Jim Elsey explains why you can't just wish and hope for pump performance.
by Jim Elsey
August 20, 2019

"Wishin’ and Hopin’”—both Dionne Warwick and Dusty Springfield had hit versions of this pop song in the early ’60s. The lyrical message is that you will not get want you want if you just sit around wishing and hoping; instead you need to take action to achieve the desired outcome.

The same advice holds true for pump performance. I witness an alarming number of people who unwittingly wish and hope their pump would perform in the proper manner, but they are wishing and hoping with total disregard of the system curve, pump capabilities, the laws of physics and the fluid properties. Time to take action.

I start my pump training courses with the simple personification that “pumps are stupid.” Put a centrifugal pump in any system, and it does not know where to operate. It is the system, not the pump, that dictates where the pump will operate on its performance curve—if the pump is even capable of operating at that point.

Further, and intended only as a comedic anecdote to help my students learn, I refer to the pump as the “husband” in this marriage with the system curve, aka the “wife.” Perhaps, and with ostensible apologies to the “PC Police,” I suggest that the marriage works best if the husband (pump) listens and obeys the wife (system). If there is a mismatch in the two, then divorce (pump and system failure) is imminent.

The pump will operate where its performance curve intersects the system curve, but we don’t always know where that point is—and just to complicate matters, it can change quickly because of many variables. Two roadblocks that make this determination difficult are:

  1. We frequently have no idea of the system curve geometry
  2. The pump is often forced off its published curve by outside factors

We will address the system curve calculation in a future article. For now, the system curve is the absolute summation of the system’s static head, pressure head, velocity head and friction head. The geometry of the system curve is directly related to the flow rate, pipe size, elevation changes and losses due to friction of all the components in the system. Note the system curve is dynamic and will change with tank elevation and system pressure changes. It will also change with valve position, system age, fouling and corrosion. This month’s article will address No. 2: How pumps can operate off their published curves, and we will look at a few common examples.

Causes for Operating Off Curve

Here are some of the common issues:

  • worn clearances
  • different or incorrect size impeller
  • different or incorrect speed
  • viscosity not corrected or accurate
  • net positive suction head (NPSH) margin insufficient
  • air entrainment and/or dual phase liquids beyond 3 percent
  • inadequate submergence (also see air entrainment)
  • partial or restricted blockage of the suction line
  • operating the pump in the wrong direction

First Things First

To determine where your pump is really operating, you will need to calculate the pump performance curve in the field “as is.”

First, obtain a copy of your pump performance curve as published or purchased. Then, using the pump’s discharge valve position, create a series of several different flow rates (recommend at least six points including shutoff head), determine and record the suction and discharge pressures for each flow condition, convert these pressures to differential head and plot them on your curve. Be careful to correct for gauge elevations, temperatures and specific gravities.

Does your curve match the published curve? If it does within 5 to 10 percent, then there is likely no problem.

Let’s look at a few cases where the curve does not agree with the published curve. Each case tells a critical story to help understand what is happening with your pump and the system.

Worn Clearances

As a pump wears, the hydraulic performance deteriorates. Most people understand the pump efficiency will drop due to wear, causing the power to increase, but not all users realize that the head and flow will also deteriorate.

Note the revised pump curve still intersects the system curve, but the meeting point is at a lower flow rate and a lower head. See my Pumps & Systems articles from January 2016 and July 2017 for more details on this subject (Image 1).

effect on pump performanceImage 1. Effect on pump performance with wear, speed, impeller size (Images courtesy of author)

It is important to understand if you are also plotting power on the curve: if the wear is simply the impeller or casing wear rings, the power will increase noticeably. But if the wear is in other areas such as internal passages, cutwater or the impeller vanes, the power will only increase a small amount.

A special note for ANSI pump designs that use the critical clearances between the impeller and the casing or stuffing box: the effect on power will be as the aforementioned pump with wear rings—that is, the power will increase noticeably.