I work at an upper Midwest water district. We are experiencing high vibrations from one of our vertical drywell sump pumps (see Figure 1).

A rubber flex joint is installed on the suction pipe leading to the pump. The flex joint has developed some pinhole leaks. We can see the leaks dripping under the flex joint when the pump is off and charged with water. We believe these pinhole leaks are entraining air into the flow when the pump operates. Could the entrained air be the cause of the cavitation?

Hoping to hear from you,

Michael P.
South Dakota

Hello Michael,

I perceive two issues in your message: high vibrations in your vertical drywell sump pump and cavitation.

A pump can vibrate for many reasons, including cavitation. One type of cavitation is air entrainment, so let’s say your drywell pump is suffering vibrations because of entrained air cavitation.


Figure 1. Michael’s pump system diagram. Courtesy of Michael P.

When a baby’s diaper fills and begins to leak, it is time to change the diaper. Likewise, pinhole leaks in the flex joint indicate the elastomeric component is rotten or at the end of its useful service life. The flex joint drips onto the floor when the pump is off and entrains air when the pump operates.

If you are not ready to purchase and install a new flex joint, you can purchase a spray can of rubber cement at any industrial supply store. (You may have seen the rubber sealant ads on TV.) Another option is to wrap the joint with plastic food wrap or duct tape to seal it. If the cavitation noise goes away, this was the culprit, and you still need to change the joint.

Cavitation magic

The cavitation noise may become even more pronounced after you stop entraining air. Let me explain with the following example. When the magician closes the curtain and his assistant disappears on stage, the assistant doesn’t really disappear, but steps through a trap door in the wall or floor before the curtain opens. There really is no magic. Likewise, an old trick used by seasoned mechanics for decades can hide or mask the cavitation noise.

I was a mechanic in the U.S. Navy about 45 years ago. I remember a particular boiler in the basement of a Navy hospital.  The boiler feedwater pumps were in violent cavitation from the deaerator tank. The noise was obvious and loud.  The boiler tech said, “Hey watch this!” He opened a little instrumentation valve on the suction side of the pump and allowed air into the flow. The cavitation noise was suddenly muffled, as though the pump had been wrapped with layers of thick insulation. The boiler tech thought he made the cavitation go away (but I didn’t). Purposefully entraining small quantities of air into the suction flow of a pump suffering obvious cavitation does not make the cavitation go away, and it doesn’t resolve the root cause of the cavitation. However, the energy is diverted to expanding and compressing the little air bubbles, rather than vaporizing and imploding the vapor pockets against the pump internals. A few months ago, I had the occasion to demonstrate this technique while diagnosing cavitation on some desalination pumps in Dubai, United Arab Emirates (see Image 1).

You still need to resolve the root cause of the cavitation.  Have your pump curve handy. Next, install suction and discharge pressure gauges on the pump with a flow meter. The gauges will indicate your pump’s position (head and flow) on the curve. The curve will also indicate the pump’s suction energy requirement, the net positive suction head required, at the pump’s duty point. Then, be sure this amount of energy plus a generous safety margin — net positive suction head available (NPSHa) — arrives into the pump. What is a generous safety margin?

You will want the NPSHa to be 5 feet or more above the NPSHr for pumps with an impeller up to an 8-inch diameter, rotating at 1,800 rpm.

You will want the NPSHa to be 7 to 10 feet or more above the NPSHr for pumps with an impeller up to a 13-inch diameter, rotating at 1,800 rpm.

You will want the NPSHa to be 15 to 20 feet or more above the NPSHr for pumps with an impeller greater than 13 inches and even smaller pumps on high speed motors (3,600 rpm).  The cavitation with cavitation-induced vibrations will go away.

Cavitation is only one type of pump vibration. You must deal with other vibrations if cavitation is no longer the issue with the pump. If I continue writing about vibrations, I’ll fill the pages of this edition of Flow Control.

So, I’ll say that some vibrations are induced or caused by inadequate pump design. For example, the classic cavitation discussed earlier in this article is a design deficiency. Figure 1 shows that your vertical pump is standing on four independent legs. This is another design weakness.


Image 2. The I-24 bridge across the Ohio River in Paducah, Kentucky. Courtesy of Bachus Inc.

Most mathematically inclined students will take a course in geometry. A geometric constant states: Any three points will define a plane. This is the reason civil engineers and surveyors mount their equipment (theodolites, levels and transits) onto tripods instead of quadripods. This is the reason most bridges and sports stadiums are constructed with triangular forms.

A vertical pump standing on three legs will always be stable. Engineers study these principles in school. Then, some design engineer puts four legs (or feet) under pumps and motors.

Other vibrations are induced by inadequate or incorrect pump operation. Starved and deadheaded pumps will vibrate. Vibrations will rise if the pump operator allows the pump to migrate or drift to the left or right extreme of the performance curve. This is like the driver (operator) allowing his car to drift into other lanes on the highway.    

And, still other vibrations are induced by inadequate or incorrect pump maintenance. A rebuilt pump will vibrate if the shop mechanic installs a bent shaft or unbalanced impeller into the pump while in the shop. Vibrations will rise with misalignment between the pump shaft and the motor shaft.

Michael, I suggest you investigate vibration equipment. Maybe a vibration sales rep can apply equipment to your pump and show you the root cause, possibly misalignment or imbalance, of the errant vibrations.

Good luck with your pump.

Flow Control will present the Pump Guy Seminar June 7-9 in Indianapolis. For more details and to register, visit flow
controlnetwork.com/pumpguy or contact Ashley Davidson at adavidson@grandviewmedia.com or 205-995-1593.

The Pump Guy is Larry Bachus, a pump consultant, lecturer and inventor, based in Nashville, Tennessee. Bachus is a retired member of the American Society of Mechanical Engineers, and lectures in both English and Spanish. You may contact him at larry@bachusinc.com. For more information, visit bachusinc.com.