I was recently hired as reliability engineer at a Texas power plant. My first task is to improve pump reliability at our site.
Our rotating equipment is composed of 252 pumps, 30 gear boxes, 20 fans, six compressors and four steam turbines. Pumps represent about 65 percent of all rotating equipment at this fossil fuel steam plant. The mean time between failures (MTBF) has been steady at 24.5 months for a few years.
I can use some refresher training on pump system basics. Do your Flow Control Pump Guy Seminars touch on pump types (other than centrifugal, positive displacement, etc.)? Does Flow Control provide Pump Guy Seminars on site?
Thanks for writing. Yes! I cover pump types and much more in the Flow Control Pump Guy Seminars. I’ll say more about this in the next few paragraphs.
About 10 percent of all pumps in a power plant are metering and injecting fuel oil, hydraulic oil and water treatment chemicals. Approximately 90 percent of the pumps in a power plant (about 225 pumps in your facility) are moving hot and cold water.
We give different names to the water in a power plant — "condensate return," "feedwater," "make-up," "chill water," "raw water" or "heavy water" (in nuclear plants). It is basically hot and cold water in various forms. And yet experience indicates many power plant pumps are problematic with excessive vibrations, overheating, cavitation, and premature bearing and seal failures. Certainly, a few pumps operate for years without problems. But too many power plant pumps are out of control.
All pump maintenance falls into one of two categories, non-repetitive pump maintenance and repetitive pump maintenance. Start with the pumps that require repetitive maintenance.
Reliable pumps (and all reliable equipment) result from the combined effort of design, maintenance and operation. A deficiency in one or more of these three disciplines results in continual pump failure.
The reliability engineer’s first goal is to pinpoint the deficiency as either design induced, maintenance induced or operation induced. Next the engineer must take steps to resolve the deficiency. Here is an example.
For more than a year, I communicated with a maintenance engineer at a plastics resin plant in Mexico. The cartridge seals in three hot oil pumps had begun failing (leaking) prematurely. One pump suffered 26 mechanical seal failures in four years, including seven failures in the first five months of 2015.
The engineer reluctantly saved a box of failed mechanical seals awaiting my visit. He had never considered failure analysis as a tool to resolve problems. He didn’t learn this in school. In June 2015, I showed the engineer how to disassemble and inspect his cartridge seals. The hot oil had spewed through the o-rings on all the seals in the box.
We disassembled a new, unused cartridge seal. My handy o-ring identifier revealed the elastomeric composition was incorrect. The spec sheet on the purchase requisition clearly stated the o-ring composition differed from the design spec.
We began asking questions. In an effort to conserve costs, the purchasing agent had changed mechanical seal suppliers. He didn’t change seal brands. He started importing pirated seals. On initial observation the copycat seals appeared identical to his original design cartridge seals.
This isn’t a maintenance or operations failure. This is a design failure. It is also a communication failure. Is the purchasing agent allowed to change the specification on a critical pump component? How does everyone tolerate this for four years without notice or action?
Sufficient evidence has not been gathered yet to declare a total success. These are only three pumps of 450 pumps in the resin plant. But the original cartridge seals with the proper o-rings have been installed on the three hot oil pumps. Since June 2015 no leaks have occurred.
To review, the reliability engineer’s first goal is to identify and resolve the pump failures caused by inadequate design, improper maintenance or poor operation. The reliability engineer’s second goal is to balance the pump’s energy with the ever-changing system energy.
For this reason, Patrick, I urge you to get training in pump system basics, rather than training in pump basics. The pump itself is a really simple machine. The pump has one moving part, the shaft assembly. The pump’s singular duty is to inject energy into the liquid. This injected energy is absorbed or consumed by the pipe system. The system’s energy constantly varies. What do I mean by this?
A certain quantity of energy is required to move a liquid through 200 ft of straight pipe. The pump develops this energy. The energy is consumed as friction in the pipe. This energy increases as the pipe takes on scale and minerals. And what happens to the energy if someone inserts a restrictor plate or orifice? The increasing energy drags the pump away from its duty coordinates.
A determined amount of energy is required to push this same liquid through a clean filter screen. The pump develops this energy. The energy is consumed as resistance through the filter. The energy increases as the filter clogs with debris. Then the energy relaxes again when a new filter screen is installed. The varying energy drags the pump away from its best efficiency zone on the curve.
When the pump is properly mated into its dynamic system, the injected energy balances the consumed energy in the system. The liquid reaches its destination, ready for the next step or phase of the process. If there is an energy imbalance, the excess energy (either added or consumed) expresses itself as errant vibrations, excess heat, component distortion, and/or noise.
This energy imbalance is the whole reason ongoing vibration trending and analysis exists. It is the reason some pumps make too much noise. It is the reason other pumps go into cavitation. It is the reason some cold water pumps overheat. Basically, energy imbalance results in runaway pump maintenance.
The words I’m using and the concepts I’m expressing are not a refresher. You didn’t study this in college. This is not taught in engineering school. The professor doesn’t express these thoughts in class. These words don’t appear in your "Fluid Mechanics" textbook except as the theoretical and illusive First Law of Thermodynamics: "Energy is neither created nor destroyed. Energy is only transformed from one form into another."
The Pump Guy is Larry Bachus, a pump consultant, lecturer, and inventor, based in Nashville, Tennessee. Bachus is a retired member of ASME, and lectures in both English and Spanish. You can contact him at email@example.com.
Flow Control will present the Pump Guy Seminar Jan. 12-14, 2016, in Houston, and June 7-9, 2016, in Indianapolis. Learn more at flowcontrolnetwork.com/pumpguy.