Due to the myriad engineering, logistical and safety
challenges involved in a refining operation, American
Petroleum Institute (API) over hung (OH) style pumps are often
overlooked as a potential source of improved
productivityor as a cause of catastrophic failure, if not
Refinery managers, maintenance engineers and production
supervisors should adhere to best practices and understand the
operational dos and donts for employing pumps
properly in refining applications. Paying heed
to pump performance can boost operational efficiency, and avoid
failures that reduce production or cause catastrophic
Pump fundamentals and monitoring.
One example of a catastrophe occurred recently at a North
American refinery that produces about 70,000
barrels per day (bpd). A fire broke out at the bottom of a
vacuum tower, forcing a three-day shut down that cost $1.5
million in damages and lost production time. An investigation
quickly identified a failed pump as the cause. This is a common
occurrence, if pumps are not operated and maintained properly.
On average, one out of every 1,000 pumps with a
failedmechanical seal leads to a fire. Due to the seriousness
and cost of such failures, maintenance engineers and production
supervisors need to understand the operation of a centrifugal
pump and how to operate it efficiently.
A centrifugal pump is rotating machine comprised of six main
parts that work together to keep the pump operating properly
(Fig. 1). They include an impeller, pump casing, bearings,
bearing frame, shaft and a mechanical seal (Fig. 2). The
operation principle of the pump is to convert mechanical energy
to pressure. In operation, a rotating impeller accelerates a
liquid and as the area of the pump casing expands, the velocity
of the fluid is converted to pressure. As a result, pressurized
fluid exits the pump discharge. Though the performance of API
pumps has improved with enhancements to design and materials,
the basic structure has changed little in decades.
1. Centrifugal pump.
2. The six main parts of a
BEP and pump performance.
Best efficiency point (BEP), the flowrate where a pump has
its highest efficiency, is a key factor in pump performance.
Few pumps operate at their exact BEP all of the time, because
process variables in a production environment are not 100% constant.
However, a pump that is properly sized for its application will
maintain a flow near peak efficiency. Maintaining a flow
between 80%110% of BEP (Figs. 3 and 4) is a good range to
maximize efficiency and minimize the risk of excessive wear or
3. Centrifugal pump characteristics
running below recommended minimum flow.
4. Centrifugal pump
When a pump operates too far off its BEP, forces inside the
pump become imbalanced, which can cause parts to deflect and
Operating to the right of BEP, a condition known as
runout, means that the flowrate is higher than the
pump was designed to maintain. The high flow increases the exit
velocity of fluid leaving the pump, thus creating a
low-pressure area inside the pump. Operating to the left of BEP
occurs when the discharge flow is restricted, causing fluid to
recirculate within the pump, also creating a low-pressure area,
which can lead to increased radical loading and low-flow
In either case, the creation of imbalanced pressure
increases the radial loads on the impeller, which can cause
shaft deflectionthe bending of the impeller shaft, which
increases vibration of the pump. The vibration and imbalance
forces can create stress on the pumps internal
components, most likely to be seen first in the bearings and/or
mechanical seals, the two parts of a centrifugal pump that fail
5. Centrifugal pump characteristics
running at preferred operating range.
Cavitation and net positive suction head.
Cavitation is another serious problem caused by the lack of
adequate net positive suction head (NPSH), the pressure
provided at the suction of the pump, less the fluids vapor
pressure. When fluid pressure on the trailing side of the
impeller blade (opposite the pump intake) falls below the
vaporization point of the fluid, vapor bubbles begin to form.
When these vapor bubbles reach an area of high pressure inside
the pump, they can collapse violentlycausing sudden,
uneven axial and radial loading on the impeller. This, in turn,
can cause shaft deflection that is random in direction and
often severe in magnitude.
While it is easy to reccomend run your pumps at the BEP with
adequate NPSH, operation within the oil and gas markets, be it
upstream to downstream, is a dynamic process. Process change
production rate change, but typically the hundreds, if not
thousands, of pumps to support these process do not change and
this can lead to improper operations. It is only when a robust
condition monitoring program in combination with operations
support, can reliable and efficient pump operation be
Managing pump performance in a refinery.
In a complex refining operation, where the focus is on
production, manually monitoring the performance and operation
of API pumps is seldom the top priority. A disconnect also
often exists between the maintenance and operations
functions, which prevents managers from seeing the complete
picture of equipment performance. A continous condition
monitoring program that involves both maintenance and
operations can help refining operations to address both
challenges. The North American refinery faced with the $1.5
million pump failure is a case in point. A root-cause analysis
revealed that the pumps mechanical seal caused the
fireand a review of maintenance records showed numerous
repairs and parts replacements consistant with off-BEP pump
operations in the weeks and months leading up to the fire.
Subsequent to the disaster, the refinery installed a continuous
monitoring system in 2009.
The system plays a key role in keeping both the operational
and maintenance sides of the refinery up to speed on pump
performance and machine health. It monitors temperatures,
vibrations, pressure and power to ensure that pumps are
operating properly, and also warns of trouble before it
The monitoring data is distributed to both the control room
and a to web-based condition monitoring platform. Simple, easy
to understand key performance indicators (KPIs) such as suction
pressure, vibration and power data allow any operator to
monitor pump operations without special training. When more
advanced diagnostics are neccessary, such as vibration
spectrums and time-wave forms, the reliability and maintence teams can
access the web-based condition monitoring system from any
web-enabled device. The system integrates maintenance and
operational data in a single dashboard, improving operational
efficiency and reducing the need for walk-around
manual monitoring. In nearly two years since the system was
installed, the refinery has required no unplanned maintenance
on its pumps.
While refinery managers, maintenance
engineers and production supervisors face continual challenges
to maintaining production, they cant lose sight of proper
pump operation and maintenance. Both play a key role in
preventing pump failures. Condition monitoring systems need to
support both machine status and process data to answer why
pumps are about to fail, not just when and where. By
implementing these best practices, operators can run their
pumps like the pros, while boosting production and reducing
risk at the same time. HP
is the manager at ITT Monitoring and Control. He earned a
BS degree in mechanical engineering from the University
of Rochester and has spent the last decade integrating
technologies such as power electronics, embedded sensors
and wireless systems with rotating equipment to improve
equipment reliability and reduce energy.
Mr. Kernan and his team have helped numerous customers in
the oil and gas industry implement pumping systems and
pump control solutions. He holds three patents or patents
pending in technology related to
improving pump reliability.