Chapter 03

TURBOMACHINERY
(Fluid Flow Devices)

Course: Fluid Mechanics-II (ME-235)

Instructor: Engr. Ali Abbas

Office No: 78 (1st Floor)

Discussion Hours: Thursday-Friday (0900 – 1200 hours)

Email: [email protected]
 Books

Fluid Mechanics; fundamentals and applications by Y. A. Cengel & J.M. Cimbala, 4th Edition
(Chapter 14)

Mechanical Engineering Department 2

 Introduction

Turbomachinery, in mechanical engineering, describes machines that transfer energy between a

rotor and a fluid, including both turbines and compressors. While a turbine transfers energy from a
fluid to a rotor, a compressor transfers energy from a rotor to a fluid. There are two broad categories
of turbomachinery, pumps and turbines.

Pumps: Any fluid machine that adds energy to a fluid. Also known as energy absorbing devices
since energy is supplied to them, and they transfer most of that energy to the fluid. The increase in
fluid energy is usually felt as an increase in the pressure of the fluid.

Turbines: are energy producing devices; they extract energy from the fluid and transfer most of that
energy to some form of mechanical energy output. The fluid at the outlet of a turbine suffers an
energy loss, typically in the form of a loss of pressure.

Mechanical Engineering Department 3

 Introduction

The purpose of a pump is to add energy to a fluid,

resulting in an increase in fluid pressure, not necessarily
an increase of fluid speed across the pump.

The purpose of a turbine is to extract energy from a fluid,

resulting in a decrease of fluid pressure, not necessarily
a decrease of fluid speed across the turbine.

Mechanical Engineering Department 4

 Introduction
Fluid machines that move liquids are called pumps, but there are several other names for machines
that move gases.

A fan is a gas pump with relatively low pressure rise and high flow rate. Examples include ceiling
fans, house fans, and propellers.

A blower is a gas pump with relatively moderate to high pressure rise and moderate to high flow
rate. Examples include centrifugal blowers and squirrel cage blowers in automobile ventilation
systems, furnaces, and leaf blowers.

A compressor is a gas pump designed to deliver a very high pressure rise, typically at low to
moderate flow rates. Examples include air compressors that run pneumatic tools and inflate tires at
automobile service stations, and refrigerant compressors used in heat pumps, refrigerators, and air
conditioners.

Mechanical Engineering Department 5

 Introduction
Pumps and turbines in which energy is supplied or extracted by a
rotating shaft are properly called turbomachines, since the Latin prefix
turbo means “spin.” Not all pumps or turbines utilize a rotating shaft,
however. The hand-operated air pump you use to inflate the tires of your
bicycle is

a prime example. The up and down reciprocating motion of a plunger or

piston replaces the rotating shaft in this type of pump, and it is more
proper to call it simply a fluid machine instead of a turbomachine. An Not all pumps have a
rotating shaft; (a) energy is
old-fashioned well pump operates in a similar manner to pump water
supplied to this manual tire
instead of air. Nevertheless, the words turbomachine and pump by the up and down
motion of a person’s arm to
turbomachinery are often used in the literature to refer to all types of
pump air; (b) a similar
pumps and turbines regardless of whether they utilize a rotating shaft or mechanism is used to pump
water with an old-fashioned
not.
well pump.
Mechanical Engineering Department 6
 Introduction
Fluid machines may also be broadly classified as either positive-displacement machines or dynamic
machines, based on the manner in which energy transfer occurs.

Positive-displacement Machines:

Fluid is directed into a closed volume. Energy transfer to the fluid is accomplished by movement of
the boundary of the closed volume, causing the volume to expand or contract, thereby sucking fluid
in or squeezing fluid out, respectively.

Examples - Reciprocating Engines, Reciprocating Pumps and Compressors

Dynamic Machines:

There is no closed volume; instead, rotating blades supply or extract energy to or from the fluid. For
pumps, these rotating blades are called impeller blades, while for turbines, the rotating blades are
called runner blades or buckets.

Examples - Axial or Radial Turbines, Centrifugal Pumps and Compressors

Mechanical Engineering Department 7

Positive-displacement Pumps: Introduction

Gear Pump Double Screw Pump Flexible-tube Peristaltic Pump Three-lobe Rotary Pump

Dynamic Pumps

Centrifugal Pump Mixed Flow Pump Axial Pump

Mechanical Engineering Department 8
 PUMPS

Some fundamental parameters are used to analyze the performance of a pump. The mass flow rate of
fluid through the pump is an obvious primary pump performance parameter. For incompressible flow,
it is more common to use volume flow rate rather than mass flow rate. In the turbomachinery
industry, volume flow rate is called capacity and is simply mass flow rate divided by fluid density;

Volume flow rate (capacity):

The performance of a pump is characterized additionally by its net head H, defined as the change in
Bernoulli head between the inlet and outlet of the pump

Mechanical Engineering Department 9

 PUMPS

The net head of a pump, H, is defined as the change in Bernoulli head

from inlet to outlet; for a liquid, this is equivalent to the change in the
energy grade line

Note:
EGL (Total Fluid Head) = Velocity Head + Pressure Head + Elevation head

Head across an operating

pump always increases

Mechanical Engineering Department 10

 PUMPS

Some Useful Parameters:

Brake horsepower: The external power supplied to the pump is called the brake horsepower.

𝒃𝒉𝒑 = 𝑾ሶ 𝒔𝒉𝒂𝒇𝒕 = 𝝎 ∗ 𝑻𝒔𝒉𝒂𝒇𝒕

Water horsepower: Useful power actually delivered to the fluid.

𝒘𝒉𝒑 = 𝑾ሶ 𝒘𝒂𝒕𝒆𝒓 𝒉𝒐𝒓𝒔𝒆𝒑𝒐𝒘𝒆𝒓 = 𝒎𝒈𝑯

ሶ ሶ
= 𝒑𝒈𝑽𝑯
Where:
𝝎 = 𝒓𝒐𝒕𝒂𝒕𝒊𝒐𝒏𝒂𝒍 𝒔𝒑𝒆𝒆𝒅 𝒐𝒇 𝒔𝒉𝒂𝒇𝒕 𝒊𝒏 𝒓𝒂𝒅/𝒔

𝑻𝒔𝒉𝒂𝒇𝒕 = 𝒕𝒐𝒓𝒒𝒖𝒆 𝒔𝒖𝒑𝒑𝒍𝒊𝒆𝒅 𝒕𝒐 𝒕𝒉𝒆 𝒕𝒐𝒓𝒒𝒖𝒆 𝒊𝒏 𝑵𝒎

Mechanical Engineering Department 11

 PUMPS

Pump Efficiency:

All pumps suffer from irreversible losses due to friction, internal leakage, flow separation on blade
surfaces, turbulent dissipation, etc. Therefore, the brake horsepower (bhp) supplied to the pump
must be larger than water horsepower (whp).

Keeping the above statement in mind; we define the pump efficiency to be the ratio of whp to bhp.

𝑾ሶ 𝒘𝒂𝒕𝒆𝒓 𝒉𝒐𝒓𝒔𝒆𝒑𝒐𝒘𝒆𝒓 𝒑𝒈𝑽𝑯 ሶ

𝜼𝒑𝒖𝒎𝒑 = =
ሶ
𝑾𝒔𝒉𝒂𝒇𝒕 𝝎 ∗ 𝑻𝒔𝒉𝒂𝒇𝒕

Increased losses will indicate decreased efficiency.

Mechanical Engineering Department 12

 Pump Performance Curves and Matching
a Pump to a Piping System

▪ The maximum volume flow rate through a

pump occurs when its net head is zero, H = 0;
this flow rate is called the pump’s free delivery.

▪ The free delivery condition is achieved when

there is no flow restriction at the pump inlet or
outlet—in other words when there is no load
on the pump.

▪ At this operating point, 𝑽ሶ is large, but H is

zero; the pump’s efficiency is zero because the
pump is doing no useful work.

Mechanical Engineering Department 13

 Pump Performance Curves and Matching
a Pump to a Piping System

▪ The At the other extreme, the shutoff head is

the net head that occurs when the volume flow
rate is zero, V = 0, and is achieved when the
outlet port of the pump is blocked off.

▪ Under these conditions, H is large but V is

zero; the pump’s efficiency is again zero,
because the pump is doing no useful work.

▪ Between these two extremes, from shutoff to

free delivery, the pump’s net head may
increase from its shutoff value somewhat as
the flow rate increases, but H must eventually
decrease to zero as the volume flow rate
increases to its free delivery value.
Mechanical Engineering Department 14
 Pump Performance Curves and Matching
a Pump to a Piping System

▪ The pump’s efficiency reaches its maximum value

somewhere between the shutoff condition and the
free delivery condition; this operating point of
maximum efficiency is appropriately called the
best efficiency point (BEP)

▪ Curves of H, 𝜼𝒑𝒖𝒎𝒑 , and bhp as functions of V are

called pump performance curves.

▪ The pump performance curves change with

rotational speed.

Mechanical Engineering Department 15

 Pump Performance Curves and Matching
a Pump to a Piping System

▪ It is important to realize that for steady

conditions, a pump can operate only along its
performance curve.

▪ Thus, the operating point of a piping system is

determined by matching system requirements
(required net head) to pump performance
(available net head).

▪ In a typical application, Hrequired and Havailable match

at one unique value of flow rate—this is the
operating point or duty point of the system.

Mechanical Engineering Department 16

 Pump Performance Curves and Matching
a Pump to a Piping System

▪ For a given piping system with its major and minor

losses, elevation changes, etc., the required net
head increases with volume flow rate.

▪ On the other hand, the available net head of most

pumps decreases with flow rate at least over the
majority of its recommended operating range.

▪ Hence, the system curve and the pump

performance curve intersect as sketched and this
establishes the operating point.

▪ If we are lucky, the operating point is at or near the

best efficiency point of the pump.

Mechanical Engineering Department 17

 Pump Performance Curves and Matching
a Pump to a Piping System
▪ In most cases, the pump does not run at its optimum efficiency. If
efficiency is of major concern, the pump should be carefully
selected (or a new pump should be designed) such that the
operating point is as close to the best efficiency point as
possible.

▪ In some cases it may be possible to change the shaft rotation

speed so that an existing pump can operate much closer to its
design point (best efficiency point).

There are unfortunate situations where the system curve and the pump performance curve
intersect at more than one operating point. This can occur when a pump that has a dip in its net
head performance curve is mated to a system that has a fairly flat system curve. Although rare,
such situations are possible and should be avoided, because the system may “hunt” for an
operating point, leading to an unsteady-flow situation.
Mechanical Engineering Department 18
 Pump Performance Curves and Matching
a Pump to a Piping System
▪ The equation for Hrequired is given is follows: 𝜶 = kinetic energy correction factor
𝑷𝟐 − 𝑷𝟏 𝜶𝟐 𝑽𝟐𝟐 − 𝜶𝟏 𝑽𝟐𝟏
𝑯𝒓𝒆𝒒𝒖𝒊𝒓𝒆𝒅 = + + 𝒛𝟐 − 𝒛𝟏 + 𝒉𝒍𝒐𝒔𝒔𝒆𝒔
𝝆𝒈 𝟐𝒈
▪ The useful pump head delivered to the fluid does 4 things;

▪ It increases the static pressure of the fluid from point 1 to point 2 (first term on the right).

▪ It increases the dynamic pressure (kinetic energy) of the fluid from point 1 to point 2 (second
term on the right).

▪ It raises the elevation (potential energy) of the fluid from point 1 to point 2 (third term on the
right).

▪ It overcomes irreversible head losses in the piping system (last term on the right).

We assume that there is no turbine in the system, although that term can be added back in, if necessary.
Mechanical Engineering Department 19
 Pump Performance Curves and Matching
a Pump to a Piping System

𝑷𝟐 − 𝑷𝟏 𝜶𝟐 𝑽𝟐𝟐 − 𝜶𝟏 𝑽𝟐𝟏
𝑯𝒓𝒆𝒒𝒖𝒊𝒓𝒆𝒅 = + + 𝒛𝟐 − 𝒛𝟏 + 𝒉𝒍𝒐𝒔𝒔𝒆𝒔
𝝆𝒈 𝟐𝒈
▪ In a general system, the change in static pressure, dynamic pressure, and elevation may be either
positive or negative, while irreversible head losses are always positive.

▪ In many mechanical and civil engineering problems in which the fluid is a liquid, the elevation
term is important but when the fluid is a gas, such as in ventilation and air pollution control
problems, the elevation term is almost always negligible.

▪ To match a pump to a system, and to determine the operating point, we equate Hrequired of to
Havailable, which is the (typically known) net head of the pump as a function of volume flow rate.

▪ The most common situation is that an engineer selects a pump that is somewhat heftier than
actually required. The volume flow rate through the piping system is then a bit larger than needed,
and a valve or damper is installed in the line so that the flow rate can be decreased as necessary.

Mechanical Engineering Department 20

 Pump Performance Curves and Matching
Example 3-1: a Pump to a Piping System

A local ventilation system (hood and exhaust duct) is used to remove air and contaminants produced
by a dry-cleaning operation. The duct is round and is constructed of galvanized steel with longitudinal
seams and with joints every 30 in (0.76 m). The inner diameter (ID) of the duct is D = 9.06 in (0.230 m),
and its total length is L = 44.0 ft (13.4 m). There are five CD3-9 elbows along the duct. The equivalent
roughness height of this duct is 0.15 mm, and each elbow has a minor (local) loss coefficient of KL = C0
= 0.21. Note the notation C0 for minor loss coefficient, commonly used in the ventilation industry
(ASHRAE, 2001). To ensure adequate ventilation, the minimum required volume flow rate through the
duct is 𝑽ሶ = 600 cfm (cubic feet per minute), or 0.283 m3 /s at 25°C. Literature from the hood
manufacturer lists the hood entry loss coefficient as 1.3 based on duct velocity. When the damper is
fully open, its loss coefficient is 1.8. A centrifugal fan with 9.0-in inlet and outlet diameters is available.
Its performance data are shown in Table on next slide, as listed by the manufacturer. Predict the
operating point of this local ventilation system, and draw a plot of required and available fan pressure
rise as functions of volume flow rate. Is the chosen fan adequate?
Mechanical Engineering Department 21
Pump Performance Curves and Matching
a Pump to a Piping System

Mechanical Engineering Department 22

 Pump Performance Curves and Matching
a Pump to a Piping System

▪ It is common practice in the pump industry to offer several

choices of impeller diameter for a single pump casing. There are
several reasons for this:

➢ to save manufacturing costs

➢ to enable capacity increase by simple impeller replacement

➢ to standardize installation mountings

➢ to enable reuse of equipment for a different application

▪ When plotting the performance of such a “family” of pumps,

pump manufacturers do not plot separate curves of H, 𝜂pump,
and bhp for each impeller diameter in the form sketched

Mechanical Engineering Department 23

 Pump Performance Curves and Matching
a Pump to a Piping System
Example 3-2:

A washing operation at a power plant requires 370 gallons per minute (gpm) of water. The required net
head is about 24 ft at this flow rate. A newly hired engineer looks through some catalogs and decides
to purchase the 8.25-in impeller option of the Taco Model 4013 FI Series centrifugal pump. If the pump
operates at 1160 rpm, as specified in the performance plot, she reasons, its performance curve
intersects 370 gpm at H = 24 ft. The chief engineer, who is very concerned about efficiency, glances at
the performance curves and notes that the efficiency of this pump at this operating point is only 70
percent. He sees that the 12.75-in impeller option achieves a higher efficiency (about 76.5 percent) at
the same flow rate. He notes that a throttle valve can be installed downstream of the pump to increase
the required net head so that the pump operates at this higher efficiency. He asks the junior engineer
to justify her choice of impeller diameter. Namely, he asks her to calculate which impeller option (8.25-
in or 12.75-in) would need the least amount of electricity to operate. Perform the comparison and
discuss.
Mechanical Engineering Department 24
Pump Performance Curves and Matching
a Pump to a Piping System

Mechanical Engineering Department 25

 Pump Cavitation and Net Positive Suction Head

▪ When pumping liquids, it is possible for the local pressure inside the
pump to fall below the vapor pressure of the liquid, Pv. (Pv is also
called the saturation pressure Psat)

▪ When P < Pv, vapor-filled bubbles called cavitation bubbles appear.

In other words, the liquid boils locally, typically on the suction side
of the rotating impeller blades where the pressure is lowest.

▪ After the cavitation bubbles are formed, they are transported

through the pump to regions where the pressure is higher, causing
rapid collapse of the bubbles. It is this collapse of the bubbles that is
undesirable, since it causes noise, vibration, reduced efficiency, and
most importantly, damage to the impeller blades.

▪ Repeated bubble collapse near a blade surface leads to pitting or

erosion of the blade and eventually catastrophic blade failure.
Mechanical Engineering Department 26
 Pump Cavitation and Net Positive Suction Head

▪ To avoid cavitation, we must ensure that the local pressure everywhere inside the pump stays
above the vapor pressure.

▪ Since pressure is most easily measured (or estimated) at the inlet of the pump, cavitation criteria
are typically specified at the pump inlet.

▪ It is useful to employ a flow parameter called net positive suction head (NPSH), defined as the
difference between the pump’s inlet stagnation pressure head and the vapor pressure head

𝑷 𝑽𝟐 𝑷𝑽
𝑵𝑷𝑺𝑯 = + −
𝝆𝒈 𝟐𝒈 𝒑𝒖𝒎𝒑 𝒊𝒏𝒍𝒆𝒕
𝝆𝒈

Mechanical Engineering Department 27

 Pump Cavitation and Net Positive Suction Head

▪ The pump manufacturer publishes a performance parameter called the required net positive suction
head (NPSHrequired), defined as the minimum NPSH necessary to avoid cavitation in the pump.

▪ The measured value of NPSHrequired varies with volume flow rate, and therefore NPSHrequired is often
plotted on the same pump performance curve as net head.

When expressed properly in units of head of the liquid being pumped,

NPSHrequired is independent of the type of liquid. However, if the required
net positive suction head is expressed for a particular liquid in pressure
units such as pascals or psi, the engineer must be careful to convert this
pressure to the equivalent column height of the actual liquid being
pumped.

▪ Note that since NPSHrequired is usually much smaller than H over the majority of the performance
curve, it is often plotted on a separate expanded vertical axis for clarity.

Mechanical Engineering Department 28

 Pump Cavitation and Net Positive Suction Head

▪ In order to ensure that a pump does not cavitate, the actual or available NPSH must be greater
than NPSHrequired.

Mechanical Engineering Department 29

 Pump Cavitation and Net Positive Suction
Example 3-3: Head
The 11.25-in impeller option of the Taco Model 4013 FI Series centrifugal
pump is used to pump water at 25°C from a reservoir whose surface is
4.0 ft above the centerline of the pump inlet. The piping system from the
reservoir to the pump consists of 10.5 ft of cast iron pipe with an internal
diameter of 4.0 in and an average inner roughness height of 0.02 in.
There are several minor losses: a sharp-edged inlet (KL = 0.5), three
flanged smooth 90° regular elbows (KL = 0.3 each), and a fully open
flanged globe valve (KL = 6.0). Calculate the available NPSH at a
volumetric flowrate of 400gpm.

Mechanical Engineering Department 30

 Positive Displacement Pumps
▪ People have designed numerous positive-displacement pumps throughout the centuries. In each
design, fluid is sucked into an expanding volume and then pushed along as that volume contracts,
but the mechanism that causes this change in volume differs greatly among the various designs.

▪ To illustrate the operation of a positive-displacement pump, we sketch four phases of half of a cycle
of a simple rotary pump with two lobes on each rotor

Mechanical Engineering Department 31

 Positive Displacement Pumps (Advantages)

▪ Positive-displacement pumps are a good choice for pumping highly viscous fluids and slurries. They
are used, for example, as automobile engine oil pumps and in the foods industry to pump heavy
liquids like syrup, tomato paste, and chocolate, and slurries like soups.

▪ Positive-displacement pumps have many advantages over dynamic pumps. For example, a positive-
displacement pump is better able to handle shear sensitive liquids since the induced shear is much
less than that of a dynamic pump operating at similar pressure and flow rate. Blood is a shear
sensitive liquid, and this is one reason why positive-displacement pumps are used for artificial
hearts.

▪ A well-sealed positive-displacement pump can create a significant vacuum pressure at its inlet, even
when dry, and is thus able to lift a liquid from several meters below the pump. We refer to this kind of
pump as a self-priming pump

▪ Finally, the rotor(s) of a positive-displacement pump run at lower speeds than the rotor (impeller) of a
dynamic pump at similar loads, extending the useful lifetime of seals
Mechanical Engineering Department 32
 Positive Displacement Pumps (Disadvantages)

▪ Their volume flow rate cannot be changed unless the rotation rate is changed. (This is not as simple
as it sounds, since most AC electric motors are designed to operate at one or more fixed rotational
speeds.)

▪ They create very high pressure at the outlet side, and if the outlet becomes blocked, ruptures may
occur or electric motors may overheat

▪ Because of their design, positive-displacement pumps sometimes deliver a pulsating flow, which
may be unacceptable for some applications.

Mechanical Engineering Department 33

 Positive Displacement Pumps

▪ Analysis of positive-displacement pumps is fairly straightforward. From the geometry of the pump,
we calculate the closed volume (𝑽𝒄𝒍𝒐𝒔𝒆𝒅 ) that is filled (and expelled) for every 𝑵 rotations of the shaft.
Volume flow rate is then equal to rotation rate 𝑵ሶ times 𝑽𝒄𝒍𝒐𝒔𝒆𝒅 divided by 𝑵.
𝑽𝒄𝒍𝒐𝒔𝒆𝒅
▪ Example 3-4: 𝑽ሶ = 𝑵ሶ
𝑵
A two-lobe rotary positive-displacement pump, moves 0.45 cm3 of SAE 30 motor oil in each lobe
volume Vlobe. Calculate the volume flow rate of oil for the case where 𝑵ሶ = 900 rpm.

Mechanical Engineering Department 34

 Dynamic Pumps

Centrifugal Pump Axial Pump Mixed Flow Pump

Fluid enters axially (in the same Fluid enters and leaves axially, A mixed-flow pump is
direction as the axis of the typically along the outer portion intermediate between centrifugal
rotating shaft) in the center of of the pump because of blockage and axial, with the flow entering
the pump, but is discharged by the shaft, motor, hub, etc. axially, not necessarily in the
radially (or tangentially) along center, but leaving at some angle
the outer radius of the pump between radially and axially.
casing. For this reason
centrifugal pumps are also called
radial-flow pumps.
Mechanical Engineering Department 35
 Centrifugal Pumps: Dynamic Pumps
A schematic diagram of a centrifugal pump is shown in Figure.
In pump terminology, the rotating assembly that consists of the shaft, the hub, the impeller
blades, and the impeller shroud is called the impeller or rotor.

Fluid enters axially through the hollow middle

portion of the pump (the eye), after which it
encounters the rotating blades. It acquires
tangential and radial velocity by momentum
transfer with the impeller blades, and acquires
additional radial velocity by so-called centrifugal
forces, which are actually a lack of sufficient
centripetal forces to sustain circular motion.
The flow leaves the impeller after gaining both
speed and pressure as it is flung radially
outward into the scroll. Mechanical Engineering Department 36
 Centrifugal Pumps: Dynamic Pumps

▪ The scroll is a snail-shaped diffuser whose purpose

is to decelerate the fast-moving fluid leaving the
trailing edges of the impeller blades, thereby further
increasing the fluid’s pressure, and to combine and
direct the flow from all the blade passages toward a
common outlet.
▪ If the flow is steady, if the fluid is incompressible,
and if the inlet and outlet diameters are the same, 𝑷𝟐 − 𝑷𝟏 𝜶𝟐 𝑽𝟐𝟐 − 𝜶𝟏 𝑽𝟐𝟏
𝑯𝒓𝒆𝒒𝒖𝒊𝒓𝒆𝒅 = + + 𝒛𝟐 − 𝒛𝟏 + 𝒉𝒍𝒐𝒔𝒔𝒆𝒔
𝝆𝒈 𝟐𝒈
the average flow speed at the outlet is identical to
that at the inlet. Thus, it is not necessarily the speed, 𝑷𝟐 − 𝑷𝟏
𝑯𝒓𝒆𝒒𝒖𝒊𝒓𝒆𝒅 = + 𝒛𝟐 − 𝒛𝟏 + 𝒉𝒍𝒐𝒔𝒔𝒆𝒔
but the pressure that increases from inlet to outlet 𝝆𝒈
through a centrifugal pump.

Mechanical Engineering Department 37

Centrifugal Pumps: Dynamic Pumps

Backward-inclined blades Radial blades Forward-inclined blades

▪ Highest efficiency due to ▪ Simplest geometry with ▪ Pressure rise remains nearly
minimal fluid turning in blade straight blades. constant over a wide range of
passages. ▪ Produces highest pressure flow rates.
▪ Can have airfoil-shaped rise initially but drops rapidly ▪ Generally lower maximum
blades for even higher after peak efficiency. efficiency compared to other
efficiency. types.
▪ Intermediate pressure rise ▪ Adaptive to varying
compared to other types. performance requirements.

Mechanical Engineering Department 38

Centrifugal Pumps: Dynamic Pumps
Assumptions for analysis;
▪ Flow is steady
▪ Flow is incompressible
▪ Flow is two dimensional (2D) i.e. Radial/Normal velocity
components (subscript n) and circumferential/tangential velocity
components (subscript t) are considered from blade in to blade
out. Axial velocity components are not considered.
▪ No leakage between blade and casing.
The volume flow rate 𝑽ሶ entering the eye of the pump passes through the 𝓥ሶ 𝟏 = 𝓥ሶ 𝟐
circumferential cross-sectional area defined by width 𝒃𝟏 at radius 𝒓𝟏 .
Conservation of mass requires that this same volume flow rate must pass
𝟐𝝅𝒓𝟏 𝒃𝟏 𝑽𝟏,𝒏 = 𝟐𝝅𝒓𝟐 𝒃𝟐 𝑽𝟐,𝒏
through the circumferential cross-sectional area defined by width 𝒃𝟐 at 𝒓𝟏 𝒃𝟏
radius 𝒓𝟐 . Using the average normal velocity components 𝑽𝟏,𝒏 and 𝑽𝟐,𝒏
𝑽𝟐,𝒏 = 𝑽𝟏,𝒏
𝒓𝟐 𝒃𝟐
defined in the given figure, we write;
Mechanical Engineering Department 39
Centrifugal Pumps: Dynamic Pumps
The tangential components of the flow through a pump are as followed:
NOTE: Both are same diagrams just from different books

Mechanical Engineering Department 40

Centrifugal Pumps: Dynamic Pumps
To evaluate the torque on the rotating shaft, we apply the angular
momentum relation for a control volume. We choose a control
volume surrounding the impeller blades, from radius r1 to radius r2.
Angles 𝛼1 and 𝛼2, defined as the angle of departure (flow angles) of
the absolute velocity vector from the normal direction at radii r1 and
r2, respectively.

We make the approximation that flow enters the control volume with
uniform absolute velocity 𝑽𝟏 around the entire circumference at

radius r1 and exits with uniform absolute velocity 𝑽𝟐 around the

entire circumference at radius r2.

Since angular momentum is defined as the cross product 𝒓 × 𝑽, only

the tangential components of 𝑽𝟏 and 𝑽𝟐 are relevant to the shaft

torque. Mechanical Engineering Department 41
Centrifugal Pumps: Dynamic Pumps
It turns out that shaft torque is equal to the change in angular momentum
from inlet to outlet; ሶ 𝟐 𝑽𝟐,𝒕 − 𝒓𝟏 𝑽𝟏,𝒕 )
𝑻𝒔𝒉𝒂𝒇𝒕 = 𝝆𝓥(𝒓
Or, in terms of angles 𝛼1 and 𝛼2 and the magnitudes of the absolute velocity
vectors; ሶ 𝟐 𝑽𝟐 𝒔𝒊𝒏𝜶𝟐 − 𝒓𝟏 𝑽𝟏 𝒔𝒊𝒏𝜶𝟏 )
𝑻𝒔𝒉𝒂𝒇𝒕 = 𝝆𝓥(𝒓
In our simplified analysis there are no irreversible losses in the pump,
hence; 𝜼𝒑𝒖𝒎𝒑 = 𝟏

𝑾ሶ 𝒘𝒂𝒕𝒆𝒓 𝒉𝒐𝒓𝒔𝒆𝒑𝒐𝒘𝒆𝒓 𝝆𝒈𝑽𝑯 ሶ

𝜼𝒑𝒖𝒎𝒑 = = → ሶ = 𝝎 ∗ 𝑻𝒔𝒉𝒂𝒇𝒕
𝝆𝒈𝑽𝑯
ሶ
𝑾𝒔𝒉𝒂𝒇𝒕 𝝎 ∗ 𝑻𝒔𝒉𝒂𝒇𝒕
→ ሶ = 𝝎 ∗ 𝝆𝓥(𝒓
𝝆𝒈𝓥𝑯 ሶ 𝟐 𝑽𝟐,𝒕 − 𝒓𝟏 𝑽𝟏,𝒕 )

𝟏
→ 𝒏𝒆𝒕 𝒉𝒆𝒂𝒅 = 𝑯 = ∗ (𝝎𝒓𝟐 𝑽𝟐,𝒕 − 𝝎𝒓𝟏 𝑽𝟏,𝒕 )
𝒈
Mechanical Engineering Department 42
 Dynamic Pumps
Example 3-5:

A centrifugal blower rotates at N = 1750 rpm (183.3 rad/s). Air enters

the impeller normal to the blades (𝛼1 = 0°) and exits at an angle of 40°
from radial (𝛼2 = 40°). The inlet radius is r1 = 4.0 cm, and the inlet
blade width 𝒃𝟏 = 5.2 cm. The outlet radius is r2 = 8.0 cm, and the
outlet blade width 𝒃𝟐 = 2.3 cm. The volume flow rate is 0.13 m3/s. For
the idealized case, i.e., 100 percent efficiency, calculate

1. The net head produced by this blower in equivalent millimeters of

water column height.

2. The required brake horsepower in watts.

Mechanical Engineering Department 43

Centrifugal Pumps: Dynamic Pumps
Impeller Blade Shape Design
In order to design the shape of the impeller blades, we must use trigonometry to obtain
expressions for 𝑽𝟏,𝒕 and 𝑽𝟐,𝒕 in terms of blade angles 𝛽1 and 𝛽2. Applying the law of cosines to the

triangle in Fig. formed by absolute velocity vector 𝑽𝟐 , relative velocity vector 𝑽𝟐,𝒓𝒆𝒍𝒂𝒕𝒊𝒗𝒆 , and the
tangential velocity of the blade at radius r2 (of magnitude 𝜔r2) we get;

𝑽𝟐𝟐 = 𝑽𝟐𝟐,𝒓𝒆𝒍𝒂𝒕𝒊𝒗𝒆 + 𝝎𝟐 𝒓𝟐𝟐 − 𝟐𝝎𝒓𝟐 𝑽𝟐,𝒓𝒆𝒍𝒂𝒕𝒊𝒗𝒆 𝒄𝒐𝒔𝜷𝟐

Mechanical Engineering Department 44

Centrifugal Pumps: Dynamic Pumps
Impeller Blade Shape Design
But we also see from Figure that 𝑽𝟐,𝒓𝒆𝒍𝒂𝒕𝒊𝒗𝒆 𝒄𝒐𝒔𝜷𝟐 = 𝝎𝒓𝟐 − 𝑽𝟐,𝒕 → 1

𝑽𝟐𝟐 = 𝑽𝟐𝟐,𝒓𝒆𝒍𝒂𝒕𝒊𝒗𝒆 + 𝝎𝟐 𝒓𝟐𝟐 − 𝟐𝝎𝒓𝟐 𝑽𝟐,𝒓𝒆𝒍𝒂𝒕𝒊𝒗𝒆 𝒄𝒐𝒔𝜷𝟐 → 2

Solving equation 1 and 2 we get
𝜷𝟐 At blade outlet
𝟏 𝟐
𝝎𝒓𝟐 𝝎𝒓𝟐 𝑽𝟐,𝒕 = 𝑽𝟐 − 𝑽𝟐𝟐,𝒓𝒆𝒍𝒂𝒕𝒊𝒗𝒆 + 𝝎𝟐 𝒓𝟐𝟐
𝑽𝟐,𝒕 𝑽𝟐,𝒓𝒆𝒍𝒂𝒕𝒊𝒗𝒆 𝒄𝒐𝒔𝜷𝟐 𝟐
Similarly at blade inlet
𝝎𝒓𝟐
𝟏 𝟐
𝝎𝒓𝟏 𝑽𝟏,𝒕 = 𝑽𝟏 − 𝑽𝟐𝟏,𝒓𝒆𝒍𝒂𝒕𝒊𝒗𝒆 + 𝝎𝟐 𝒓𝟐𝟏
𝟐

𝟏
𝑯= ∗ 𝑽𝟐𝟐 − 𝑽𝟐𝟏 + 𝝎𝟐 𝒓𝟐𝟐 − 𝝎𝟐 𝒓𝟐𝟏 − 𝑽𝟐𝟐,𝒓𝒆𝒍𝒂𝒕𝒊𝒗𝒆 − 𝑽𝟐𝟏,𝒓𝒆𝒍𝒂𝒕𝒊𝒗𝒆
𝟐𝒈

Mechanical Engineering Department 45

Centrifugal Pumps: Dynamic Pumps
Impeller Blade Shape Design
𝟏
𝑯= ∗ 𝑽𝟐𝟐 − 𝑽𝟐𝟏 + 𝝎𝟐 𝒓𝟐𝟐 − 𝝎𝟐 𝒓𝟐𝟏 − 𝑽𝟐𝟐,𝒓𝒆𝒍𝒂𝒕𝒊𝒗𝒆 − 𝑽𝟐𝟏,𝒓𝒆𝒍𝒂𝒕𝒊𝒗𝒆 → 3
𝟐𝒈

In words, Eq. 3 states that in the ideal case (no irreversible losses), the net head is proportional to
the change in absolute kinetic energy plus the rotor-tip kinetic energy change minus the change in
relative kinetic energy from inlet to outlet of the impeller. 𝑷 𝑽𝟐 𝑷 𝑽𝟐
𝑯= + +𝒛 − + +𝒛 → 4
Equating Eq. 3 and 4 we get; 𝝆𝒈 𝟐𝒈 𝟐
𝝆𝒈 𝟐𝒈 𝟏

𝑷 𝑽𝟐𝒓𝒆𝒍𝒂𝒕𝒊𝒗𝒆 𝝎𝟐 𝒓𝟐 𝑷 𝑽𝟐𝒓𝒆𝒍𝒂𝒕𝒊𝒗𝒆 𝝎𝟐 𝒓𝟐 𝑷 𝑽𝟐𝒓𝒆𝒍𝒂𝒕𝒊𝒗𝒆 𝝎𝟐 𝒓𝟐

+ − +𝒛 = + − +𝒛 → 5 + − + 𝒛 = 𝒄𝒐𝒏𝒔𝒕𝒂𝒏𝒕
𝝆𝒈 𝟐𝒈 𝟐𝒈 𝟐
𝝆𝒈 𝟐𝒈 𝟐𝒈 𝟏
𝝆𝒈 𝟐𝒈 𝟐𝒈

Eq. 5 is commonly called the Bernoulli equation in a rotating reference frame and is valid only for
the ideal case in which there are no irreversible losses through the impeller. Nevertheless, it is
valuable as a first-order approximation for flow through the impeller of a centrifugal pump.

Mechanical Engineering Department 46

Centrifugal Pumps: Dynamic Pumps
𝟏
Impeller Blade Shape Design 𝑯 = ∗ (𝝎𝒓𝟐 𝑽𝟐,𝒕 − 𝝎𝒓𝟏 𝑽𝟏,𝒕 )
𝒈
We now examine the equation for net head, more closely. Since the term containing 𝑽𝟏,𝒕 carries a
negative sign, we obtain the maximum H by setting 𝑽𝟏,𝒕 to zero. (We are assuming that there is no
mechanism in the eye of the pump that can generate a negative value of 𝑽𝟏,𝒕 .) Thus, a first-order
approximation for the design condition of the pump is to set 𝑽𝟏,𝒕 = 0. In other words, we select the
blade inlet angle 𝛽1 such that the flow into the impeller blade is purely radial from an absolute
reference frame, and 𝑽𝟏,𝒏 = 𝑽𝟏 . Using some trigonometry we see that;
𝑽𝟏,𝒏 𝑽𝟏,𝒏
𝑽𝟏,𝒕 = 𝝎𝒓𝟏 − → 𝝎𝒓𝟏 = → 𝑽𝟏,𝒏 = 𝝎𝒓𝟏 𝒕𝒂𝒏𝜷𝟏
𝒕𝒂𝒏𝜷𝟏 𝒕𝒂𝒏𝜷𝟏

𝓥ሶ = 𝟐𝝅𝒃𝟏 𝝎𝒓𝟐𝟏 𝒕𝒂𝒏𝜷𝟏 → 6

Eq. 6 can be used for preliminary design of the impeller blade shape

Mechanical Engineering Department 47

 Dynamic Pumps
Example 3-6:

A centrifugal pump is being designed to pump liquid refrigerant R-

134a ρ= 1226kg/m3 (at room temperature and atmospheric pressure).
The impeller inlet and outlet radii are r1=100mm and r2=180mm,
respectively. The impeller inlet and outlet widths are b1 = 50 and b2 =
30 mm (into the page of Fig). The pump is to deliver 0.25 m3/s of the
liquid at a net head of 14.5 m when the impeller rotates at 1720 rpm.
Design the blade shape for the case in which these operating
conditions are the design conditions of the pump ( 𝑽𝟏,𝒕 = 0, as
sketched in the figure); specifically, calculate

1. Angles 𝛽1 and 𝛽2

2. Also predict the horsepower required by the pump.

Mechanical Engineering Department 48

 Dynamic Pumps
Pump Scaling Laws:

Turbomachinery provides a very practical example of the power and

usefulness of dimensional analysis. We apply the method of repeating
variables to the relationship between;
▪ Gravity times net head (gH) [treated as one variable]
▪ Pump Properties
▪ ሶ
Volume flow rate (𝓥)
▪ Some characteristic length, typically the diameter of the impeller
blades (D)
▪ Blade surface roughness height (𝜀)
▪ Impeller rotational speed (𝜔)
▪ Fluid Properties
▪ Density (𝜌)
▪ Viscosity (𝜇)
Mechanical Engineering Department 49
 Dynamic Pumps
Pump Scaling Laws:
The dimensionless Pi groups obtained by the dimensional analysis
are;
𝒈𝑯 𝓥ሶ 𝝆𝝎𝑫𝟐 𝜺
𝜫𝟏 = 𝟐 𝟐 𝜫𝟐 = 𝜫𝟑 = 𝜫𝟒 =
𝝎 𝑫 𝝎𝑫𝟑 𝝁 𝑫
Following relationships are established between these dimensionless
parameters;

𝒈𝑯 𝓥ሶ 𝝆𝝎𝑫𝟐 𝜺 𝒃𝒉𝒑 𝓥ሶ 𝝆𝝎𝑫𝟐 𝜺

𝟐 𝟐
=𝒇 𝟐 𝟐
, , 𝟑 𝟓
=𝒇 𝟐 𝟐
, ,
𝝎 𝑫 𝝎 𝑫 𝝁 𝑫 𝝆𝝎 𝑫 𝝎 𝑫 𝝁 𝑫

𝝆𝝎𝑫𝟐
𝑹𝒆 = → As 𝝎𝑫 is a characteristics velocity
𝝁
𝜺
→ Dimensionless roughness parameter [you are already familiar with this term]
𝑫
Mechanical Engineering Department 50
 Dynamic Pumps
Pump Scaling Laws:

Dimensionless pump parameters:

𝒈𝑯
▪ 𝑪𝑯 = 𝑯𝒆𝒂𝒅 𝑪𝒐𝒆𝒇𝒇𝒊𝒄𝒊𝒆𝒏𝒕 = 𝝎𝟐 𝑫𝟐

𝓥ሶ
▪ 𝑪𝑸 = 𝑪𝒂𝒑𝒂𝒄𝒊𝒕𝒚 𝑪𝒐𝒆𝒇𝒇𝒊𝒄𝒊𝒆𝒏𝒕 = 𝝎𝑫𝟑

𝒃𝒉𝒑
▪ 𝑪𝒑 = 𝑷𝒐𝒘𝒆𝒓 𝑪𝒐𝒆𝒇𝒇𝒊𝒄𝒊𝒆𝒏𝒕 =
𝝆𝝎𝟑 𝑫𝟓

𝒈𝑵𝑷𝑺𝑯𝒓𝒆𝒒𝒖𝒊𝒓𝒆𝒅
▪ 𝑪𝑵𝑷𝑺𝑯 = 𝑺𝒖𝒄𝒕𝒊𝒐𝒏 𝑯𝒆𝒂𝒅 𝑪𝒐𝒆𝒇𝒇𝒊𝒄𝒊𝒆𝒏𝒕 = 𝝎𝟐 𝑫𝟐

𝒈𝑯 𝓥ሶ 𝝆𝝎𝑫𝟐 𝜺
=𝒇 , , → 7 Lets understand the interpretation of Eq. 7 and 8 for the sake
𝝎𝟐 𝑫 𝟐 𝝎𝟐 𝑫𝟐 𝝁 𝑫
corelating to similar pumps, in the next slide. Note that these
𝒃𝒉𝒑 𝓥ሶ 𝝆𝝎𝑫𝟐 𝜺 equations contain all the Pi groups so we are actually
=𝒇 , , → 8
𝝆𝝎𝟑 𝑫𝟓 𝝎𝟐 𝑫𝟐 𝝁 𝑫 discussing all those groups.

Mechanical Engineering Department 51

 Dynamic Pumps
Pump Scaling Laws:

If two pumps, A and B, are geometrically similar (pump A is

geometrically proportional to pump B, although they may be of
different sizes), and if the independent П’s are equal to each other (in
this case if (𝑪𝑸, 𝑨 = 𝑪𝑸, 𝑩 , 𝑹𝒆, 𝑨 = 𝑹𝒆, 𝑩 , and 𝜺𝑨 /𝑫𝑨 = 𝜺𝑩 /𝑫𝑩 ), then the
dependent П’s are guaranteed to also be equal to each other as well.
If such conditions are established, the two pumps are said to be
dynamically similar. When dynamic similarity is achieved, the
operating point on the pump performance curve of pump A and the
corresponding operating point on the pump performance curve of
pump B are said to be homologous.

Mechanical Engineering Department 52

 Dynamic Pumps
Pump Scaling Laws:

The requirement of equality of all three of the independent dimensionless parameters can be relaxed
somewhat;
▪ If the Reynolds numbers of both pump A and pump B exceed several thousand, turbulent flow
conditions exist inside the pump. It turns out that for turbulent flow, if the values of 𝑹𝒆, 𝑨 and 𝑹𝒆, 𝑩
are not equal, but not too far apart, dynamic similarity between the two pumps is still a reasonable
approximation. This fortunate condition is due to Reynolds number independence.
▪ In most cases of practical turbomachinery engineering analysis, the effect of differences in the
roughness parameter is also small, unless the roughness differences are large, as when one is
scaling from a very small pump to a very large pump (or vice versa).
Thus, for many practical problems, we may neglect the effect of both Re and 𝜀/D.
Eq. 7 and 8 then reduce to;
𝒈𝑯 𝓥ሶ 𝒃𝒉𝒑 𝓥ሶ
=𝒇 → 𝑪𝑯 ≈ 𝒇(𝑪𝑸 ) =𝒇 → 𝑪𝑷 ≈ 𝒇(𝑪𝑸 ) → 9
𝝎𝟐 𝑫𝟐 𝝎𝑫𝟑 𝝆𝝎𝟑 𝑫𝟓 𝝎𝑫𝟑
Mechanical Engineering Department 53
Pump Scaling Laws: Dynamic Pumps
As always, dimensional analysis cannot predict the shape of the
functional relationships of Eq. 9, but once these relationships are
obtained for a particular pump, they can be generalized for geometrically
similar pumps that are of different diameters, operate at different
rotational speeds and flow rates, and operate even with fluids of different
density and viscosity.
We can also write the pump efficiency equation in terms of these
dimensionless parameters;

ሶ
𝝆𝒈𝓥𝑯 𝝆 𝓥ሶ (𝒈𝑯) 𝝆 𝑪𝑸 𝝎𝑫𝟑 𝑪𝑯 𝝎𝟐 𝑫𝟐 𝑪𝑸 𝑪𝑯
𝜼𝒑𝒖𝒎𝒑 = = = → 𝜼𝒑𝒖𝒎𝒑 = ≈ 𝒇(𝑪𝑸 ) → 10
𝒃𝒉𝒑 𝒃𝒉𝒑 𝑪𝒑 𝝆𝝎𝟑 𝑫𝟓 𝑪𝒑

Since CH, CP, and 𝜂pump are approximated as functions only of CQ, we often plot these three parameters
as functions of CQ on the same plot, generating a set of nondimensional pump performance curves.

Mechanical Engineering Department 54

 Dynamic Pumps
Pump Scaling Laws:

The simplified similarity laws of Eq. 9 and 10 break down when the full-scale prototype is significantly
larger than its model; the prototype’s performance is generally better. There are several reasons for this:
▪ The prototype pump often operates at high Reynolds numbers that are not achievable in the
laboratory. We know from the Moody chart that the friction factor decreases with Re, as does
boundary layer thickness. Hence, the influence of viscous boundary layers is less significant as
pump size increases, since the boundary layers occupy a less significant percentage of the flow path
through the impeller.
▪ In addition, the relative roughness (𝜀/D) on the surfaces of the prototype impeller blades may be
significantly smaller than that on the model pump blades unless the model surfaces are micro
polished.
▪ Large full-scale pumps have smaller tip clearances relative to the blade diameter; therefore, tip losses
and leakage are less significant.

Mechanical Engineering Department 55

 Dynamic Pumps
Pump Scaling Laws:

Some empirical equations have been developed to account for the increase in efficiency between a
small model and a full-scale prototype. One such equation was suggested by Moody (1926) for turbines,
but it can be used as a first-order correction for pumps as well,
Moody efficiency correction equation for pumps:

𝟏/𝟓
𝑫𝒎𝒐𝒅𝒆𝒍
𝜼𝒑𝒖𝒎𝒑, 𝒑𝒓𝒐𝒕𝒐𝒕𝒚𝒑𝒆 ≅ 𝟏 − 𝟏 − 𝜼𝒑𝒖𝒎𝒑, 𝒎𝒐𝒅𝒆𝒍
𝑫𝒑𝒓𝒐𝒕𝒐𝒕𝒚𝒑𝒆

Note:
𝑫𝒎𝒐𝒅𝒆𝒍
Higher 𝜼𝒑𝒖𝒎𝒑, 𝒎𝒐𝒅𝒆𝒍 and lower 𝑫 will ultimately result
𝒑𝒓𝒐𝒕𝒐𝒕𝒚𝒑𝒆
in higher 𝜼𝒑𝒖𝒎𝒑, 𝒑𝒓𝒐𝒕𝒐𝒕𝒚𝒑𝒆

Mechanical Engineering Department 56

 Dynamic Pumps
Pump Scaling Laws:

Pump Specific Speed:

Pump specific speed is used to characterize the operation of a pump at its optimum conditions (best
efficiency point) and is useful for preliminary pump selection and/or design.
𝟏
𝓥ሶ 𝟐
𝟏
𝟏/𝟐
𝑪𝑸 𝝎𝑫𝟑 𝝎𝓥ሶ 𝟐
𝑵𝑺𝒑 = 𝟑/𝟒
= 𝟑 = 𝟑
𝑪𝑯 𝒈𝑯 𝟒 𝒈𝑯 𝟒
𝝎𝟐 𝑫𝟐

Note:
Remember than pump specific speed
is a dimensionless parameter

Mechanical Engineering Department 57

 Dynamic Pumps
Pump Scaling Laws:

Affinity Laws / Similarity Rules:

It is convenient to summarize the similarity relationships as ratios. Some authors call these
relationships similarity rules, while others call them affinity laws.
For any two homologous states A and B
𝟑
𝓥ሶ 𝑩 𝝎𝑩 𝑫𝑩
= 11 ▪ Equations 11 apply to both pumps and turbines.
ሶ𝓥𝑨 𝝎𝑨 𝑫𝑨
▪ States A and B can be any two homologous states between any two
𝟐 𝟐
𝑯𝑩 𝝎𝑩 𝑫𝑩 geometrically similar turbomachines, or even between two
=
𝑯𝑨 𝝎𝑨 𝑫𝑨 homologous states of the same machine.
𝟑 𝟓 ▪ Examples include changing rotational speed or pumping a different
𝒃𝒉𝒑𝑩 𝝆𝑩 𝝎𝑩 𝑫𝑩
= fluid with the same pump.
𝒃𝒉𝒑𝑨 𝝆𝑨 𝝎𝑨 𝑫𝑨

Mechanical Engineering Department 58

 Dynamic Pumps
Pump Scaling Laws:

Affinity Laws / Similarity Rules:

For the simple case of a given pump in which 𝜔 is varied, but the same fluid is pumped, DA = DB, and 𝜌A
= 𝜌B. In such a case, Equations 11 reduce to the forms

The pump affinity laws are quite useful as a design tool. In particular,
𝓥ሶ 𝑩 𝝎𝑩 𝒏𝑩 suppose the performance curves of an existing pump are known, and
= = 12
ሶ𝓥𝑨 𝝎𝑨 𝒏𝑨 the pump operates with reasonable efficiency and reliability. The
𝟐 𝟐 pump manufacturer decides to design a new, larger pump for other
𝑯𝑩 𝝎𝑩 𝒏𝑩
= = applications, e.g., to pump a much heavier fluid or to deliver a
𝑯𝑨 𝝎𝑨 𝒏𝑨
substantially greater net head. Rather than starting from scratch,
𝟑 𝟑
𝒃𝒉𝒑𝑩 𝝎𝑩 𝒏𝑩 engineers often simply scale up an existing design. The pump affinity
= =
𝒃𝒉𝒑𝑨 𝝎𝑨 𝒏𝑨 laws enable such scaling to be accomplished with a minimal amount
of effort.

Mechanical Engineering Department 59

 Dynamic Pumps
Example 3-7:

After graduation, you work for a pump manufacturing company. One of your
company’s best-selling products is a water pump, which we shall call pump A.
Its impeller diameter is DA = 6.0 cm, and its performance data when operating
at nA = 1725 rpm are shown in Table (next slide). The marketing research
department is recommending that the company design a new product, namely,
a larger pump (which we shall call pump B) that will be used to pump liquid
refrigerant R-134a at room temperature. The pump is to be designed such that
its best efficiency point occurs as close as possible to a volume flow rate of VB
= 2400 cm3/s and at a net head of HB = 450 cm (of R-134a). The chief engineer
(your boss) tells you to perform some preliminary analyses using pump
scaling laws to determine if a geometrically scaled-up pump could be designed
and built to meet the given requirements. Calculate the required pump
diameter DB, rotational speed nB, and brake horsepower bhpB for the
new product.

Mechanical Engineering Department 60

 Practice Problems

Fluid Mechanics; fundamentals and applications by Y. A. Cengel & J.M. Cimbala, 4th
Edition (Chapter 14)

Mechanical Engineering Department 61