Why Does a Suction Pump Form Vortices at the Intake?

A visible air-core vortex forms above the suction inlet, the pump becomes noisy, and delivered flow falls well below capacity. In the documented arrangement, two 18-inch HDPE suction lines moved a stated 5,300 gpm with 18 ft of suction lift and 5 ft of submergence. Raising the upstream level by 1.5 ft reduced the lift to 16.5 ft; the vortex disappeared and pump output increased from 35% capacity to more than 60%.

Reject the quick fixes until the operating point is known

Do not start by cutting slots, drilling holes, moving pipes, or treating every noise as cavitation. Each action addresses a different mechanism, and an intake near its minimum-submergence limit can change behavior after a small level adjustment.

Quick fix	Why it may fail	What to check first
Move the suction lines apart	Separation may reduce interaction between vortices, but it does not remove swirl already entering the manhole or correct inadequate submergence.	Run each pump separately and observe where rotation begins.
Add a vortex breaker	A breaker can interrupt an air core, but a crowded manhole may lack clearance. It also will not recover inadequate NPSH.	Measure liquid level, lift, branch flow, and approach-flow direction.
Cut slots or drill holes in the suction pipe	The modification changes inlet velocity distribution permanently. Poorly distributed openings can introduce local losses or draw air near the surface.	Prove that inlet concentration, rather than low liquid level or upstream swirl, is the controlling problem.
Call the noise cavitation	Air passing through the pump and mismatched flow into the impeller eye can also produce noise.	Determine whether an air core reaches the inlet and compare NPSH available with pump NPSH required.

Check before continuing: Record the untouched arrangement and baseline operating condition. If the vortex, level, lift, and individual pump flow are not recorded together, stop changing hardware.

Map the intake and establish the baseline

Sketch the manhole in plan and elevation. Show all suction pipes, the incoming line, pipe-end elevations, wall clearances, spacing, liquid level, pump elevations, strainers, fittings, and the direction from which flow enters. The installation contained three 18-inch pipes in the manhole, leaving no practical room for the proposed baffle; that physical constraint must be part of the decision.

1. Mark the measured submergence from the liquid surface to the controlling point at each suction inlet. Use the same reference point every time.
2. Measure static suction lift with the pump stopped and operating suction conditions with each pump running.
3. Observe the surface and subsurface approach flow. Note rotation that exists before an air core forms.
4. Run pump 1 alone, pump 2 alone, and then both together. Record level, flow, noise, vibration, suction pressure, and vortex location for every state.
5. Confirm whether 5,300 gpm is total system flow or flow through each 18-inch branch. Parallel pumps connected to a common header do not necessarily divide flow equally.

Incoming swirl can seed a vortex even when the inlet geometry and calculated minimum submergence appear acceptable. Pipe separation helps only when interaction or wall proximity is the cause; it does not straighten rotating approach flow.

Check before continuing: The sketch and test sheet must identify which pump or combination creates the vortex and whether rotation starts upstream of the suction openings.

Calculate velocity with the correct branch flow and diameter

Minimum-submergence methods based on velocity and methods based on inside diameter plus flow should agree only when supplied the same hydraulic inputs. Use actual HDPE inside diameter, not the nominal 18-inch designation, and use the measured flow in one suction branch rather than total flow unless one branch truly carries the full amount.

Calculate area and velocity as follows:

A = pi x D_i^2 / 4
v = Q_branch / A

For a labeled screening calculation only, assume the inside diameter is exactly 18 inches, or 1.5 ft. The area is then approximately 1.767 ft². A flow of 5,300 gpm equals approximately 11.81 ft³/s.

Flow interpretation	Branch flow	Calculated velocity with assumed 18-inch ID
5,300 gpm passes through one line	5,300 gpm	Approximately 6.68 ft/s
5,300 gpm splits equally between two lines	2,650 gpm	Approximately 3.34 ft/s per line
Parallel pumps split unequally	Measure each branch	Calculate separately; do not use the average

This factor-of-two velocity difference can move the required submergence materially. Recalculate with the measured inside diameter and individual branch flow, then plot the operating point on the same graph used during commissioning.

Check before continuing: The velocity calculation, graph, and calculator must use the same inside diameter and the same per-branch flow. Resolve any disagreement before modifying the intake.

Raise the level and prove the submergence margin

The installation operated at about 5 ft of submergence and was described as being on the calculated limit. A controlled level test supplied the fastest diagnostic: surcharging the main line by 1.5 ft reduced suction lift from 18 ft to 16.5 ft. The vortex disappeared, and flow rose from 35% to more than 60% of pump capacity.

1. Hold pump configuration and discharge restrictions constant.
2. Raise the upstream liquid level in controlled increments within the system's allowable limits.
3. At each level, record submergence, suction lift, individual flow, noise, and whether a surface depression or air core reaches the inlet.
4. Lower the level carefully toward the original condition to identify where the vortex returns. Avoid repeated operation that sends air through the pump.
5. Set the temporary operating level above the observed transition with allowance for normal level fluctuation, waves, and unequal branch flow.

A level increase changes two conditions at once: it provides more geometric submergence above the inlet and reduces suction lift. The test therefore proves that liquid level is effective, but the NPSH check must separate air-core suppression from suction-pressure recovery.

Check before continuing: Repeat the stable run at 16.5 ft lift or the approved equivalent level. The pass condition is no air core and sustained flow above the previous 35% condition.

Check NPSH before calling the problem solved

Vortexing and cavitation are different faults. A vortex can entrain air into the suction, while cavitation occurs when local pressure falls low enough for liquid to vaporize. The two can coexist, and both can reduce flow and damage the pump.

NPSH available = absolute pressure head at the liquid surface
               + static suction head
               - vapor-pressure head
               - suction-line and strainer losses

Use the correct sign for a suction lift: lift reduces NPSH available. Calculate with site atmospheric pressure, liquid temperature, actual suction-pipe losses, fittings, and clean or dirty strainer loss. Compare the result at measured pump flow with the manufacturer's NPSH-required curve at that same flow.

One installation estimate placed NPSH available below 10 ft at the original 18 ft lift before counting strainer loss. If the strainer consumed about 3 ft, the remaining estimate was 7 ft or less. Those figures are screening values from this installation, not substitutes for a site calculation. Raising the level by 1.5 ft adds approximately 1.5 ft of static pressure head when the surface and pump references remain otherwise unchanged.

If the air core disappears but noise, unstable suction pressure, or poor flow remains, inspect the strainer and recalculate NPSH. Where lift cannot be reduced and NPSH available remains inadequate, engineering review may consider a low-NPSH booster pump ahead of each main pump; it is a system redesign, not a quick field adjustment.

Check before continuing: Document positive NPSH margin using the pump curve and worst credible level, temperature, flow, and strainer condition. Stop the test if suction pressure becomes unstable or the pump continues passing air.

Correct approach flow before adding intake hardware

If adequate level removes the air core only temporarily, inspect the approach flow. Rotation entering the manhole can start vortex formation at the suction inlet. Straightening the incoming flow may be more effective than increasing pipe spacing.

1. Watch the intake while each pump operates alone. A vortex tied to one inlet suggests local geometry or branch flow; rotation across the whole manhole points toward incoming swirl.
2. Check whether the incoming pipe discharges tangentially or creates a circulating path around the manhole.
3. Use temporary, removable flow guides only where they cannot detach, obstruct the line, or create an unsafe surcharge condition.
4. Recheck the level and flow after every temporary geometry change.
5. Select permanent guides, strakes, or vortex breakers through a hydraulic review when observation proves that approach flow remains the cause.

Pipe slots with total opening area around twice the inner-pipe cross-sectional area were suggested for this arrangement, as were numerous drilled holes. That geometry was not reported as installed or verified. Do not cut an 18-inch suction line on that basis alone; opening placement, free area, structural strength, air-entry clearance, debris loading, and added losses require design.

Do not count the third pipe as a baffle without a controlled test. A solid obstruction can redirect circulation, worsen asymmetry, or trap debris as easily as it can interrupt swirl.

Check before continuing: Accept an approach-flow correction only when repeated runs show uniform entry, no persistent rotation, and stable branch flow at the required liquid level.

Verify the full operating range before returning service

1. Run each pump separately from startup through its required operating flow.
2. Run both pumps together and measure each branch rather than assuming equal division.
3. Test the lowest permitted liquid level and the highest expected suction demand.
4. Record submergence, lift, flow, suction pressure, discharge pressure, noise, vibration, and visible air entrainment.
5. Inspect and account for the strainer condition. A clean-strainer test alone does not represent the worst suction loss.
6. Compare every operating point with the minimum-submergence method and the pump's NPSH-required curve.
7. Reinspect temporary level-control or flow-straightening measures before leaving the system unattended.

Get production running with the proven 1.5 ft level increase only if the upstream system can safely tolerate the surcharge. Then develop the permanent correction around the measured cause: operating-level control, reduced branch velocity, approach-flow straightening, revised inlet geometry, or a reviewed booster-pump arrangement.

Final check: Pass only when the pump holds required flow without an air core, suction-pressure instability, or abnormal noise during single- and dual-pump operation at the lowest approved level.

Frequently Asked Questions