eye to discharge, with its minimum at the inlet. Bubbles form only in the region (red band) where that minimum dips below the Pvap line. So P_local < Pvap is the physical condition for cavitation — which is just another way of stating NPSHa < NPSHr.
🙋
When I raise the speed slider, it seems like more bubbles appear. Is higher speed dangerous?
🎓
Sharp eye. Higher speed raises the inlet tip velocity, so local pressure drops further and the effective NPSHr increases, making cavitation more likely. That's why engineers add an inducer (a helical pre-stage) to pressurize first, or choose a slower large pump. In the animation, the impeller rotation and bubble spawn rate both scale with speed. On site, the rule is to act the moment the "Caution" badge appears.
Physical Model & Key Equations
Available NPSH (NPSHa) expresses the suction-side energy margin that prevents the fluid from vaporizing at the pump inlet, measured as liquid head in meters.
Cavitation condition (local static pressure $P_{local}$ at the eye falls below vapor pressure):
$$P_{local} < P_{vap} \iff NPSH_a < NPSH_r$$
$NPSH_r$: minimum NPSH required by the pump [m] (manufacturer catalog value, defined at the 3% head-drop point). When NPSHa falls below NPSHr, bubbles form at the eye and collapse at the high-pressure rim, and head breaks down sharply. Always keep a safety margin $M$ (typically 0.5 to 1.0 m).
Real-World Applications
Chemical plants and refineries: NPSH calculations are essential for pumps handling hot solvents or light hydrocarbons, where rising vapor pressure makes cavitation at the impeller eye likely. Underestimating NPSHa at design time leads to impeller erosion after start-up.
Building HVAC and district heating/cooling: Engineers use this check when selecting chilled-water or cooling-water circulation pumps and designing static head and pipe diameter so NPSHa stays adequate even in summer heat.
Power plants (condenser systems): Large seawater or condensate pumps suffer major damage from collapse shocks under cavitation, so designers include margins for tidal head changes and increasing pipe losses over time.
Food and pharmaceutical processes: Sanitary piping often has many valves and filters, increasing friction losses. Combined with high-temperature sterilization, vapor pressure rises and NPSHa can drop sharply.
Frequently Asked Questions
Where on the impeller do cavitation bubbles form and collapse?
Bubbles form on the suction side of the blade at the impeller eye (inlet), where velocity is highest and local static pressure is lowest. When NPSHa<NPSHr the pressure drops below the saturated vapor pressure and the liquid vaporizes. The bubbles travel along the blade toward the high-pressure rim and collapse (implode) where pressure recovers, eroding the blade with shock waves and micro-jets. The animation shows this form → travel → collapse sequence.
How much margin between NPSHa and NPSHr is safe?
General design guidelines recommend a margin of NPSHa ≥ NPSHr + 0.5 to 1.0 m. HI (Hydraulic Institute) recommends at least NPSHa ≥ 1.10 × NPSHr. For high-temperature liquids, volatile fluids, or systems with large flow fluctuations, a margin of 2 to 3 m may be used. This tool flags "Full cavitation / head breakdown" below 0 m margin, "Incipient" from 0 to 1.0 m, and "Safe" above 1.0 m.
Why does head drop sharply once NPSHa falls below NPSHr?
Because bubbles forming at the eye choke the blade passages. NPSHr is defined as the point where head drops by 3% — the knee point. As NPSHa falls further, vapor occupies more of the passages, the effective flow plummets, and head falls off a cliff. This is head breakdown. Check where the operating point (●) sits relative to the cliff on the "Head vs NPSH" chart.
Why does higher speed make cavitation more likely?
Higher speed raises the inlet tip velocity $u_1$, so $NPSH_r \approx \sigma_c u_1^2 / 2g$ increases, while the local static pressure at the eye drops further. NPSHa is then more likely to fall below NPSHr and the bubble count rises. In the animation the impeller rotation and bubble spawn rate scale with the speed slider. An inducer is effective for lowering NPSHr.
How can NPSHa be improved?
① Install the pump lower to gain static head Hs. ② Shorten and enlarge the suction pipe to cut loss hf (doubling the inner diameter cuts velocity to 1/4 and loss to 1/16). ③ Lower the fluid temperature to reduce Pvap. ④ Add a suction drum to raise suction pressure. ⑤ Select a pump with an inducer to cut NPSHr by 30 to 50%. Move each slider to see how the operating point (●) pulls away from the breakdown cliff.
Why is NPSH a problem at high altitude?
As altitude increases, atmospheric pressure Patm decreases. Since NPSHa = (Patm - Pvap)/ρg + Hs - hf, a drop in Patm directly reduces NPSHa. At 3,000 m, atmospheric pressure is about 70% of sea level, so the same piping arrangement loses substantial NPSHa and cavitation becomes likely at the impeller eye. Always apply altitude correction for liquid-transfer systems in mountainous regions.
Common Misconceptions and Points of Caution
People often assume that "if NPSHa is greater than NPSHr, cavitation will never occur." In reality NPSHr is defined at the 3% head-drop point, so incipient cavitation is already present there. Always add a safety factor (typically 1.0 to 1.5 m) and design with margin.
People assume that "vapor pressure depends only on water temperature," but the fluid type and dissolved gases matter too. For hot water or hydrocarbon fluids, applying a pure-water vapor-pressure table directly can introduce large errors.
People assume that "pipe loss is constant," but friction loss varies with the square of flow rate. NPSHa calculations must consider partial-load and start-up flow variations, not just the rated point, or you may miss unexpected cavitation risk.
