GRUNDFOS PUMP User Manual

PUMP HANDBOOK

GRUNDFOS Management A/S

Poul Due Jensens Vej 7
DK-8850 Bjerringbro
Tel: [phone]

www.grundfos.com

Being responsible is our foundation

Thinking ahead makes it possible

Innovation is the essence

96563258 1104

PUMP HANDBOOK

Copyright 2004 GRUNDFOS Management A/S. All rights reserved.

Copyright law and international treaties protect this material. No part of this material
may be reproduced in any form or by any means without prior written permission from
GRUNDFOS Management A/S.

Disclaimer
All reasonable care has been taken to ensure the accuracy of the contents of this material,
however GRUNDFOS Management A/S shall not be reliable for any loss whether direct,
indirect, incidental or consequential arising out of the use of or reliance upon any of the
content of the material.

Foreword

The manufacturing industry places heavy demand on pumps, when it comes to
optimum operation, high reliability and low energy consumption. Therefore,
Grundfos has developed the Pump handbook, which in a simple manner deals
with various considerations when dimensioning pumps and pump systems.
We have elaborated a handbook for engineers and technicians who work with
design and installation of pumps and pump systems, containing answers to a
wide range of technical pump specific questions. The Pump handbook can either
be read from one end to the other or partly on specific topics.

The handbook is divided into 5 chapters which deal with different phases when
designing pump systems.

Throughout chapter 1 we make a general presentation of different pump types
and components. Here we also describe which precautions to adopt when dealing
with viscous liquids. Further, the most used materials as well as different types of
corrosion are presented here. The most important terminologies in connection
with reading the pump performance are presented in chapter 2. Chapter 3 deals
with system hydraulics and some of the most important factors to consider to obtain
optimum operation of the pump system. As it is often necessary to adjust the pump
performance by means of different adjustment methods, these are dealt with in
chapter 4. Chapter 5 describes the life cycle costs as energy consumption plays an
important role in today’s pumps and pump systems.

We sincerely hope that you will make use of The pump handbook and find it useful in
your daily work.

Segment Director Business Development Manager

Allan Skovgaard Claus Bærnholdt Nielsen

Chapter 1 Design of pumps and motors ......................7

Section 1.1 Pump construction ............................................................ 8

1.1.1 The centrifugal pump .................................................................. 8

1.1.2 Pump curves ................................................................................................9

1.1.3 Characteristics of the centrifugal pump ......................... 11

1.1.4 Most common end-suction and

in-line pump types .............................................................................. 12

1.1.5 Impeller types (axial forces) .......................................................14

1.1.6 Casing types (radial forces) ......................................................... 15

1.1.7 Single-stage pumps ........................................................................... 15

1.1.8 Multistage pumps ...............................................................................16

1.1.9 Long-coupled and close-coupled pumps ........................16

Section 1.2 Types of pumps ...................................................................17

1.2.1 Standard pumps .................................................................................... 17

1.2.2 Split-case pumps .................................................................................. 17

1.2.3 Hermetically sealed pumps ........................................................18

1.2.4 Sanitary pumps ..................................................................................... 20

1.2.5 Wastewater pumps ............................................................................21

1.2.6 Immersible pumps .............................................................................. 22

1.2.7 Borehole pumps ....................................................................................23

1.2.8 Positive displacement pumps ...................................................24

Section 1.3 Mechanical shaft seals ..................................................27

1.3.1 The mechanical shaft seal’s

components and function ............................................................29

1.3.2 Balanced and unbalanced shaft seals .............................. 30

1.3.3 Types of mechanical shaft seals .............................................31

1.3.4 Seal face material combinations ...........................................34

1.3.5 Factors affecting the seal performance ...........................36

Section 1.4 Motors .................................................................................... 39

1.4.1 Standards .................................................................................................... 40

1.4.2 Motor start-up .......................................................................................46

1.4.3 Voltage supply ........................................................................................47

1.4.4 Frequency converter ..........................................................................47

1.4.5 Motor protection ................................................................................. 49

Section 1.5 Liquids .......................................................................................53

1.5.1 Viscous liquids ........................................................................................54

1.5.2 Non-Newtonian liquids .................................................................. 55

1.5.3 The impact of viscous liquids on the

performance of a centrifugal pump .................................... 55

1.5.4 Selecting the right pump for a liquid

with antifreeze .......................................................................................56

1.5.5 Calculation example ..........................................................................58

1.5.6 Computer aided pump selection for dense and

viscous liquids .........................................................................................58

Section 1.6 Materials ................................................................................ 59

1.6.1 What is corrosion? .............................................................................. 60

1.6.2 Types of corrosion ................................................................................61

1.6.3 Metal and metal alloys ...................................................................65

1.6.4 Ceramics ....................................................................................................... 71

1.6.5 Plastics ............................................................................................................ 71

1.6.6 Rubber ............................................................................................................72

1.6.7 Coatings ........................................................................................................73

Chapter 2 Installation and performance

reading .............................................................................................................75

Section 2.1 Pump installation ............................................................76

2.1.1 New installation ....................................................................................76

2.1.2 Existing installation-replacement ........................................76

2.1.3 Pipe flow for single-pump installation ............................. 77

2.1.4 Limitation of noise and vibrations .......................................78

2.1.5 Sound level (L) .........................................................................................81

Section 2.2 Pump performance ........................................................ 83

2.2.1 Hydraulic terms .....................................................................................83

2.2.2 Electrical terms ...................................................................................... 90

2.2.3 Liquid properties ...................................................................................93

Table of Contents

Chapter 3 System hydraulic ....................................................95

Section 3.1 System characteristics .................................................96

3.1.1 Single resistances .................................................................................97

3.1.2 Closed and open systems ............................................................ 98

Section 3.2 Pumps connected in series and parallel

...................101

3.2.1 Pumps in parallel ................................................................................101

3.2.2 Pumps connected in series ....................................................... 103

Chapter 4 Performance adjustment

of pumps ..................................................................................................... 105

Section 4.1 Adjusting pump performance ..............................106

4.1.1 Throttle control ....................................................................................107

4.1.2 Bypass control .......................................................................................107

4.1.3 Modifying impeller diameter ................................................. 108

4.1.4 Speed control ........................................................................................ 108

4.1.5 Comparison of adjustment methods ...............................110

4.1.6 Overall efficiency of the pump system ........................... 111

4.1.7 Example: Relative power consumption

when the flow is reduced by 20% ........................................ 111

Section 4.2 Speed-controlled pump solutions .................... 114

4.2.1 Constant pressure control ..........................................................114

4.2.2 Constant temperature control ...............................................115

4.2.3 Constant differential pressure in a

circulating system ............................................................................. 115

4.2.4 Flow-compensated differential

pressure control ...................................................................................116

Section 4.3 Advantages of speed control .................................117

Section 4.4 Advantages of pumps with integrated

frequency converter ............................................................................... 118

4.4.1 Performance curves of speed-controlled

pumps ...........................................................................................................119

4.4.2 Speed-controlled pumps in different systems ........119

Section 4.5 Frequency converter .................................................... 122

4.5.1 Basic function and characteristics ......................................122

4.5.2 Components of the frequency converter .....................122

4.5.3 Special conditions regarding frequency

converters ................................................................................................124

Chapter 5 Life cycle costs calculation ....................... 127

Section 5.1 Life cycle costs equation ............................................ 128

5.1.1 Initial costs, purchase price (Cic) ........................................... 129

5.1.2 Installation and commissioning costs (Cin) ................ 129

5.1.3 Energy costs (Ce) ................................................................................. 130

5.1.4 Operating costs (Co) ......................................................................... 130

5.1.5 Environmental costs (C

env

) ......................................................... 130

5.1.6 Maintenance and repair costs (Cm) .....................................131

5.1.7 Downtime costs, loss of production (Cs) .......................131

5.1.8 Decommissioning and disposal costs (Co) ...................131

Section 5.2 Life cycle costs calculation

– an example ................................................................................................132

Appendix .........................................................................................................133

A) Notations and units .........................................................................134

B) Unit conversion tables ...................................................................135

C) SI-prefixes and Greek alphabet ............................................ 136

D) Vapour pressure and density of water at

different temperatures ................................................................. 137

E) Orifice ..........................................................................................................138

F) Change in static pressure due to change

in pipe diameter ................................................................................. 139

G) Nozzles ........................................................................................................140

H) Nomogram for head losses in

bends, valves, etc. ..............................................................................141

I) Pipe loss nomogram for clean water 20˚C ..................142

J) Periodical system ................................................................................143

K) Pump standards ................................................................................. 144

L) Viscosity for different liquids as a function

of liquid temperature ....................................................................145

Index ..................................................................................................................151

Chapter 1. Design of pumps and motors

Section 1.1: Pump construction

1.1.1 The centrifugal pump

1.1.2 Pump curves

1.1.3 Characteristics of the centrifugal pump

1.1.4 Most common end-suction and in-line
pump types

1.1.5 Impeller types (axial forces)

1.1.6 Casing types (radial forces)

1.1.7 Single-stage pumps

1.1.8 Multistage pumps

1.1.9 Long-coupled and close-coupled pumps

Section 1.2: Types of pumps

1.2.1 Standard pumps

1.2.2 Split-case pumps

1.2.3 Hermetically sealed pumps

1.2.4 Sanitary pumps

1.2.5 Wastewater pumps

1.2.6 Immersible pumps

1.2.7 Borehole pumps

1.2.8 Positive displacement pumps

Section 1.1
Pump construction

1.1.1 The centrifugal pump

In 1689 the physicist Denis Papin invented the centrifugal
pump and today this kind of pump is the most used around
the world. The centrifugal pump is built on a simple
principle: Liquid is led to the impeller hub and by means
of the centrifugal force it is flung towards the periphery of
the impellers.
The construction is fairly inexpensive, robust and simple
and its high speed makes it possible to connect the pump
directly to an asynchronous motor. The centrifugal pump
provides a steady liquid flow, and it can easily be throttled
without causing any damage to the pump.

Now let us have a look at figure 1.1.1, which shows the
liquid’s flow through the pump. The inlet of the pump
leads the liquid to the centre of the rotating impeller from
where it is flung towards the periphery. This construction
gives a high efficiency, and is suitable for handling pure
liquids. Pumps, which have to handle impure liquids, such
as wastewater pumps, are fitted with an impeller that is
constructed especially to avoid that objects get stocked
inside the pump, see section 1.2.5.

If a pressure difference occurs in the system while the
centrifugal pump is not running, liquid can still pass
through it due to its open design.

As you can tell from figure 1.1.2, the centrifugal pump can
be categorised in different groups: Radial flow pumps,
mixed flow pumps and axial flow pumps. Radial flow
pumps and mixed flow pumps are the most common
types used. Therefore, we will only concentrate on these
types of pumps on the following pages.

However, we will briefly present the positive displacement
pump in section 1.2.8.

The different demands on the centrifugal pump’s
performance, especially with regards to head, flow, and
installation, together with the demands for economical
operation, are only a few of the reasons why so many types
of pump exist. Figure 1.1.3 shows the different pump types
with regard to flow and pressure.

Fig. 1.1.2: Different kinds of centrifugal pumps

Fig. 1.1.1: The liquid’s flow through the pump

Radial flow pump

Mixed flow pump

Axial flow pump

Fig. 1.1.3: Flow and head for different types of centrifugal pumps

1 22446610

1

10

1

2

4

6

10

2

2

4

6

10

3

2

4

6

10

4

H

[m]

Q [m

3

/s]

2 4 6 10

2

2 4 6 10

3

2 4 6 10

4

2 4 6 10

5

Multistage radial
flow pumps

Single-stage radial
flow pumps

Mixed flow pumps

Axial flow pumps

8

1.1.2 Pump curves

Before we dig any further into the world of pump
construction and pump types, we will present the
basic characteristics of pump performance curves. The
performance of a centrifugal pump is shown by a set
of performance curves. The performance curves for a
centrifugal pump are shown in figure 1.1.4. Head, power
consumption, efficiency and NPSH are shown as a function
of the flow.

Normally, pump curves in data booklets only cover the
pump part. Therefore, the power consumption, the P2­value, which is listed in the data booklets as well, only
covers the power going into the pump – see figure 1.1.4.
The same goes for the efficiency value, which only covers
the pump part (η = ηP).

In some pump types with integrated motor and possibly
integrated frequency converter, e.g. canned motor pumps
(see section 1.2.3), the power consumption curve and the
η-curve cover both the motor and the pump. In this case it
is the P1-value that has to be taken into account.

In general, pump curves are designed according to ISO
9906 Annex A, which specifies the tolerances of the
curves:

• Q +/- 9%,

• H +/-7%,

• P +9%

• -7%.

What follows is a brief presentation of the different pump
performance curves.

Head, the QH-curve

The QH-curve shows the head, which the pump is able
to perform at a given flow. Head is measured in meter
liquid column [mLC]; normally the unit meter [m] is
applied. The advantage of using the unit [m] as the unit
of measurement for a pump’s head is that the QH-curve is
not affected by the type of liquid the pump has to handle,
see section 2.2 for more information.

H

[m]

η

[%]

50

40

70

Efficiency

60

50

40

20

10

2

12

4

6

8

10

0

30

30

20

10

0

10

0

2

4

6

8

0 10 20 30 40 50 60 70

Q [m3/h]

P

2

[kW]

NP

SH

(m)

Power consumption

NP

SH

Fig. 1.1.4: Typical performance curves for a centrifugal
pump. Head, power consumption, efficiency and NPSH
are shown as a function of the flow

Fig. 1.1.5: The curves for power consumption and efficiency will
normally only cover the pump part of the unit – i.e. P2 and η

P

P

1

P

2

HM

3~

η

M η

P

Q

H

[m]

50

60

40

30

20

10

0

0 10 20 30 40 50 60 70 80

Q [m3/h]

Fig. 1.1.6: A typical QH-curve for a centrifugal pump;
low flow results in high head and high flow results
in low head

9

Efficiency, the η-curve

The efficiency is the relation between the supplied power
and the utilised amount of power. In the world of pumps,
the efficiency ηP is the relation between the power, which
the pump delivers to the water (PH) and the power input
to the shaft (P2 ):

where:

ρ is the density of the liquid in kg/m

3

,
g is the acceleration of gravity in m/s2,
Q is the flow in m3/s and H is the head in m.

For water at 20oC and with Q measured in m3/h and H in m,
the hydraulic power can be calculated as :

As it appears from the efficiency curve, the efficiency depends
on the duty point of the pump. Therefore, it is important to
select a pump, which fits the flow requirements and ensures
that the pump is working in the most efficient flow area.

Power consumption, the P2-curve

The relation between the power consumption of the pump
and the flow is shown in figure 1.1.8. The P2-curve of most
centrifugal pumps is similar to the one in figure 1.1.8 where
the P2 value increases when the flow increases.

NPSH-curve (Net Positive Suction Head)

The NPSH-value of a pump is the minimum absolute
pressure (see section 2.2.1) that has to be present at the
suction side of the pump to avoid cavitation.
The NPSH-value is measured in [m] and depends on the
flow; when the flow increases, the NPSH-value increases
as well; figure 1.1.9. For more information concerning
cavitation and NPSH, go to section 2.2.1.

50

60

70

80

40

30

20

10

0

0 10 20 30 40 50 60 70

Q [m3/h]

η

[%]

8

10

6

4

2
0

0 10 20 30 40 50 60 70

Q [m3/h]

P

2

[kW]

Fig. 1.1.7: The efficiency curve of a typical centrifugal pump

Fig. 1.1.8: The power consumption curve of a typical
centrifugal pump

10

0

2

4

6

8

0 10 20 30 40 50 60 7

0

Q [m3/h]

NPSH

[m]

η

p

=

PH
P

2

=

ρ . g . Q . H

P

2

Fig. 1.1.9: The NPSH-curve of a typical centrifugal pump

PH = 2.72

.

Q . H [W]

10

Section 1.1
Pump construction

1.1.3 Characteristics of the
centrifugal pump

The centrifugal pump has several characteristics and in
this section, we will present the most important ones.
Later on in this chapter we will give a more thorough
description of the different pump types.

• The number of stages

Depending on the number of impellers in the pump, a
centrifugal pump can be either a single-stage pump or a
multistage pump.

• The position of the pump shaft

Single-stage and multistage pumps come with horizontal
or vertical pump shafts. These pumps are normally
designated horizontal or vertical pumps. For more
information, go to section 1.1.4.

• Single-suction or double-suction impellers

Depending on the construction of the impeller, a pump can
be fitted with either a single-suction impeller or a double­suction impeller. For more information, go to section 1.1.5.

• Coupling of stages

The pump stages can be arranged in two different ways: in
series and in parallel, see figure 1.1.10.

• Construction of the pump casing

We distinguish between two types of pump casing: Volute
casing and return channel casing with guide vanes. For
more information, go to section 1.1.6.

Fig 1.1.10: Twin pump with parallel-coupled impellers

11

1.1.4 Most common end-suction and in-line pump types

End-suction pump = The liquid runs directly into the impeller. Inlet and outlet have a
90° angle. See section 1.1.9

In-line pump = The liquid runs directly through the pump in-line. The suction pipe and the discharge
pipe are placed opposite one another and can be mounted directly in the piping system

Split-case pump = Pump with an axially divided pump housing. See section 1.2.2

Horizontal pump = Pump with a horizontal pump shaft

Vertical pump = Pump with a vertical pump shaft

Single-stage pump = Pump with a single impeller. See section 1.1.7

Multistage pump = Pump with several series-coupled stages. See section 1.1.8

Long-coupled pump = Pump connected to the motor by means of a flexible coupling. The motor and
the pump have separate bearing constructions. See section 1.1.9

Close-coupled pump = A pump connected to the motor by means of a rigid coupling. See section 1.1.9

Horizontal

Close-coupled Close-coupled

End-suction

Single-stage

Long-coupled

Multistage

12

Section 1.1
Pump construction

13

Multistage

Horizontal / Vertical

Single-stage

Long-coupled Close-coupled

Close-coupled

In-line

Split-case

Single-stage

Long-coupled

Horizontal

14

1.1.5 Impeller types (axial forces)

A centrifugal pump generates pressure that exerts forces
on both stationary and rotating parts of the pump.
Pump parts are made to withstand these forces.
If axial and radial forces are not counterbalanced in the
pump, the forces have to be taken into consideration when
selecting the driving system for the pump (angular contact
bearings in the motor). In pumps fitted with single-suction
impeller, large axial forces may occur, figures 1.1.11 and

1.1.12. These forces are balanced in one of the following
ways:

• Mechanically by means of thrust bearings. These types
of bearings are specially designed to absorb the axial
forces from the impellers

• By means of balancing holes on the impeller,
see figure 1.1.13

• By means of throttle regulation from a seal ring
mounted on the back of the impellers, see figure 1.1.14

• Dynamic impact from the back of the impeller, see
figure 1.1.15

• The axial impact on the pump can be avoided by either
using double-suction impellers (see figure 1.1.16).

Fig. 1.1.11: Single-suction
impeller

Fig. 1.1.12: Standard pump with
single-suction impeller

Fig. 1.1.13: Balancing the axial forces
in a single-stage centrifugal pump
with balancing holes only

Fig. 1.1.14: Balancing the axial forces
in a single-stage centrifugal pump
with sealing gap at discharge side and
balancing holes

Fig. 1.1.15: Balancing the axial forces in
a single-stage centrifugal pump with
blades on the back of the impellers

Fig. 1.1.16: Balancing the axial
forces in a double-suction
impeller arrangement

Axial forces

Section 1.1
Pump construction

1.1.6 Casing types (radial forces)

Radial forces are a result of the static pressure in the
casing. Therefore, axial deflections may occur and lead
to interference between the impeller and the casing. The
magnitude and the direction of the radial force depend on
the flow rate and the head.

When designing the casing for the pump, it is possible
to control the hydraulic radial forces. Two casing types
are worth mentioning: the single-volute casing and the
double-volute casing. As you can tell from figure 1.1.18,
both casings are shaped as a volute. The difference between
them is, that the double-volute has an guide vane.

The single-volute pump is characterised by a symmetric
pressure in the volute at the optimum efficiency point,
which leads to zero radial load. At all other points,
the pressure around the impeller is not regular and
consequently a radial force is present.

As you can tell from figure 1.1.19, the double-volute casing
develops a constant low radial reaction force at any capacity.

Return channels (figure 1.1.20) are used in multistage pumps
and have the same basic function as volute casings. The
liquid is led from one impeller to the next and at the same
time, the rotation of water is reduced and the dynamic
pressure is transformed into static pessure. Because of the
return channel casing’s circular design, no radial forces are
present.

1.1.7 Single-stage pumps

Generally, single-stage pumps are used in applications,
which do not require a total head of more than 150 m.
Normally, single-stage pumps operate in the interval of
2-100 m.

Single-stage pumps are characterised by providing a low
head relative to the flow, see figure 1.1.3. The single-stage
pump comes in both a vertical and a horizontal design, see
figures 1.1.21 and 1.1.22.

15

Q/Qopt1.0

Volute casing

Double-volute
casing

Radial force

Fig. 1.1.18: Single-volute casing Double-volute casing

Fig. 1.1.22: Vertical single-stage
in-line close-coupled pump

Fig. 1.1.21: Horizontal single-stage
end-suction close-coupled pump

Radial forces

Fig. 1.1.17: Single-suction

impeller

Fig. 1.1.19: Radial force for single- and double-volute casing

Fig. 1.1.20: Vertical multistage
in-line pump with return
channel casing

Return channel

Fig. 1.1.25: Long-coupled pump
with basic coupling

Fig. 1.1.26: Long-coupled pump with spacer coupling

16

1.1.8 Multistage pumps

Multistage pumps are used in installations where a high
head is needed. Several stages are connected in series and
the flow is guided from the outlet of one stage to the inlet
of the next. The final head that a multistage pump can
deliver is equal to the sum of pressure each of the stages
can provide.

The advantage of multistage pumps is that they provide
high head relative to the flow. Like the single-stage pump,
the multistage pump is available in both a vertical and a
horizontal version, see figures 1.1.23 and 1.1.24.

1.1.9 Long-coupled and close-coupled
pumps

Long-coupled pumps

Long-coupled pumps are pumps with a flexible coupling
that connects the pump and the motor. This kind of
coupling is available either as a basic coupling or as a
spacer coupling.

If the pump is connected to the motor by a basic coupling,
it is necessary to dismount the motor when the pump
needs service. Therefore, it is necessary to align the pump
upon mounting, see figure 1.1.25.

On the other hand, if the pump is fitted with a spacer
coupling, it is possible to service the pump without
dismounting the motor. Alignment is thus not an issue,
see figure 1.1.26.

Close-coupled pumps

Close-coupled pumps can be constructed in the following
two ways: Either the pump has the impeller mounted
directly on the extended motor shaft or the pump has a
standard motor and a rigid or a spacer coupling, see figures

1.1.27 and 1.1.28.

Fig. 1.1.24: Horizontal multistage
end-suction pump

Fig. 1.1.23: Vertical
multistage in-line pump

Fig. 1.1.27: Close-coupled pump with
rigid coupling

Basic coupling
type

Long-coupled
pump with
flexible coupling

Close-coupled
pump with
rigid coupling

Spacer coupling
(option)

Fig. 1.1.28: Different coupling types

Section 1.1
Pump construction

17

Fig. 1.2.1: Long-coupled standard pump

Fig. 1.2.2: Bare shaft standard pump

Fig. 1.2.3: Long-coupled split-case pump

Fig. 1.2.4: Split-case pump
with double-suction impeller

1.2.1 Standard pumps

Few international standards deal with centrifugal pumps.
In fact, many countries have their own standards, which
more or less overlap one another. A standard pump is
a pump that complies with official regulations as to for
example the pump’s duty point. What follows, are a couple
of examples of international standards for pumps:

• EN 733 (DIN 24255) applies to end-suction centrifugal
pumps, also known as standard water pumps with a
rated pressure (PN) of 10 bar.

• EN 22858 (ISO 2858) applies to centrifugal pumps, also
known as standard chemical pumps with a rated
pressure (PN) of 16 bar, see appendix K.

The standards mentioned above cover the installation
dimensions and the duty points of the different pump
types. As to the hydraulic parts of these pump types, they
vary according to the manufacturer – thus, no international
standards are set for these parts.

Pumps, which are designed according to standards, provide
the end-user with advantages with regard to service, spare
parts and maintenance.

1.2.2 Split-case pumps

A split-case pump is a pump with the pump housing
divided axially into two parts. Figure 1.2.4 shows a single­stage split-case pump with a double-suction impeller.
The double-inlet construction eliminates the axial
forces and ensures a longer life span of the bearings.
Usually, split-case pumps have a rather high efficiency,
are easy to service and have a wide performance range.

Section 1.2
Types of pumps

Section 1.2
Types of pumps

18

1.2.3 Hermetically sealed pumps

It comes as no surprise that a pump’s shaft lead-in has to be
sealed. Usually, this is done by means of a mechanical shaft
seal, see figure 1.2.5. The disadvantage of the mechanical
shaft seal is its poor properties when it comes to handling
of toxic and aggressive liquids, which consequently leads
to leakage. These problems can to some extent be solved
by using a double mechanical shaft seal. Another solution
to these problems is to use a hermetically sealed pump.

We distinguish between two types of hermetically sealed
pumps: Canned motor pumps and magnetic-driven pumps.
In the following two sections, you can find additional
information about these pumps.

Canned motor pumps

A canned motor pump is a hermetically sealed pump with
the motor and pump integrated in one unit without a seal,
see figures 1.2.6 and 1.2.7. The pumped liquid is allowed to
enter the rotor chamber that is separated from the stator
by a thin rotor can. The rotor can serves as a hermetically
sealed barrier between the liquid and the motor. Chemical
pumps are made of materials, e.g. plastics or stainless
steel that can withstand aggressive liquids.

The most common canned motor pump type is the
circulator pump. This type of pump is typically used in
heating circuits because the construction provides low
noise and maintenance-free operation.

Fig. 1.2.5: Example of a standard pump with mechanical shaft seal

Liquid

Atmosphere

Seal

Fig. 1.2.7: Circulator pump
with canned motor

Motor can

Fig. 1.2.6: Chemical pump with canned motor

Motor can

Magnetic-driven pumps

In recent years, magnetic-driven pumps have become
increasingly popular for transferring aggressive and toxic
liquids.

As shown in figure 1.2.8, the magnetic-driven pump is
made of two groups of magnets; an inner magnet and
an outer magnet. A non-magnetizable can separate these
two groups. The can serves as a hermetically sealed barrier
between the liquid and the atmosphere. As it appears
from figure 1.2.9, the outer magnet is connected to the
pump drive and the inner magnet is connected to the
pump shaft. Hereby the torque from the pump drive is
transmitted to the pump shaft. The pumped liquid serves
as lubricant for the bearings in the pump. Therefore,
sufficient venting is crucial for the bearings.

19

Fig. 1.2.8: Construction of magnetic drive

Fig. 1.2.9: Magnetic-driven multistage pump

Can

Inner magnets

Outer magnets

Can

Inner magnets

Outer magnets

20

Fig. 1.2.10: Sanitary pump

Fig.1.2.11: Sanitary self-priming side-channel pump

1.2.4 Sanitary pumps

Sanitary pumps are mainly used in the food, beverage,
pharmaceutical and bio-technological industries where it
is important that the pumped liquid is handled in a gentle
manner and that the pumps are easy to clean.

In order to meet process requirements in these industries,
the pumps have to have a surface roughness between

3.2 and 0.4 µm Ra. This can be best achieved by using
forged or

deep-drawn rolled stainless steel as materials

of construction,

see figure 1.2.12. These materials have a
compact pore-free surface finish that can be easily worked
up to meet the various surface finish requirements.

The main features of a sanitary pump are ease of cleaning
and ease of maintenance.

The leading manufacturers of sanitary pumps have
designed their products to meet the following standards:

EHEDG –

European Hygienic Equipment Design Group

QHD – Qualified Hygienic Design

3-A – Sanitary Standards:

3A0/3A1: Industrial/Hygienic Standard

Ra ≤ 3.2 µm

3A2: Sterile Standard

Ra ≤ 0.8 µm

3A3: Sterile Standard

Ra ≤ 0.4 µm

Sand casting

Precision casting

Rolled steel

Fig.1.2.12: Roughness of material surfaces

Section 1.2
Types of pumps

1.2.5 Wastewater pumps

A wastewater pump is an enclosed unit with a pump
and a motor. Due to this construction the wastewater
pump is suitable for submersible installation in pits.
In submersible installations with auto-coupling systems
double rails are normally used. The auto-coupling system
facilitates maintenance, repair and replacement of the
pump. Because of the construction of the pump, it is
not necessary to enter the pit to carry out service. In
fact, it is possible to connect and disconnect the pump
automatically from the outside of the pit. Wastewater
pumps can also be installed dry like conventional pumps
in vertical or horizontal installations. Likewise this type of
installation provides easy maintenance and repair like it
provides uninterrupted operation of the pump in case of
flooding of the dry pit, see figure 1.2.14.

Normally, wastewater pumps have to be able to handle large
particles. Therefore they are fitted with special

impellers that
make it possible to avoid blockage and clogging. Different
types of impellers exist; single-channel impellers, double­channel impellers three and four-channel impellers and
vortex impellers.

Figure 1.2.15 shows the different designs of

these impellers.

Wastewater pumps usually come with a dry motor, which
is IP68 protected (for more information on IP-classes,
go to section 1.4.1). Motor and pump have a common
extended shaft with a double mechanical shaft seal system
in an intermediate oil chamber, see figure 1.2.13.
Wastewater pumps are able to operate either intermittently
or continuously depending on the installation in question.

Fig. 1.2.14: Wastewater pump for dry installations

Fig. 1.2.15: Impeller types for wastewater

Vortex

impeller

Single-channel

impeller

Double-channel

impeller

21

Fig.1.2.13: Detail of a sewage pump
for wet installations

1.2.6 Immersible pumps

An immersible pump is a pump type where the pump part
is immersed in the pumped liquid and the motor is kept
dry. Normally, immersible pumps are mounted on top of
or in the wall of tanks or containers. Immersible pumps
are for example used in the machine tool industry

for
example in spark machine tools, grinding machines,
machining

centres and cooling units or in other industrial
applications involving tanks or containers, such as
industrial washing and filtering systems.

Pumps for machine tools can be divided in two groups:
Pumps for the clean side of the filter and pumps for the
dirty side of the filter. Pumps with closed impellers are
normally used for the clean side of the filter, because they
provide a high efficiency and a high pressure if necessary.
Pumps with open or semi-open impellers are normally
used for the dirty side of the filter, because they can
handle metal chips and particles.

Fig. 1.2.16: Immersible pump

22

Section 1.2
Types of pumps

1.2.7 Borehole pumps

Two types of borehole pumps exist: The submerged
borehole pump type with a submersible motor, and the
deep well pump with a dry motor, which is connected
to the pump by a long shaft. These pumps are normally
used in connection with water supply and irrigation. Both
pump types are made to be installed in deep and narrow
boreholes and have thus a reduced diameter, which makes
them longer than other pump types, see figure 1.2.17.

The borehole pumps are specially designed to be
submerged in a liquid and are thus fitted with a
submersible motor, which is IP68 protected. The pump
comes in both a single-stage and a multistage version (the
multistage version being the most common one), and is
fitted with a non-return valve in the pump head.

Today, the deep well pump has been more or less replaced
by the submerged pump type. The long shaft of the deep
well pump is a drawback, which makes it difficult to
install and carry out service. Because the deep well pump
motor is air-cooled, the pump is often used in industrial
applications to pump hot water from open tanks. The
submersible pump cannot handle as high temperatures
because the motor is submerged in the liquid, which has
to cool it.

Fig. 1.2.17: Submersible pump

23

1.2.8 Positive displacement pumps

The positive displacement pump provides an approximate
constant flow at fixed speed, despite changes in the
counterpressure. Two main types of positive displacement
pumps exist:

• Rotary pumps

• Reciprocating pumps

The difference in performance between a centrifugal
pump, a rotary pump and a reciprocating is

illustrated

to the right, figure 1.2.18.

Depending on which of these
pumps you are dealing with, a small change in the pump’s
counterpressure results in differences in the flow.

The flow of a centrifugal pump will change considerably,
the flow of a rotary pump will change a little, while the
flow of a reciprocating pump will hardly change at all.
But, why is there a difference between the pump curves
for reciprocating pumps and rotary pumps? The actual
seal face surface is larger for rotary pumps than for
reciprocating pumps. So, even though the two pumps are
designed with the same tolerances, the gap loss of the
rotary pump is larger.

The pumps are typically designed with the finest tolerances
possible to obtain the highest possible efficiency and
suction capability. However, in some cases, it is necessary
to increase the tolerances, for example when the pumps
have to handle highly viscous liquids, liquids containing
particles and liquids of high temperature.

Positive displacement pumps are pulsate, meaning that
their volume flow within a cycle is not constant.
The variation in flow and speed leads to pressure fluctuations
due to resistance in the pipe system and in valves.

Q

H

H

23

1

3

2 1

Fig. 1.2.19: Classification of positive displacement pumps

Simplex

Duplex

Simplex

Duplex

Triplex

Multiplex

Fig. 1.2.18: Typical relation between
flow and head for 3 different pump types:

1) Centrifugal pumps

2) Rotary pumps

3) Reciprocating pumps

24

Section 1.2
Types of pumps

Reciprocating

Rotary

Plunger

Diaphragm

Steam Double-acting

Power

Single-acting

Double-acting

Gear

Lobe

Circumferential piston

Screw

Vane

Piston

Flexible member

Screw

Single rotor

Multiple rotor

Positive
displacement
pumps

+

Dosing pumps

The dosing pump belongs to the positive displacement
pump family and is typically of the diaphragm type. Diaphragm
pumps are leakage-free, because the diaphragm forms
a seal between the liquid and the surroundings.

The diaphragm pump is fitted with two non-return valves
– one on the suction side and one on the discharge side
of the pump. In connection with smaller diaphragm pumps,
the diaphragm is activated by the connecting rod, which is
connected to an electromagnet. Thereby the coil receives
the exact amount of strokes needed, see figure 1.2.21.

In connection with larger diaphragm pumps the
diaphragm is typically mounted on the connecting

rod,
which is activated by a camshaft. The camshaft is turned by
means of a standard asynchronous motor, see figure 1.2.22.

The flow of a diaphragm pump is adjusted by either
changing the stroke length and/or the frequency of the
strokes. If it is necessary to enlarge the operating area,
frequency converters can be connected to the larger
diaphragm pumps, see figure 1.2.22.

Yet, another kind of diaphragm pump exists. In this case,
the diaphragm is activated by means of an excentrically
driven connecting rod powered by a stepper motor or a
synchronous motor, figures 1.2.20 and 1.2.23. By using a
stepper motor drive the pump’s dynamic area is increased
and its

accuracy is improved considerably. With this

construction

it is no longer necessary to adjust the pump’s
stroke length because the connection rod is mounted
directly on the diaphragm. The result is optimised suction
conditions and excellent operation features.

So therefore, it is simple to control both the suction
side and the discharge side of the pump. Compared to
traditional electromagnetic-driven diaphragm pumps
which provide powerful pulsations, stepper motor-driven
diaphragm pumps make it possible to get a much more
steady dosage of additive.

Fig.1.2.21: Solenoid spring return

1.2.22: Cam-drive spring return

1.2.23: Crank drive

Fig. 1.2.20: Dosing pump

+

25

Chapter 1. Design of pumps and motors

Section 1.3: Mechanical shaft seals

1.3.1 The mechanical shaft seal’s components
and function

1.3.2 Balanced and unbalanced shaft seals

1.3.3 Types of mechanical shaft seals

1.3.4 Seal face material combinations

1.3.5 Factors affecting the seal performance

Section 1.3

Mechanical shaft seals

From the middle of the 1950s mechanical shaft seals
gained ground in favour of the traditional sealing method

- the stuffing box. Compared to stuffing boxes, mechani­cal shaft seals provide the following advantages:

• They keep tight at smaller displacements and vibrations

in the shaft

• They do not require any adjustment

• Seal faces provide a small amount of friction and thus,

minimise the power loss

• The shaft does not slide against any of the seal’s

components and thus, is not damaged because of
wear (reduced repair costs).

The mechanical shaft seal is the part of a pump that
separates the liquid from the atmosphere. In figure 1.3.1
you can see a couple of examples where the mechanical
shaft seal is mounted in different types of pumps.

The majority of mechanical shaft seals are made
according to the European standard EN 12756.

Before choosing a shaft seal, there are certain things
you need to know about the liquid and thus the
seal’s resistance to the liquid:

• Determine the type of liquid

• Determine the pressure that the shaft seal is exposed to

• Determine the speed that the shaft seal is exposed to

• Determine the built-in dimensions

On the following pages we will present how a mechanical shaft
seal works, the different types of seal, which kind of materials
mechanical shaft seals are made of and which factors that
affect the mechanical shaft seal’s performance.

28

Fig. 1.3.1: Pumps with mechanical shaft seals

1.3.1 The mechanical shaft seal’s
components and function

The mechanical shaft seal is made of two main components:
a rotating part and a stationary part; and consists of the parts
listed in figure 1.3.2. Figure 1.3.3 shows where the different
parts are placed in the seal.

• The stationary part of the seal is fixed in the pump

housing. The rotating part of the seal is fixed on the
pump shaft and rotates when the pump operates.

• The two primary seal faces are pushed against each other

by the spring and the liquid pressure. During operation
a liquid film is produced in the narrow gap between the
two seal faces. This film evaporates before it enters the
atmosphere, making the mechanical shaft seal liquid tight,
see figure 1.3.4.

• Secondary seals prevent leakage from occurring

between the assembly and the shaft.

• The spring presses the seal faces together mechanically.

• The spring retainer transmits torque from the shaft to

the seal. In connection with mechanical bellows shaft
seals, torque is transferred directly through the bellows.

Seal gap

During operation the liquid forms a lubricating film
between the seal faces. This lubricating film consists of a
hydrostatic and a hydrodynamic film.

• The hydrostatic element is generated by the pumped

liquid which is forced into the gap between the seal faces.

•

The hydrodynamic lubricating film is created by

pressure generated by the shaft’s rotation.

Fig. 1.3.4: Mechanical shaft seal in operation

Lubrication film

Liquid force

Spring force

Vapour

Evaporation
begins

Fig. 1.3.3: Main components of the
mechanical shaft seal

Rotating part

Stationary part

Shaft

Primary seal

Secondary seal

Primary seal

Secondary seal

Spring

Spring retainer

Mechanical shaft seal Designation

Seal face (primary seal)

Secondary seal
Spring

Spring retainer (torque transmission

)

Seat (seal faces, primary seal)

Static seal (secondary seal

)

Rotating par

t

Stationary par

t

Fig. 1.3.2:

The mechanical shaft seal’s components

29

1.3.2 Balanced and unbalanced shaft seals

To obtain an acceptable face pressure between the
primary seal faces, two kind of seal types exist: a balanced
shaft seal and an unbalanced shaft seal.

Balanced shaft seal

Figure 1.3.6 shows a balanced shaft seal indicating where
the forces interact on the seal.

Unbalanced shaft seal

Figure 1.3.7 shows an unbalanced shaft seal indicating
where the forces interact on the seal.

Several different forces have an axial impact on the seal
faces. The spring force and the hydraulic force from the
pumped liquid press the seal together while the force from
the lubricating film in the seal gap counteracts this. In
connection with high liquid pressure, the hydraulic forces
can be so powerful that the lubricant in the seal gap cannot
counteract the contact between the seal faces. Because the
hydraulic force is proportionate to the area that the liquid
pressure affects, the axial impact can only be reduced by
obtaining a reduction of the pressure-loaded area.

The thickness of the lubricating film depends on the pump
speed, the liquid temperature, the viscosity of the liquid
and the axial forces of the mechanical shaft seal. The
liquid is continuously changed in the seal gap because of

• evaporation of the liquid to the atmosphere

• the liquid’s circular movement

Figure 1.3.5 shows the optimum ratio between fine
lubrication properties and limited leakage. As you can tell,
the optimum ratio is when the lubricating film covers the
entire seal gap, except for a very narrow evaporation

zone

close to the atmospheric side of the mechanical shaft seal.

Leakage due to deposits on the seal faces is often observed.
When using coolant agents, deposits are built up quickly by
the evaporation at the atmosphere side of the seal. When
the

liquid evaporates in the evaporation zone, microscopic

solids

in the liquid remain in the seal gap as deposits
creating wear.
These deposits are seen in connection with most types
of liquid. But when the pumped liquid has a tendency
to crystallise, it can become a problem. The best way to
prevent wear is to select seal faces made

of hard material,
such as tungsten carbide (WC) or silicon carbide (SiC).
The narrow seal gap between these materials (app. 0.3 µm
Ra) minimises the risk of solids entering the seal gap and
thereby minimises the amount of deposits building up.

Pressure

LiquidPump pressure

Stationary
seal face

Rotating
seal face

Vapour Atmosphere

Entrance

in seal

Exit into

atmospher

e

Start of
ev

aporation

1 atm

Fig. 1.3.6: Interaction of
forces on the balanced
shaft seal

Fig. 1.3.7: Interaction of
forces on the unbalanced
shaft seal

A

Spring forces

Hydraulic forces

Contact area of seal faces

B

A B

Hydraulic forces

Contact area of seal faces

Fig. 1.3.5: Optimum ratio between fine lubrication
properties and limited leakage

30

Section 1.3

Mechanical shaft seals

31

The balancing ratio (K) of a mechanical shaft seal is defined
as the ratio between the area A and the area (B) : K=A/B

K = Balancing ratio
A = Area exposed to hydraulic pressure
B = Contact area of seal faces

For balanced shaft seals the balancing ratio is usually
around K=0.8 and for unbalanced shaft seals the balancing
ratio is normally around K=1.2.

1.3.3 Types of mechanical shaft seals

What follows is a brief outline of the main types of
mechanical shaft seals: O-ring seal, bellows seal, and the
one-unit seal - the cartridge seal.

O-ring seals

In an O-ring seal, sealing between the rotating shaft and
the rotating seal face is effected by an O-ring (figure 1.3.9).
The O-ring must be able to slide freely in the axial direction
to absorb axial displacements as a result of changes
in temperatures and wear. Incorrect positioning of the
stationary seat may result in rubbing and thus unnecessary
wear on the O-ring and on the shaft. O-rings are made of
different types of rubber material, such as NBR, EPDM and
FKM, depending on the operating conditions.

Bellows seals

A common feature of bellows seals is a rubber or metal
bellows which functions as dynamic sealing element
between the rotating ring and the shaft.

Rubber bellows seals

The bellows of a rubber bellows seal (see figure 1.3.10) can
be made of different types of rubber material, such as NBR,
EPDM and FKM, depending on the operating conditions.
Two different geometric principles are used for the
design of rubber bellows:

• Folding bellows

• Rolling bellows.

Fig. 1.3.8: Wear rate for different balancing ratios

Temperature (oC)

0 20 40 60 80 100 120 140

Wear rate comparative

Temperature (oC)

K = 1.15
K = 1.00
K = 0.85

Fig. 1.3.9: O-ring seal

Fig. 1.3.10: Rubber bellows seal

Advantages and
disadvantages of
O-ring seal

Advantages:

Suitable in hot liquid and
high pressure applications

Disadvantages:

Deposits on the shaft,
such as rust, may prevent
the O-ring shaft
seal from moving axially

Advantages and
disadvantages of
rubber bellows seal

Advantages:
Not sensitive to deposits,
such as rust, on the shaft

Suitable for pumping
solid-containing liquids

Disadvantages:

Not suitable in hot liquid and
high pressure applications

Rubber bellows seal with folding
bellows geometry

32

Metal bellows seals

In an ordinary mechanical shaft seal, the spring produces
the closing force required to close the seal faces. In a
metal bellows seal (figure 1.3.11) the spring has been
replaced by a metal bellows with a similar force.
Metal bellows act both as a dynamic seal between the
rotating ring and the shaft and as a spring. The bellows
have a number of corrugations to give them the desired
spring force.

Cartridge seals

In a cartridge mechanical shaft seal, all parts form a compact
unit on a shaft sleeve, ready to be installed. A cartridge seal
offers many benefits compared to conventional mechanical
shaft seals, figure 1.3.12.

Flushing

In certain applications it is possible to extend the
performance of the mechanical shaft seal by installing
flushing, see figure 1.3.13. Flushing can lower the
temperature of the mechanical shaft seal and prevent
deposits from occurring. Flushing can be installed either
internally or externally. Internal flushing is done when a
small flow from the pump’s discharge side is bypassed
to the seal area. Internal flushing is primarily used to
prevent further heat generation from the seal in heating
applications. External flushing is done by a flushing liquid
and is used to ensure trouble-free operation when handling
liquids that are abrasive or contain clogging solids.

Fig. 1.3.11: Cartridge metal
bellows seal

Advantages and
disadvantages of cartridge
metal bellows seal

Advantages:

Not sensitive to deposits,
such as rust and lime on
the shaft

Suitable in hot liquid and
high-pressure applications

Low balancing ratio leads
to low wear rate and
consequently longer life

Disadvantages:

Fatique failure of the
mechanical shaft seal may
occur when the pump is not
aligned correctly

Fatique may occur as a
result of excessive
temperatures or pressures

Fig. 1.3.12: Cartridge seal

Advantages of the
cartridge seal:

• Easy and fast service

• The design protects the
seal faces

• Preloaded spring

• Safe handling

Fig 1.3.13: Flushing device of a
single mechanical shaft seal

Section 1.3

Mechanical shaft seals

33

Double mechanical shaft seals

Double mechanical shaft seals are used when the life
span of single mechanical shaft seals is insufficient due
to wear caused by solids or too high/low pressure and
temperature. Further, double mechanical shaft seals are
used in connection with toxic, aggressive and explosive liquids
to protect the surroundings. Two types of double mechanical
shaft seals exist: The double seal in a tandem arrangement
and the double seal in a back-to-back arrangement.

Double seal in tandem

This type of double seal consists of two mechanical shaft
seals mounted in tandem, that is one behind the other,
placed in a separate seal chamber, see figure 1.3.14.

The seal type is used when a pressurised double mechanical
shaft seal mounted in a back-to-back arrangement is not
necessary.

The tandem seal arrangement has to be fitted with a
quenching liquid system which

•

absorbs leakage

•

monitors the leakage rate

•

lubricates and cools the outboard seal to prevent icing

•

protects against dry-running

•

stabilises the lubricating film

•

prevents air from entering the pump in case of vacuum

The pressure of the quenching liquid must always be lower
than the pumped liquid pressure.

Tandem - circulation

Circulation of quenching liquid via a pressureless tank, see
figure 1.3.14. Quenching liquid from the elevated tank is
circulated by thermosiphon action and/or by pumping action
in the seal.

Tandem - dead end

Quenching liquid from an elevated tank, see figure 1.3.15. No
heat is dissipated from the system.

Tandem - drain

The quenching liquid runs directly through the seal chamber
to be collected for reuse, or directed to drain, see figure 1.3.16.

Fig. 1.3.16: Tandem seal arrangement with quench liquid to drain

•

•

Quench liquid

Pumped liquid

•

•

•

•

Quench liquid

Pumped liquid

Quench liquid

Pumped liquid

•

•

•

•

•

•

Quench liquid

Pumped liquid

Quench liquid

Pumped liquid

Pumped
liquid

Fig. 1.3.15: Tandem seal arrangement with quench liquid dead

end

Fig. 1.3.14: Tandem seal arrangement with quench liquid

circulation

34

1.3.4 Seal face material combinations

What follows is a description of the most important
material pairings used in mechanical shaft seals for
industrial applications: Tungsten carbide/tungsten
carbide, silicon carbide/silicon carbide and carbon/
tungsten carbide or carbon/silicon carbide.

Tungsten carbide/tungsten carbide
(WC/WC)

Cemented tungsten carbide covers the type of hard metals
that are based on a hard tungsten carbide (WC) phase and
usually a softer metallic binder phase. The correct technical
term is cemented tungsten carbide, however, the abbreviated
term tungsten carbide (WC) is used for convenience.

Cobalt-bonded (Co) WC is only corrosion resistant in water if
the pump incorporates base metal, such as cast iron.

Chromium-nickel-molybdenum-bounded WC has a
corrosion resistance equal to EN 14401.

Sintered binderless WC has the highest corrosion
resistance. However, its resistance to corrosion in liquids,
such as hypochlorite, is not as high. The material pairing
WC/WC has the following features:

• Extremely wear resistant

• Very robust, resists rough handling

•

Poor dry-running properties. In case of dry-running, the
temperature increases to several hundred degrees Celsius
in very few minutes and consequently damages the O-rings.

If a certain pressure and a certain temperature are
exceeded, the seal may generate noise. Noise is an
indication of poor seal operating conditions that in the
long term may cause wear of the seal. The limits of use
depend on seal face diameter and design.

To a WC/WC seal face pairing, the running-in wear period
where noise is to be expected may last 3-4 weeks, although
typically, no noise occurs during the first 3-4 days.

Double seal in back-to-back

This type of seal is the optimum solution for handling
abrasive, aggressive, explosive or sticky liquids, which
would either wear out, damage or block a mechanical
shaft seal.

The back-to-back double seal consists of two shaft
seals mounted back-to-back in a separate seal chamber,
see figure 1.3.17. The back-to-back double seal protects
the surrounding environment and the people working
with the pump.

The pressure in the seal chamber must be 1-2 bar higher
than the pump pressure. The pressure can be generated
by:

• An existing, separate pressure source. Many

applications incorporate pressurised systems.

• A separate pump, e.g. a dosing pump.

•

Fig. 1.3.17:
Back-to-back seal arrangement

Seal chamber with

barrier pressure liquid

Pumped liquid

Barrier pressure liquid

Section 1.3

Mechanical shaft seals

Silicon carbide/silicon carbide
(SiC/SiC)

Silicon carbide/silicon carbide (SiC/SiC) is an alternative to
WC/WC and is used where higher corrosion resistance is
required .

The SiC/SiC material pairing has the following features:

• Very brittle material requiring careful handling

• Extremely wear resistant

• Extremely good corrosion resistance. SiC (Q

1

s

, Q

1

P

and

Q

1

G

) hardly corrodes, irrespective of the pumped liquid

type. However, an exception is water with very poor

conductivity, such as demineralised water, which attacks

the SiC variants Q

1

s

and Q

1

P

, whereas Q

1

G

is corrosion-

resistant also in this liquid

•

In general, these material pairings have poor dry-running

properties however, the Q

1

G

/ Q

1

G

material withstands a
limited period of dry-running on account of the
graphite content of the material

For different purposes, various SiC/SiC variants exist:

Q

1

s

, dense-sintered, fine-grained SiC

A direct-sintered, fine-grained SiC with a small amount of
tiny pores.

For a number of years, this SiC variant was used as a
standard mechanical shaft seal material. Pressure and
temperature limits are slightly below those of WC/WC.

Q

1

P

, porous, sintered, fine-grained SiC

A variant of the dense-sintered SiC. This SiC variant has
large circular closed pores. The degree of porosity is 5-15%
and the size of the pores 10-50 µm Ra.
The pressure and temperature limits exceed those of WC/WC.

35

Consequently, in warm water, the Q

1

P

/ Q

1

P

face material
pairing generates less noise than the WC/WC pairing.
However, noise from porous SiC seals must be expected
during the running-in wear period of 3-4 days.

Q

1

G

self-lubricating, sintered SiC

Several variants of SiC materials containing dry lubricants
are available on the market. The designation Q

1

G

applies
to a SiC material, which is suitable for use in distilled or
demineralised water, as opposed to the above materials.

Pressure and temperature limits of Q

1

G

/ Q

1

G

are similar to

those of Q

1

P

/ Q

1

P

.

The dry lubricants, i.e. graphite, reduce the friction in case
of dry-running, which is of decisive importance to the
durability of a seal during dry-running.

Carbon/tungsten carbide or carbon/
silicon carbide features

Seals with one carbon seal face have the following
features:

• Brittle material requiring careful handling

• Worn by liquids containing solid particles

• Good corrosion resistance

• Good dry-running properties (temporary dry-running)

• The self-lubricating properties of carbon make the

seal suitable for use even with poor lubricating
conditions (high temperature) without generating
noise. However, such conditions will cause wear of
the carbon seal face leading to reduced seal life. The
wear depends on the pressure, temperature, liquid
diameter and seal design.
Low speeds reduce the lubrication between the seal
faces; as a result, increased wear might have been
expected. However, this is normally not the case because
the distance that the seal faces have to move is reduced.

36

• Metal-impregnated carbon (A) offers limited corro-

sion resistance, but improved mechanical strength,
heat conductivity and thus reduced wear

• With reduced mechanical strength, but higher

corrosion resistance, synthetic resin-impregnated
carbon (B) covers a wide application field. Synthetic

resin-impregnated carbon is approved for drinking

water

•

The use of carbon/SiC for hot water applications may
cause heavy wear of the SiC, depending on the
quality of the carbon and water. This type of wear
primarily applies to Q

1

S

/carbon. The use of Q

1

P

,

Q

1

G

or a carbon/WC pairing causes far less wear. Thus,

carbon/WC, carbon/Q

1

P

or carbon/Q

1

G

are recommended

for hot water systems

1.3.5 Factors affecting the seal
performance

As mentioned previously, no seal is completely tight. On
the next pages, we will present the following factors,
which have an impact on the seal performance: Energy
consumption, noise and leakage. These factors will be
presented individually. However, it is important to stress
that they are closely interrelated, and thus must be
considered as a whole.

Energy consumption

It comes as no surprise that power is needed to make the
seal rotate. The following factors contribute to the power
consumption, that is the power loss of a mechanical shaft
seal:

• The centrifugal pumping action of the rotating parts.

The power consumption increases dramatically with
the speed of rotation (to the third power).

• The seal face friction.

Friction between the two seal faces consists of
– friction in the thin liquid film and
– friction due to points of contact between the
seal faces.

The level of power consumption depends on seal design,
lubricating conditions and seal face materials.

Figure 1.3.18 is a typical example of the power consumption
of a mechanical shaft seal. The figure shows that up to 3600
rpm friction is the major reason for the mechanical shaft
seal’s energy consumption.

Speed (rpm)

0

0

50

100

150

200

250

2000 4000 6000 8000 10000 12000

Power loss (W)

3600

Fig. 1.3.18: Power consumption of a 12 mm
mechanical shaft seal

Speed (rpm)

Pumping
action

Friction

Section 1.3

Mechanical shaft seals

37

Bar

25

20

15

10

5

0

10 20 30 40 50 60 70 80 90 100 110 °C

Noise

Duty range

Fig. 1.3.20: Relation between duty range and speed

Speed at 3000 rpm

Speed at

1800 rpm

Speed at

1200 rpm

Speed at 600 rpm

Energy consumption is, especially in connection with
stuffing boxes, an important issue. As you can tell from
the example, replacing a stuffing box by a mechanical
shaft seal leads to considerable energy savings, see
figure 1.3.19.

Noise

The choice of seal face materials is decisive for the
function and the life of the mechanical shaft seal.

Noise
is generated as a result of the poor lubricating conditions
in seals handling low viscosity liquids. The viscosity
of water decreases with increasing temperature. This
means that the lubricating conditions decrease as the
temperature rises. If the pumped liquid reaches or
exceeds boiling temperature, the liquid on part of the
seal face evaporates, which results in a further decrease
in the lubricating conditions. A speed reduction has the
same effect, see figure 1.3.20.

Leakage

The pumped liquid lubricates the seal face of a
mechanical shaft seal. Thus, better lubrication means
less friction and increased leakage. Conversely, less
leakage means worse lubricating conditions and
increased friction.

In practice, the amount of leakage
and power loss that occur in mechanical shaft seals can
vary. The reason is that leakage depends on factors which
are impossible to quantify theoretically because of type
of seal faces, type of liquid, spring load, etc. Therefore,
figure 1.3.21 should be perceived as a guideline.

To read the leakage rate curve correctly (figure 1.3.21),
you have to go through four steps:

Step 1: Read the pressure - in this case 5 bar

Step 2

: 30 mm unbalanced shaft seal

Step 3

: Speed 3000 rpm

Step 4

: Leakage rate 0.06 ml/h

Dw (mm) 100 B

100 U

B = balanced
U = unbalanced

n (min

-1

) 3600

3000

1800

1500

80 U

60 U

40 U

0.

001 0.01 0.06 0.1 1 Leakage Q (ml/h)

Differential pressure to be sealed p (bar)1 10

5

100

30 U

20 U

80 B

60 B

40 B

20 B

30 B

Fig. 1.3.21: Leakage rates

Fig. 1.3.19: Stuffing box versus mechanical shaft seal

Standard pump 50 mLC; 50 mm shaft and 2900 rpm

Energy consumption

Stuffing box 2.0 kW

h

Mechanical shaft seal 0.3 kW

h

Leakag

e

Stuffing box 3.0 l/h (when mounted correctly

)

Mechanical shaft seal 0.8 ml/h

Chapter 1. Design of pumps and motors

Section 1.4: Motors

1.4.1 Standards

1.4.2 Motor start-up

1.4.3 Voltage supply

1.4.4 Frequency converter

1.4.5 Motor protection

Motors are used in many applications all over the world.
The purpose of the electric motor is to create rotation, that is
to convert electric energy into mechanical energy. Pumps are
operated by means of mechanical energy, which is provided
by electric motors.

1.4.1 Standards

NEMA

The National Electrical Manufacturers Association (NEMA)
sets standards for a wide range of electric products, including
motors. NEMA is primarily associated with motors used in
North America. The standards represent general industry
practices and are supported by the manufacturers of electric
equipment. The standards can be found in NEMA Standard
Publication No. MG1. Some large motors may not fall under
NEMA standards.

IEC

The International Electrotechnical Commission (IEC) sets
standards for motors used in many countries around the
world. The IEC 60034 standard contains recommended
electrical practices that have been developed by the
participating IEC countries.

Fig. 1.4.1: Electric motor

Fig. 1.4.2: NEMA and IEC standards

Section 1.4
Motors

40

Directives and methods of protection –
Ex-motors

ATEX (ATmosphère EXplosible) refers to two EU directives
about danger of explosion within different areas. The ATEX
directive concerns electrical, mechanical, hydraulic and
pneumatic equipment. As to the mechanical equipment,
the safety requirements in the ATEX directive ensure that
pump components, such as shaft seals and bearings do not
heat up and ignite gas and dust. The first ATEX directive
(94/9/EC) deals with requirements put on equipment for
use in areas with danger of explosion. The manufacturer
has to fulfil the requirements and mark his products with
categories. The second ATEX directive (99/92/EC) deals
with the minimum safety and health requirements that
the user has to fulfil, when working in areas with danger of
explosion. Different techniques are used to prevent electric
equipment from becoming a source of ignition. In the
case of electric motors, protection types d (flameproof), e
(increased safety) and nA (non-sparking) are applied
in connection with gas, and DIP (dust ignition proof) is
applied in connection with dust.

Flameproof motors - protection type EExd (de)

First of all, flameproof EExd (type de) motors are category
2G equipment for use in zone 1. The stator housing and
the flanges enclose the flameproof motor parts that
can ignite a potentially explosive atmosphere. Because
of the enclosure, the motor can withstand the pressure
that goes along with the explosion of an explosive
mixture inside the motor. Propagation of the explosion
to the atmosphere that surrounds the enclosure is hereby
avoided because the explosion is cooled down by means
of flame paths. The size of the flame paths is defined in
the EN 50018 standard. The surface temperature of the
flameproof enclosure should always be in accordance with
the temperature classes.

Increased safety motors - protection type EEx (e)

Increased safety motors (type e) are category 2G equipment
for use in zone 1. These motors are not flameproof and not
built to withstand an internal explosion. The construction
of such a motor is based on increased security against

5

User

Manufacturer

Zones:
Gas (G): 0, 1 and 2
Dust (D): 20, 21 and 22

Minor

danger

Po

tentiel

danger

Constant

danger

Category 3
equipment
(3G/3D)

Category 2
equipment
(2G/2D)

Category 1
equipment
(1G/1D)

Zone:

0 or 20

Zone:
1 or 21

Zone:
1 or 21

Zone:
2 or 22

Zone:

2 or 22

Fig 1.4.4: The explosion occurs
inside the motor and is lead
out of the motor through the
flame paths. The temperature
classification for flameproof
EExd motors is valid for
external surfaces.

Fig 1.4.3: The link
between zones and
equipment categories is
a minimum requirement.
If the national rules are
more strict, they are the
ones to follow.

Fig 1.4.5: For increased safety
motors EExe, no sparks may
occur. The temperature
classification covers both
internal and external surfaces.

Fig 1.4.6: With non-sparking
motors ExnA, no ignition is
likely to occur.

41

possible excessive temperatures and occurrence of sparks
and arcs during normal operation and when a predictable
error occurs. The temperature classification for increased
safety motors is valid for both internal and external
surfaces, and therefore, it is important to observe the
stator winding temperature.

Non-sparking motors - protection type Ex (nA)

Non-sparking motors (type nA) are category 3G equipment
for use in zone 2. These motors cannot by any means ignite
a potential explosive atmosphere under normal operation,
see figure 1.4.6.

Dust Ignition Proof (DIP)

Two types of Dust Ignition Proof motors exist: 2D/category
2 equipment and 3D/category 3 equipment.

2D/category 2 equipment

In order to avoid static electricity to cause ignition, the
cooling fan on a category 2 DIP motor for use in zone 21
(area with potential danger of explosion) is made of metal.
Likewise, to minimise the risk of ignition, the external
ground terminal is subject to more severe demands of
construction. The external surface temperature of the
enclosure, which is indicated on the motor nameplate and
corresponds to the running performance during the worst
conditions allowed for the motor. Motors for use in zone 21
(areas with potential danger of explosion) have to be IP65
protected, that is completely protected against dust.

3D/category 3 equipment

The temperature indicated on a category 3 DIP motor for
use in zone 22 (areas with minor danger of explosion)

Type of

pr

otection

Code

Standar

ds

Use in A TEX

category/

Zone

Principle Application

CENELEC

EN

IE

C

60079

General
requirements

- 50014 - 0

-

Basic electrical requirements All equipment

Oil immersion

o 50015 - 6

Category 2
Zone 1

Electrical components immersed in oil
excluding explosive atmosphere from
igniting

Transformers

Pressurised

p 50016 - 2

Category 2
Zone 1

Enclosure housing equipment is purged to
remove explosive atmosphere and pres­surised to prevent ingress of
surrounding atmosphere

Switching and
control cabinets,
large motors

Powder filled

q 50017 - 5

Category 2
Zone 1

Electrical parts are surrounded with
power, e.g. quartz to prevent contact with
an explosive atmosphere

Electronic devices,
e.g. capacitors, fuses

Flameproof

d 50018 - 1

Category 2
Zone 1

Enclosure housing electrical equipment
which, if there is an internal explosion,
will not ignite surrounding atmosphere

control panels,

light fittings

Increased
safety

e 50019 - 7

Category 2
Zone 1

Additional methods are used to
eliminate arcs, sparks and hot surface
capable of igniting flammable
atmosphere

AC motors

AC motors,

, terminal
and connection boxes,
light fittings, squirrel
cage motors

Intrinsic safety

i

a

i

b

50020

50020

- 11

- 11

Category 1
Zone 0

Category 2
Zone 1

Electrical energy in equipment is limited
so that circuits cannot ignite an
atmosphere by sparking or heating

Measurement and
control equipment,
e.g. sensors,
instrumentation

Encapsulation

m 50028 - 18

Category 2
Zone 1

Electrical components embedded in
approved material to prevent contact
with explosive atmosphere

Measurement and
control devices,
solenoid valves

Type of
protection n

nA 50021 - 15

Category 3
Zone 2

Non-arcing and non-sparking

AC mot

ors, terminal

boxes, light fittings

Note: Group II Dust atmospher es are covered by CEN ELEC EN 50281-1-1 and EN50281-1-2

Fig 1.4.7: Standards and methods of protection

42

Section 1.4
Motors

corresponds to the running performance under the worst
conditions allowed for that specific motor. A motor for
use in zone 22 has to be IP55 protected, that is protected
against dust. The IP protection is the only difference
between 2D/category 2 equipment and 3D/category 3
equipment.

Mounting (International Mounting - IM)

Three different ways of mounting the motor exist: Foot­mounted motor, flange-mounted motor with free-hole
flange (FF) and flange-mounted motor with tapped-hole
flange (FT). Figure 1.4.8 shows the different ways of
mounting a motor and the standards that apply for the
mountings. The mounting of motors is stated according to
the following standards:

• IEC 60034-7, Code I,

i.e. designation IM followed by the previously
used DIN 42590 code

• IEC 60034-7, Code II

Enclosure class
(Ingress Protection - IP)

The enclosure class states the degrees of protection of
the motor against ingress of solid objects and water.
The enclosure class is stated by means of two letters IP
followed by two digits, for example IP55. The first digit
stands for protection against contact and ingress of solid
objects and the second digit stands for protection against
ingress of water, see figure 1.4.9.

Drain holes enable the escape of water which has entered
the stator housing for instance through condensation.
When the motor is installed in a damp environment, the
bottom drain hole should be opened. Opening the drain
hole changes the motor’s enclosure class from IP55 to
IP44.

IM B3

IM 1001

IM B5

IM 3001

IM V1

IM 3011

IM B14

IM 3601

IM V18

IM 3611

First digit Second digit

Protec

tion against contact and

ingress of solid objects

Protection against

ingress of water

0 No special protection

1 The motor is protected against

solid objects bigger than
55 mm, e.g. a hand

2 The motor is protected against
objects bigger than 12 mm, e.g.
a finger

3 The motor is protected against

solid objects bigger than 25 mm,

i.e. wires, tools, etc.

4 The motor is protected against
solid objects bigger than 1 mm,
e.g. wires

5 The motor is protected against
the ingress of dust

6 The motor is completely
dust-proof

0 No special protection

1 The motor is protected against

v

ertically falling drops of water,

s

uch as condensed water

2 The motor is protected against
v

ertically falling drops of water,

e

ven

if the motor is tilted at an

angle of 15°

3 The motor is protected against
water spray falling at an angle
of 60

°

from vertical

4 The motor is protected against
water splashing from any
direction

5 The motor is protected against
water being projected from a
nozzle from any direction

6 The motor is protected against
heavy

seas or high-pressure

water jets from any direction

7 The motor is protected when
submerged from 15 cm to 1 m in
water for a period specified by
the manufacturer

8 The motor is protected against
continuous submersion in water
under conditions specified by
the manufacturer

Fig 1.4.8: Different mounting methods

Foot-mounted
motor

Flange-mounted
motor with
free-hole flange

Flange-mounted
motor with
tapped-hole flange

Fig 1.4.9: The enclosure class is stated by means of two digits IP
followed by two letters; for example IP55

IM B35

IM 2001

43

Frame size

Figure 1.4.11 gives an overview of the relation between
frame size, shaft end, motor power and flange type and
size. For motors in frame sizes 63 up to and including 315M,
the relationship is specified in EN 50347. For motors with
frame size 315L and larger, no standard covers this relation.
The figure shows where on the motor the different values
that make up the frame size are measured.

Flanges and shaft end comply with EN 50347 and
IEC 60072-1. Some pumps have a coupling, which requires
a smooth motor shaft end or a special shaft extension
which is not defined in the standards.

Insulation class

The insulation class is defined in the IEC 60085 standard
and tells something about how robust the insulation
system is to temperatures. The life of an insulation
material is highly dependent on the temperature to which
it is exposed. The various insulation materials and systems
are classified into insulation classes depending on their
ability to resist high temperatures.

IEC 100L (In this case L = 140mm)

100mm

Distance between
holes

B3

Maximum temperature increase

Hot-spot overtemperature

Maximum ambient

temperature

10

80 105 125

40

B

[˚C] 180

155

130
120

40

F H

40 40

10

15

Class

B

F

H

Maximum ambient

temper

ature

(˚C)

40

40

40

Maximum

temper

ature increase

(K)

80

105

125

Hot-spot

overtemperature

(K)

10

10

15

Maximum

winding temperature

(Tmax)

(˚C)

130

155

180

Fig 1.4.12:

Different insulation classes and their temperature increase at

nominal voltage and load

Fig 1.4.10: Frame size

140mm

44

Section 1.4
Motors

1

Frame si

ze

4-pole 6-pole 8-pol

e

Free-hole

flange

Tapped-hole

flange

[kW] [kW] [kW] (FF) (FT)

0.06;0.09 FF100 FT65

0.12 ; 0.18 FF115 FT75

0.25; 0.37 FF130 FT85

0.55; 0.7

5

0.37; 0.55

FF165 FT100

1.1 0.75 0.37 FF165 FT115

1.5 1.1 0.55 FF165 FT115

2.2; 3 1.5 0.75; 1.1 FF215 FT13

0

4 2.2 1.5 FF215 FT130

5.5

3 2.2 FF265 FT165

7.5 4; 5.5 3 FF265 FT165

11 7.5 4; 5.5 FF300 FT215

15 11 7.5 FF300 FT215

18.5 - - FF30

0

22 15 11 FF30

0

30

18.5; 22

15 FF350

37 30 18.5 FF400

45 - 2

2 FF400

55 37 30 FF500

75 45 37 FF500

90 55 45 FF500

110 75 55 FF600

132 90 7

5 FF600

FF600

315; 355; 400; 450; 500

FF740

560; 630; 710

FF840

56

63

71

80

90S

90L

100L

112M

132S

132M

160M

160L

180M

180L

200L

225S

225M

250M

280S

280M

315S

315M

315L

355

400

450

2-pole

[mm]

9

11

14

19

24

24

28

28

38

38

42

42

48

48

55

55

55

60

65

65

65

65

65

75

80

90

4-, 6-, 8-pole

[mm]

9

11

14

19

24

24

28

28

38

38

42

42

48

48

55

60

60

65

75

75

80

80

80

10

0

10

0

12

0

2-pole

[kW]

0.09; 0.12

0.18; 0.2

5

0.37; 0.55

0.75; 1.1

1.5

2.2

3

4

5.5; 7.5

-

11; 1

5

18.5

22

-

30; 37

-

45

55

75

90

110

13

2

160; 200; 250

315; 355; 400; 450; 500

560; 630; 710

800; 900; 1000 800; 900; 1000

FF940

4

Flange size

Shaft end diameter Rated power

32

Fig 1.4.11: The relation between frame size and power input

45

Starting method

Direct-on-line starting (DOL) Simple and cost-efficient.

Safe starting.

High locked-rotor current.

Current pulses when switching over from star to delta.
Not suitable if the load has a low inertia.
Reduced locked-rotor torque.

Star/delta starting (SD)
(Y/∆)

Reduction of starting current by factor 3.

Autotransformer starting Reduction of locked-rotor current and torque. Current pulses when switching from reduced to full voltage.

Reduced locked-rotor torque.

Soft starter "Soft" starting. No current pulses.

Less water hammer when starting a pump.
Reduction of locked-rotor current as required,
typically 2-3 times.

Reduced locked-rotor torque.

Frequency converter starting No current pulses.

Less water hammer when starting a pump.
Reduction of locked-rotor current as required,
typically 2 to 3 times.
Can be used for continuous feeding of the motor.

Reduced locked-rotor torque.
Expensive

Pros Cons

Direct-on-line starting (DOL)

As the name suggests, direct-on-line starting means that
the motor is started by connecting it directly to the supply
at rated voltage. Direct-on-line starting is suitable for stable
supplies and mechanically stiff and well-dimensioned
shaft systems, for example pumps. Whenever applying
the direct-on-line

starting method, it is important to consult

local authorities.

Star/delta starting

The objective of this starting method, which is used with
three-phase induction motors, is to reduce the starting
current. In one position, current supply to the stator
windings is connected in star (Y) for starting. In other
positions, current supply is reconnected to the windings in
delta (∆) once the motor has gained speed.

Autotransformer starting

As the name states, autotransformer starting makes use
of an autotransformer. The autotransformer is placed in
series with the motor during start and varies the voltage
up to nominal voltage in two to four steps.

Soft starter

A soft starter is, as you would expect, a device which
ensures a soft start of a motor. This is done by raising the
voltage to a preset voltage raise time.

Frequency converter starting

Frequency converters are designed for continuous feeding
of motors, but they can also be used for soft starting.

1.4.2 Motor start-up

We distinguish between different ways of starting up
the motor: Direct-on-line starting, star/delta starting,
autotransformer starting, soft starter and frequency
converter starting. Each of these methods have their pros
and cons, see figure 1.4.13.

Fig 1.4.13:

Starting method

46

Section 1.4
Motors

1.4.3 Voltage supply

The motor’s rated voltage lies within a certain voltage
range. Figure 1.4.14 shows typical voltage examples for 50
Hz and 60 Hz motors.

According to the international standard IEC 60038, the
motor has to be able to operate with a main voltage
tolerance of ± 10%.

For motors that are designed according to the IEC 60034­1 standard with a wide voltage range, e.g. 380-415 V, the
main voltage may have a tolerance of ± 5%.

The permissible maximum temperature for the actual
insulation class is not exceeded when the motor is
operated within the rated voltage range. For conditions
at the extreme boundaries the temperature typically rises
approx. 10 Kelvin.

1.4.4 Frequency converter

Frequency converters are often used for speed controlling
pumps, see chapter 4. The frequency converter converts
the mains voltage into a new voltage and frequency,
causing the motor to run at a different speed. This way of
regulating the frequency might result in some problems:

• Acoustic noise from the motor, which is sometimes

transmitted to the system as disturbing noise

• High voltage peaks on the output from the frequency

converter to the motor

50 Hz 60 Hz

-

-

-

460 V

+_

10%

Mains voltage according to IEC 60038

230 V

+_

10%

400 V

+_

10%

690 V

+_

10%

-

Typical voltage examples

50 Hz

50 Hz motors come with the following voltages:

• 3 x 220 – 240 ∆/ 380 – 415 Y

• 3 x 200 – 220 ∆ / 346 – 380 Y

• 3 x 200 ∆/ 346 Y

• 3 x 380 – 415 ∆

• 1 x 220 – 230 / 240

60 Hz
60 Hz motors come with the following voltages:

• 3 x 200 – 230 ∆ / 346 – 400 Y

• 3 x 220 – 255 ∆ / 380 – 440 Y

• 3 x 220 – 277 ∆ / 380 – 480 Y

• 3 x 200 – 230 ∆ / 346 – 400 Y

• 3 x 380 – 480 ∆

Fig 1.4.14:

Typical voltages

Fig 1.4.15:

Mains voltage according to IEC 60038

47

Insulation for motors with frequency
converter

In connection with motors with frequency converters we
distinguish between different kinds of motors, with different
kinds of insulation.

Motors without phase insulation

For motors constructed without the use of phase
insulation, continuous voltages (RMS) above 460 V
can increase the risk of disruptive discharges in the
windings and thus destruction of the motor. This applies
to all motors constructed according to these principles.
Continuous operation with voltage peaks above 650 V can
cause damage to the motor.

Motors with phase insulation

In three-phase motors, phase insulation is normally
used and consequently, specific precautions are not
necessary if the voltage supply is smaller than 500 V.

Motors with reinforced insulation

In connection with supply voltages between 500 and
690 V, the motor has to have reinforced insulation or be
protected with delta U /delta t filters. For supply voltages
of 690 V and higher the motor has to be fitted with both
reinforced insulation and delta U /delta t filters.

Motors with insulated bearings

In order to avoid harmful current flows through the
bearings, the motor bearings have to be electrically

insulated.

This applies for motors from frame size 280 and up.

Fig 1.4.16: Stator with phase insulation

Phase insulation also referred

to as phase paper

48

Section 1.4
Motors

Motor efficiency

Generally speaking, electric motors are quite efficient.
Some motors have electricity-to-shaft power efficiencies
of 80-93% depending on the motor size and sometimes
even higher for bigger motors. Two types of energy losses
in electric motors exist: Load dependent losses and load
independent losses.

Load dependent losses vary with the square of the current
and cover:

• Stator winding losses (copper losses)

• Rotor losses (slip losses)

• Stray losses (in different parts of the motor)

Load independent losses in the motor refer to:

• Iron losses (core losses)

• Mechanical losses (friction)

Different motor classifications categorise motors according
to efficiency. The most important are CEMEP in the EU (EFF1,
EFF2 and EFF3) and EPAct in the US.

Motors can fail because of overload for a longer period of
time and therefore most motors are intentionally oversized
and only operate at 75% to 80% of their full load capacity.
At this level of loading, motor efficiency and power factor
remain relatively high. But when the motor load is less
than 25%, the efficiency and the power factor decrease.

The motor efficiency drops quickly below a certain
percentage of the rated load. Therefore, it is important to
size the motor so that the losses associated with running
the motor too far below its rated capacity are minimised.
It is common to choose a pump motor that meets the
power requirements of the pump.

1.4.5 Motor protection

Motors are nearly always protected against reaching temperatures,
which can damage the insulation system. Depending on
the construction of the motor and the application, thermal
protection can also have other functions, e.g. prevent damaging
temperatures in the frequency converter if it is mounted on
the motor.

The type of thermal protection varies with the motor type.
The construction of the motor together with the power
consumption must be taken into consideration when choosing
thermal protection. Generally speaking, motors have to be
protected against the following conditions:

Errors causing slow temperature increases in
the windings:

• Slow overload

• Long start-up periods

• Reduced cooling / lack of cooling

• Increased ambient temperature

• Frequent starts and stops

• Frequency fluctuation

• Voltage fluctuation

Errors causing fast temperature increases in
the windings:

• Blocked rotor

• Phase failure

Percent of rated load

Percent

0 25 50 75 150125100

100

20

40

60

80

1

0.2

0.4

0.6

0.8

Efficiency
Power factor

Cos ϕ

Percent of rated load

0 25 50 7

5 150125 175100

100

20

10

0

30

50

90

80

75 kW

7.5 kW

0.75 kW

70

60

40

Efficiency %

Fig 1.4.17: Efficiency vs. load
power factor vs. load
(schematic drawing)

Fig 1.4.18: The relation
between efficiency and
rated load of different
sized motors (schematic
drawing)

49

Thermal protection (TP)

According to the IEC 60034-11 standard, the thermal
protection of the motor has to be indicated on the
nameplate with a TP designation. Figure 1.4.19 shows an
overview of the TP designations.

PTC thermistors

PTC thermistors (Positive Temperature Coefficient
Thermistors) can be fitted into the windings of a motor
during production or retrofitted afterwards. Usually 3
PTCs are fitted in series; 1 in each phase of the winding.
They can be purchased with trip temperatures ranging
from 90°C to 180°C in 5 degrees steps. PTCs have to be
connected to a thermistor relay, which detects the rapid
increase in resistance of the thermistor when it reaches its
trip temperature. These devices are non-linear. At ambient
temperatures the resistance of a set of 3 will be about 200­300 ohms and this will increase rapidly when the thermistor
reaches its trip temperature. If the temperature increases
any further, the PTC thermistor can reach several thousand
ohms. The thermistor relays are usually set to trip at 3000
ohms or are preset to trip according to what the DIN 44082
standard prescribes. The TP designation for PTCs for motors
smaller than 11kW is TP211 if the PTCs are fitted into the
windings. If the PTCs are retrofitted the TP designation is
TP111. The TP designation for PTCs for motors larger than
11kW is normally TP111.

Indication of the permissible temperature level when the motor is exposed to thermal
overload. Category 2 allows higher temperatures than category 1 does.

Technical overload with

va

riation (1 digit)

Only slow

(i.e. constant

overload)

Only fast

(i.e. blocked condition)

Slow and fast

(i.e. constant overload

and blocked condition )

2 levels at emergency

signal and cutoff

1 level at cutoff

2 levels at emergency

signal and cutoff

1 level at cutoff

1 level at cutoff

Number af levels and

function area (2 digits)

Symbol

TP 111
TP 112
TP 121
TP 122
TP 211
TP 212
TP 221
TP 222
TP 311
TP 312

Category 1

(3 digits)

1
2
1
2

1
2

1
2
1
2

Fig 1.4.19: TP designations

Thermal switch and thermostats

Thermal switches are small bi-metallic switches that switch
due to the temperature. They are available with a wide range
of trip temperatures; normally open and closed types. The
most common type is the closed one. One or two, in series,
are usually fitted in the windings like thermistors and can be
connected directly to the circuit of the main contactor coil.
In that way no relay is necessary. This type of protection is
cheaper than thermistors, but on the other hand, it is less
sensitive and is not able to detect a locked rotor failure.

Thermal switches are also referred to as Thermik, Klixon
switches and PTO (Protection Thermique à Ouverture).
Thermal switches always carry a TP111 designation.

Single-phase motors

Single-phase motors normally come with incorporated
thermal protection. Thermal protection usually has an
automatic reclosing. This implies that the motor has to
be connected to the mains in a way that ensures that
accidents caused by the automatic reclosing are avoided.

Three-phase motors

Three-phase motors have to be protected according to local
regulations. This kind of motor has usually incorporated
contacts for resetting in the external control circuit.

50

51

Standstill heating

A heating element ensures the standstill heating of
the motor. The heating element is especially used in
connection with applications that struggle with humidity
and condensation. By using the standstill heating, the
motor is warmer than the surroundings and thereby the
relative air humidity inside the motor is always lower than
100%.

Maintenance

The motor should be checked at regular intervals. It is
important to keep the motor clean in order to ensure
adequate ventilation. If the pump is installed in a dusty
environment, the pump must be cleaned and checked
regularly.

Bearings

Normally, motors have a locked bearing in the drive end and
a bearing with axial play in the non-drive end. Axial play is
required due to production tolerances, thermal expansion
during operation, etc. The motor bearings are held in place
by wave spring washers in the non-drive end, see figure

1.4.21.

The fixed bearing in the drive end can be either a deep­groove ball bearing or an angular contact bearing.

Bearing clearances and tolerances are stated according
to ISO 15 and ISO 492. Because bearing manufacturers
have to fulfil these standards, bearings are internationally
interchangeable.

In order to rotate freely, a ball bearing must have a certain
internal clearance between the raceway and the balls.
Without this internal clearance, the bearings can either
be difficult to rotate or it may even seize up and be unable
to rotate. On the other hand, too much internal clearance
will result in an unstable bearing that may generate
excessive noise or allow the shaft to wobble.

Depending on which pump type the motor is fitted, the
deep-groove ball bearing in the drive end must have C3
or C4 clearance. Bearings with C4 clearance are less heat
sensitive and have increased axial load-carrying capacity.

The bearing carrying the axial forces of the pump can have
C3 clearance if:

• the pump has complete or partial hydraulic relief

• the pump has many brief periods of operation

• the pump has long idle periods

C4 bearings are used for pumps with fluctuating high axial
forces. Angular contact bearings are used if the pump exerts
strong one-way axial forces.

Non-drive end

Drive end

Non-driv

e end bearingSpring washer Drive end bearing

Fig 1.4.21: Cross-sectional drawing of motor

1.4.20: Stator with heating element

52

Motors with permanently lubricated bearings

For closed permanently lubricated bearings, use one of the
following high temperature resistant types of grease:

• Lithium-based grease

• Polyurea-based grease

The technical specifications must correspond to the
standard

DIN - 51825 K2 or better. The basic oil viscosity

must be higher than:

• 50 cSt (10

-6m2

/sec) at 40°C and

• 8 cSt (mm

2

/sec) at 100°C

For example Klüberquiet BQH 72-102 with a grease
filling ratio of: 30 - 40%.

Motors with lubrication system

Normally, frame size 160 motors and upwards have
lubricating nipples for the bearings both in the drive end
and the non-drive end.

The lubricating nipples are visible and easily accessible.
The motor is designed in such a way that:

• there is a flow of grease around the bearing

• new grease enters the bearing

• old grease is removed from the bearing

Motors with lubricating systems are supplied with a
lubricating instruction, for instance as a label on the
fan cover. Apart from that, instructions are given in the
installation and operating instructions.

The lubricant is often lithium-based, high temperature
grease, for instance EXXON UNIREX N3 or Shell Alvania
Grease G3. The basic oil viscosity must be

• higher than 50 cSt (10-6m2/sec) at 40°C and

• 8 cSt (mm2/sec) at 100°C

Moderate to strong forces.
Primarily outward pull on
the shaft end

Fixed deep-groove ball bearing (

C4)

Strong outward pull
on the shaft end

Small forces
(flexible coupling)

Strong inward
pressure

Axial forces Bearing types and recommended clearance

Drive-end Non-drive-end

Moderate forces.
Primarily outward pull on
the shaft end (partly
hydraulically relieved in
the pump)

Deep-groove ball bearing (C4)

Fixed deep-groove ball bearing (

C3)

Fixed deep-groove ball bearing (

C3)

Fixed angular contact bearing

Deep-groove ball bearing (C3)

Fixed angular contact bearing

Deep-groove ball bearing (C3)

Deep-groove ball bearing (C3)

Deep-groove ball bearing (C3)

Fig:1.4.22: Typical types of bearings in pump motors

Section 1.4
Motors

Chapter 1. Design of pumps and motors

Section 1.5: Liquids

1.5.1 Viscous liquids

1.5.2 Non-Newtonian liquids

1.5.3 The impact of viscous liquids on the
performance of a centrifugal pump

1.5.4 Selecting the right pump for a liquid
with antifreeze

1.5.5 Calculation example

1.5.6 Computer aided pump selection for
dense and viscous liquids

Section 1.5
Liquids

1.5.1 Viscous liquids

There is no doubt about it; water is the most common
liquid that pumps handle. However, in a number of
applications, pumps have to handle other types of liquids,
e.g. oil, propylene glycol, gasoline. Compared to water,
these types of liquids have different density and viscosity.

Viscosity is a measure of the thickness of the liquid.

The higher the viscosity, the thicker the liquid. Propylene
glycol and motor oil are examples of thick or high viscous
liquids. Gasoline and water are examples of thin, low
viscous liquids.

Two kinds of viscosity exist:

• The dynamic viscosity (µ), which is normally measured
in Pa⋅s or Poise. (1 Poise = 0.1 Pa⋅s)

• The kinematic viscosity (ν), which is normally measured
in centiStokes or m2/s (1 cSt = 10-6 m2/s)

The relation between the dynamic viscosity (µ) and the
kinematic viscosity (ν) is shown in the formula on your
right hand side.

On the following pages, we will only focus on kinematic
viscosity (ν).

The viscosity of a liquid changes considerably with the
change in temperature; hot oil is thinner than cold oil. As
you can tell from figure 1.5.1, a 50% propylene glycol liquid
increases its viscosity 10 times when the temperature
changes from +20 to –20 oC.
For more information concerning liquid viscosity, go to
appendix L.

54

ν =

µ
ρ

ρ = density of liquid

Kinematic

viscosity

ν

[cSt]

Density

ρ

[kg/m3]

Liquid

temperature

t [˚C]

Liquid

Water 20

998 1.004
Gasoline 20 733 0.75
Olive oil

20 900 93

50% Propylene glycol

20 1043 6.4

50% Propylene glycol

-20 1061 68.7

Fig. 1.5.1: Comparison of viscosity values for water and a few other liquids.
Density values and temperatures are also shown

1.5.2 Non-Newtonian liquids

The liquids discussed so far are referred to as Newtonian
fluids. The viscosity of Newtonian liquids is not affected by
the magnitude and the motion that they are exposed to.
Mineral oil and water are typical examples of this type of
liquid. On the other hand, the viscosity of non-Newtonian
liquids does change when agitated.

This calls for a few examples:

• Dilatant liquids like cream – the viscosity increases
when agitated

• Plastic fluids like catsup – have a yield value, which has
to be exceeded before flow starts. From that point on,
the viscosity decreases with an increase in agitation

• Thixotrophic liquids like non-drip paint - exhibit a
decreasing viscosity with an increase in agitation

The non-Newtonian liquids are not covered by the viscosity
formula described earlier in this section.

1.5.3 The impact of viscous liquids on the
performance of a centrifugal pump

Viscous liquids, that is liquids with higher viscosity and/
or higher density than water, affect the performance of
centrifugal pumps in different ways:

• Power consumption increases, i.e. a larger motor may
be required to perform the same task

• Head, flow rate and pump efficiency are reduced

Let us have a look at an example. A pump is used for
pumping a liquid in a cooling system with a liquid
temperature below 0oC. To avoid that the liquid freezes,
an antifreeze agent like propylene glycol is added to
the water. When glycol or a similar antifreeze agent is
added to the pumped liquid, the liquid obtains properties,
different from those of water. The liquid will have:

• Lower freezing point, tf [°C]

• Lower specific heat, cp [kJ/kg.K]

• Lower thermal conductivity, λ [W/m.K]

• Higher boiling point, tb [°C]

• Higher coefficient of expansion, β [m/°C]

• Higher density, ρ [kg/m3]

• Higher kinematic viscosity, ν [cSt]

These properties have to be kept in mind when designing
a system and selecting pumps. As mentioned earlier,
the higher density requires increased motor power and
the higher viscosity reduces pump head, flow rate and
efficiency resulting in a need for increased motor power,
see figure 1.5.2.

Q

H, P, η

H

P

η

Fig. 1.5.2: Changed head, efficiency and power input for
liquid with higher viscosity

55

1.5.4 Selecting the right pump for a
liquid with antifreeze

Pump characteristics are usually based on water at around
20°C, i.e. a kinematic viscosity of approximately 1 cSt and
a density of approximately 1,000 kg/m³.
When pumps are used for liquids containing antifreeze
below 0°C, it is necessary to examine whether the pump
can supply the required performance or whether a
larger motor is required. The following section presents
a simplified method used to determine pump curve
corrections for pumps in systems that have to handle a
viscosity between 5 - 100 cSt and a density of maximum
1,300 kg/m³. Please notice that this method is not as
precise as the computer aided method described later in
this section.

Pump curve corrections for pumps handling high
viscous liquid

Based on knowledge about required duty point, QS, HS, and
kinematic viscosity of the pumped liquid, the correction
factors of H and P2 can be found, see figure 1.5.3.

Fig. 1.5.3: It is possible to determine the correction factor for head and
power consumption at different flow, head and viscosity values

10

20

30

40

50

60

70

80

90

100

110

120

130

140

0.

9

1.0

1.1

1.2

1.3

1.4

1.5

1.6

1.7

1.8

1.9

1.00

1.05

1.20

1.15

1.10

1.25

1.30

1.35

0

H = 6

m

H = 10

m

H = 20

m

H = 40

m

H = 60

m

10 cS

t

20 cS

t

40 cS

t

60 cSt

100 cSt

5 cS

t

10 cSt

20 cSt

40 cSt

60 cSt

100 cSt

5 cSt

K

H

K

P2

Q [m3/h]

56

Section 1.5
Liquids

Figure 1.5.3 is read in the following way:

When kH and kP2 are found in the figure, the equivalent
head for clean water HW and the corrected actual shaft
power P2S can be calculated by the following formula

where
HW : is the equivalent head of the pump if the pumped
liquid is “clean” water

P

2W

: is the shaft power at the duty point (QS,HW) when

the pumped liquid is water

HS : is the desired head of the pumped liquid
(with agents)

P2S : is the shaft power at the duty point (Qs,Hs) when
the pumped liquid is water (with agents)

ρs : is the density of the pumped liquid

ρw : is the density of water = 998 kg/m

3

The pump selection is based on the normal data sheets/
curves applying to water. The pump should cover the duty
point Q,H = QS,HW, and the motor should be powerful
enough to handle P2S on the shaft.

Figure 1.5.4 shows how to proceed when selecting a pump
and testing whether the motor is within the power range
allowed.

ρ

P

2S

= K

P2

.

P

2w

.

( )

s

w

Water

Water

Mixture

Mixture

H

w

H

w

= k

H

.

H

S

2

1

H

H

s

P

2s

P

P

2w

Q

s

Q

Q

5

3

4

ρ

Fig. 1.5.4: Pump curve correction when choosing the right pump for
the system

The pump and motor selecting procedure contains the
following steps:

• Calculate the corrected head Hw
(based on HS and kH), see figure 1.5.4 1-2

• Choose a pump capable of providing performance
according to the corrected duty point (QS, HW)

• Read the power input P2W in the duty point (QS,Hw),
see figure 1.5.4 3-4

• Based on P2W , kP2 , ρW , and ρS calculate the
corrected required shaft power P2S , see figure 1.5.4 4-5

• Check if P2S < P

2 MAX

of the motor. If that is the case
the motor can be used. Otherwise select a more
powerful motor

HW = k

H . HS

ρ

s

ρ

w

P2S = k

P2 . P2w

.

( )

57

1.5.5 Calculation example

A circulator pump in a refrigeration system is to pump a
40% (weight) propylene glycol liquid at –10°C. The desired
flow is QS = 60 m3/h, and the desired head is HS = 12 m.
Knowing the required duty point, it is possible to find the QH­characteristic for water and choose a pump, which is able to
cover the duty point. Once we have determined the needed
pump type and size we can check if the pump is fitted with a
motor, which can handle the specific pump load.

The liquid has kinematic viscosity of 20 cSt and a density
of 1049 kg/m3. With QS = 60 m3/h, HS = 12 m and ν = 20 cSt,
the correction factors can be found in figure 1.5.3.

kH = 1.03

kP2 = 1.15

HW = kH · HS = 1.03 · 12 = 12.4 m

QS = 60 m3/h

The pump has to be able to cover a duty point equivalent
to Q,H = 60 m3/h, 12.4m. Once the necessary pump size is
determined, the P2 value for the duty point is found, which
in this case is P2W = 2.9 kW. Now it is possible to calculate
the required motor power for propylene glycol mixture:

The calculation shows, that the pump has to be fitted with
a 4 kW motor, which is the smallest motor size able to
cover the calculated P2S = 3.5 kW.

1.5.6 Computer aided pump selection for
dense and viscous liquids

Some computer aided pump selection tools include a
feature that compensates for the pump performance
curves based on input of the liquid density and viscosity.
Figure 1.5.5 shows the pump performance curves from the
example we just went through.

The figure shows both the performance curves for the
pump when it handles viscous liquid (the full lines) and
the performance curves when it handles water (the broken
lines). As indicated head, flow and efficiency are reduced,
resulting in an increase in power consumption.
The value of P2 is 3.4 kW, which corresponds to the result
we got in the calculation example in section 1.5.4.

H

[m]

η

[%]

0

1

2

3

4

0

2

4

6

8

10

12

14

0 10 2

0

0

10

20

30

40

50

60

70

30 40 50 60 70 80

Q [m3/h]

Q [m3/h]

P

2

[kW]

Fig. 1.5.5: Pump performance curves

ρ

S

ρ

w

P2S = k

P2 . P2w

.

P

2S

= 1.15

.

2.9

.

1049

998

= 3.5 kW

58

Section 1.5
Liquids

Chapter 1. Design of pumps and motors

Section 1.6: Materials

1.6.1 What is corrosion?

1.6.2 Types of corrosion

1.6.3 Metal and metal alloys

1.6.4 Ceramics

1.6.5 Plastics

1.6.6 Rubber

1.6.7 Coatings

60

Section 1.6
Materials

In this section you can read about different materials that
are used for pump construction. Our main focus will be
on the features that every single metal and metal alloy
have to offer. But before we dig any further into the
world of materials, we will have a closer look at corrosion.
Besides explaining what corrosion is, we will examine the
different types of corrosion and what can be done to
prevent corrosion from occurring.

1.6.1 What is corrosion?

Corrosion is usually referred to as the degradation of the
metal by chemical or electrochemical reaction with its
environment, see figure 1.6.1. When considered broadly,
corrosion may be looked upon as the tendency of the
metal to revert to its natural state similar to the oxide
from which it was originally melted. Only precious metals,
such as gold and platinum are found in nature in their
metallic state.

Some metals produce a tight protective oxide layer on the
surface, which hinders further corrosion. If the surface
layer is broken it is self-healing. These metals are passivated.
Under atmospheric conditions the corrosion products
of zinc and aluminium form a fairly tight layer and
further corrosion is prevented.
Likewise, on the surface of stainless steel a tight layer of
iron and chromium oxide is formed and on the surface of
titanium a layer of titanium oxide is formed. The protective
layer of these metals explains their good corrosion
resistance. Rust, on the other hand, is a non-protective
corrosion product on steel.
Rust is porous, not firmly adherent and does not prevent
continued corrosion, see figure 1.6.2.

pH (acidity)

Oxidizing agents (such as oxygen)
Temperature
Concentration of solution constituents

(such as chlorides)

Biological activity

Operating conditions
(such as velocity, cleaning procedures and shutdowns)

Environmental variables that affect the
corrosion resistance of metals and alloys

Fig. 1.6.1: Environmental variables that affect the corrosion
resistance of metals and alloys

Non-protective corrosion product

Protective corrosion product

Fig. 1.6.2: Examples of corrosion products

Rust on steel

Oxide layer on stainless steel

1.6.2 Types of corrosion

Generally, metallic corrosion involves the loss of metal at
a spot on an exposed surface. Corrosion occurs in various
forms ranging from uniform attacks over the entire surface
to severe local attacks.
The environment’s chemical and physical conditions
determine both the type and the rate of corrosion attacks. The
conditions also determine the type of corrosion products that
are formed and the control measures that need to be taken. In
many cases, it is impossible or rather expensive to completely
stop the corrosion process; however, it is usually possible to
control the process to acceptable levels.

On the following pages we will go through the different
forms of corrosion in order to give you an idea of their
characteristics.

Uniform corrosion

Uniform or general corrosion is characterised by corrosive
attacks proceeding evenly over the entire surface, or on a
large part of the total area. General thinning continues until
the metal is broken down. Uniform corrosion is the type of
corrosion where the largest amount of metal is wasted.

Examples of metals, which are subject to uniform
corrosion:

• Steel in aerated water

• Stainless steel in reducing acids (such as EN 1.4301
(AISI 304) in sulfuric acid)

Pitting corrosion

Pitting corrosion is a localised form of corrosive attacks.
Pitting corrosion forms holes or pits on the metal surface.
It perforates the metal while the total corrosion, measured
by weight loss, might be rather minimal. The rate of
penetration may be 10 to 100 times that of general
corrosion depending on the aggressiveness of the liquid.
Pitting occurs more easily in a stagnant environment.

Example of metal that is subject to pitting corrosion:

• Stainless steel in seawater

Fig. 1.6.3: Uniform corrosion

Fig. 1.6.4: Pitting corrosion

61

1. Design of pumps and motors

1.1 Pump construction, (10)

Crevice corrosion

Crevice corrosion - like pitting corrosion - is a localised
form of corrosion attack. However, crevice corrosion starts
more easily than pitting. Crevice corrosion occurs at narrow
openings or spaces between two metal surfaces or between
metals and non-metal surfaces and is usually associated
with a stagnate condition in the crevice. Crevices, such as
those found at flange joints or at threaded connections,
are thus often the most critical spots for corrosion.

Example of metal that is subject to crevice corrosion:

• Stainless steel in seawater

Intergranular corrosion

As the name implies, intergranular corrosion occurs at
grain boundaries. Intergranular corrosion is also called
intercrystalline corrosion. Typically, this type of corrosion
occurs when chromium carbide precipitates at the grain
boundaries during the welding process or in connection
with insufficient heat treatment. A narrow region around
the grain boundary may therefore deplete in chromium
and become less

corrosion resistant than the rest of the
material. This is unfortunate because chromium plays an
important role in corrosion resistance.

Examples of metals that are subject to intergranular corrosion:

• Stainless steel - which is insufficiently welded or
heat-treated

• Stainless steel EN 1.4401 (AISI 316) in concentrated
nitric acid

Selective corrosion

Selective corrosion is a type of corrosion which attacks
one single element of an alloy and dissolves the element
in the alloy structure. Consequently, the alloy’s structure
is weakened.

Examples of selective corrosion:

• The dezincification of unstabilised brass, whereby a
weakened, porous copper structure is produced

• Graphitisation of gray cast iron, whereby a brittle
graphite skeleton is left because of the dissolution
of iron

Fig. 1.6.5: Crevice corrosion

Fig. 1.6.6: Intergranular corrosion

Fig. 1.6.7: Selective corrosion

Copper

Zinc corrosion products

Brass

62

Section 1.6
Materials

Erosion corrosion

Erosion corrosion is a process that involves corrosion
and erosion. The rate of corrosion attack is accelerated
by the relative motion of a corrosive liquid and a
metal surface. The attack is localised in areas with
high velocity or turbulent flow. Erosion corrosion
attacks are characterised by grooves with directional
pattern.

Examples of metals which are subject to erosion corrosion:

• Bronze in seawater

• Copper in water

Cavitation corrosion

A pumped liquid with high velocity reduces the pressure.
When the pressure drops below the liquid vapour
pressure, vapour bubbles form (the liquid boils).
In the areas where the vapour bubbles form, the
liquid is boiling. When the pressure raises again, the
vapour bubbles collapse and produce intensive
shockwaves. Consequently, the collapse of the vapour
bubbles remove metal or oxide from the surface.

Examples of metals that are subject to cavitation:

• Cast iron in water at high temperature

• Bronze in seawater

Stress corrosion cracking (SCC)

Stress corrosion cracking (SCC) refers to the combined
influence of tensile stress (applied or internal) and
corrosive environment. The material can crack without
any significant deformation or obvious deterioration of
the material. Often, pitting corrosion is associated with
the stress corrosion cracking phenomena.

Examples of metals that are subject to stress corrosion
cracking:

• Stainless steel EN 1.4401 (AISI 316) in chlorides

• Brass in ammonia

Fig. 1.6.8: Erosion corrosion

Fig. 1.6.9: Cavitation corrosion

Fig. 1.6.10: Stress corrosion cracking

Flow

63

1. Design of pumps and motors

1.1 Pump construction, (10)

<

Corrosion fatigue

Pure mechanical fatigue is when a material subjected
to a cyclic load far below the ultimate tensile strength
can fail. If the metal is simultaneously exposed to a
corrosive environment, the failure can take place at an
even lower stress and after a shorter time. Contrary to
a pure mechanical fatigue, there is no fatigue limit in
corrosion-assisted fatigue.

Example of metals that are subject to corrosion fatigue:

• Aluminium structures in corrosive atmosphere

Galvanic corrosion

When a corrosive electrolyte and two metallic materials
are in contact (galvanic cell), corrosion increases on the
least noble material (the anode) and decreases on the
noblest (the cathode). The increase in corrosion is called
galvanic corrosion. The tendency of a metal or an alloy
to corrode in a galvanic cell is determined by its position
in the galvanic series. The galvanic series indicates the
relative nobility of different metals and alloys in a given
environment (e.g. seawater, see figure 1.6.12).
The farther apart the metals are in the galvanic series,
the greater the galvanic corrosion effect will be. Metals or
alloys at the upper end are noble, while those at the lower
end are least noble.

Examples of metal that are subject to galvanic corrosion:

• Steel in contact with 1.4401

• Aluminium in contact with copper

The principles of galvanic corrosion are used in cathodic
protection. Cathodic protection is a means of reducing
or preventing the corrosion of a metal surface by the
use of sacrificial anodes (zinc or aluminum) or impressed
currents.

Fig. 1.6.11: Corrosion fatigue

Fig. 1.6.12: Galvanic corrosion

Fig. 1.6.13: Galvanic series for metals and alloys in seawater

Aluminium - less noble Copper - most noble

64

Section 1.6
Materials

1.6.3 Metal and metal alloys

On the following pages, you can read about the features
of different metals and metal alloys, used for construction
of pumps.

Ferrous alloys

Ferrous alloys are alloys where iron is the prime constituent.
Ferrous alloys are the most common of all materials
because of their availability, low cost, and versatility.

Steel

Steel is a widely used material primarily composed of iron
alloyed with carbon. The amount of carbon in steel varies
in the range from 0.003% to 1.5% by weight. The content of
carbon has an important impact on the material’s strength,
weldability, machinability, ductility, and hardness. As a
rule-of-thumb, an increase in carbon content will lead to
an increase in strength and hardness but to a decrease in
ductility and weldability. The most common type of steel is
carbon steel. Carbon steel is grouped into four categories,
see figure 1.6.14.

Steel is available in wrought as well as in cast condition.
The general characteristics of steel castings are closely
comparable to those of wrought steels. The most obvious
advantage of steel is that it is relatively inexpensive to make,
form and process. On the other hand, the disadvantage
of steel is that its corrosion resistance is low compared to
alternative materials, such as stainless steel.

Cavitation corrosion of bronze impeller

Erosion corrosion of cast iron impeller

Pitting corrosion of EN 1.4401 (AISI 316)

Intergranular corrosion of
stainless steel

Crevice corrosion of
EN 1.4462 (SAF 2205)

1 mm

Type of steel Content of carbon

Low carbon or mild steel 0.003% to 0.30% of carbon

Medium carbon steel 0.30% to 0.45% of carbon

High carbon steel 0.45% to 0.75% of carbon

Very high carbon steel 0.75% to 1.50% of carbon

Fig 1.6.14: Four types of carbon steel

65

Nodular (ductile) iron

Nodular iron contains around 0.03-0.05% by weight of
magnesium. Magnesium causes the flakes to become
globular so the graphite is dispersed throughout a ferrite
or pearlite matrix in the form of spheres or nodules. The
graphite nodules have no sharp features. The round shape
of nodular graphite reduces the stress concentration and
consequently, the material is much more ductile than grey
iron. Figure 1.6.16 clearly shows that the tensile strength is
higher for nodular iron than is the case for grey iron. Nodular
iron is normally used for pump parts with high strength
requirements (high pressure or high temperature
applications).

Stainless steel

Stainless steel is chromium containing steel alloys. The
minimum chromium content in standardised stainless
steel is 10.5%. Chromium improves the corrosion resistance
of stainless steel. The higher corrosion resistance is due
to a chromium oxide film that is formed on the metal
surface. This extremely thin layer is self-repairing under
the right conditions.
Molybdenum, nickel and nitrogen are other examples of
typical alloying elements. Alloying with these elements
brings out different crystal structures, which enable
different properties in connection with machining, forming,
welding, corrosion resistance, etc. In general, stainless
steel has a higher resistance to chemicals (i.e. acids) than
steel and cast iron have.

Cast iron

Cast iron can be considered an alloy of iron, silicon
and carbon. Typically, the concentration of carbon is
between 3-4% by weight, most of which is present
in insoluble form (e.g. graphite flakes or nodules).
The two main types are grey cast iron and nodular
(ductile) cast iron. The corrosion resistance of cast iron is
comparable to the one for steel; and sometimes even better.
Cast iron can be alloyed with 13-16% by weight silicon or
15-35% by weight nickel (Ni-resist) respectively in order to
improve corrosion resistance. Various types of cast irons are
widely used in industry, especially for valves, pumps, pipes
and automotive parts. Cast iron has good corrosion resistance
to neutral and alkaline liquids (high pH) . But its resistance to
acids (low pH) is poor.

Grey iron

In grey iron, the graphite is dispersed throughout a ferrite
or pearlite matrix in the form of flakes. Fracture surfaces
take on a grey appearance (hence the name!). The graphite
flakes act as stress concentrators under tensile loads,
making it weak and brittle in tension, but strong
and ductile in compression. Grey iron is used for the
construction of motor blocks because of its high vibration
damping ability. Grey iron is an inexpensive material and
is relatively easy to cast with a minimal risk of shrinkage.
That is why grey iron is often

used for pump parts with

moderate strength requirements.

1. Design of pumps and motors

1.1 Pump construction, (10)

Fig 1.6.15: Comparison and designations of grey iron

Fig 1.6.16: Comparison and designations of nodular iron

66

Section 1.6
Materials

In environments containing chlorides, stainless steel can
be attacked by localised corrosion, e.g. pitting corrosion
and crevice corrosion. The resistance of stainless steel to
these types of corrosion is highly dependent on its
chemical composition. It has become rather common to
use the so-called PRE (Pitting Resistance Equivalent)
values as a measure of pitting resistance for stainless
steel. PRE values are calculated by formulas in which the
relative influence of a few alloying elements (chromium,

Chemical composition of stainless steel [w%]

Microstructure Designation % % % % % PRE

5)

EN/AISI/UNS Carbon max. Chromium Nickel Molybdenum Other

Ferritic

1.4016/430/ S43000 0.08 16-18 17

Martensitic 1.4057/431/ S43100 0.12-0.22 15-17 1.5-2.5 16

Austenitic 1.4305/303/ S30300 0.1 17-19 8-10 S 0.15-0.35 18

Austenitic 1.4301/304/ S30400 0.07 17-19.5 8-10.5 18

Austenitic 1.4306/304L/ S30403 0.03 18-20 10-12 18

Austenitic 1.4401/316/ S31600 0.07 16.5-18.5 10-13 2-2.5 24

Austenitic 1.4404/316L/ S31603 0.03 16.5-18.5 10-13 2-2.5 24

Austenitic 1.4571/316Ti/ 0.08 16.5-18.5 10.5-13.5 2-2.5 Ti > 5 x carbon 24
S31635 Ti < 0.70

Austenitic 1.4539/904L/ N08904 0.02 19-21 24-26 4-5 Cu 1.2-2 34

Austenitic 1.4547/none / 0.02 20 18 6.1 N 0.18-0.22 43
S 31254

3)

Cu 0.5-1

Ferritic/

1.4462/ none/ 0.03 21-23 4.5-6.5 2.5-3.5 N 0.10-0.22 34

austenitic S32205 2)

Ferritic/

1.4410/none/ 0.03 25 7 4 N 0.24-0.32 43

austenitic S 32750

4)

Microstructure Designation % % % % % PRE
EN/ASTM/UNS Carbon max. Chromium Nickel Molybdenum Other

Austenitic 1) 1.4308/CF8/ J92600 0.07 18-20 8-11 19

Austenitic 1) 1.4408/CF8M/ J92900 0.07 18-20 9-12 2-2.5 26

Austenitic 1) 1.4409/CF3M/ J92800 0.03 18-20 9-12 2-2.5 N max. 0.2 26

Austenitic 1.4584/none/ none 0.025 19-21 24-26 4-5

N max. 0.

2

35

Cu 1-3

Ferritic/
Austenitic 1.4470/CD3MN/ J92205 0.03 21-23 4.5-6.5 2.5-3.5 N 0.12-0.2 35

Ferritic/ 1.4517/CD4MCuN/ N 0.12-0.22
Austenitic J93372 0.03 24.5-26.5 2.5-3.5 2.5-3.5 Cu 2.75-3.5 38

molybdenum and nitrogen) on the pitting resistance is
taken into consideration. The higher the PRE, the higher
the resistance to localised corrosion. Be aware that the PRE
value is a very rough estimate of the pitting resistance of
a stainless steel and should only be used for comparison/
classification of different types of stainless steel.
In the following, we will present the four major types of
stainless steel: ferritic, martensitic, austenitic and duplex.

1)

Contains some ferrite 2) Also known as SAF 2205, 3) Also known as 254 SMO, 4) Also known as SAF 2507

5)

Pitting Resistance Equivalent (PRE): Cr% + 3.3xMo% + 16xN%.

Fig 1.6.17: Chemical composition of stainless steel

67

Ferritic (magnetic)

Ferritic stainless steel is characterised by quite good
corrosion properties, very good resistance to stress
corrosion cracking and moderate toughness. Low alloyed
ferritic stainless steel is used in mild environments
(teaspoons, kitchen sinks, washing machine drums,
etc.) where it is a requirement that the component is
maintenance-free and non-rusting.

Martensitic (magnetic)

Martensitic stainless steel is characterised by high strength
and limited corrosion resistance. Martensitic steels are
used for springs, shafts, surgical instruments and for
sharp-edged tools, such as knives and scissors.

Austenitic (non-magnetic)

Austenitic stainless steel is the most common type
of stainless steel and is characterised by a high
corrosion resistance, very good formability, toughness
and weldability. Austenitic stainless steel, especially
the EN 1.4301 and EN 1.4401 are used for almost any
type of pump components in the industry. This kind of
stainless steel can be either wrought or cast.

EN 1.4305 is one of the most popular stainless steel types
of all the free machining stainless steel types. Due to its
high sulphur content (0.15-0.35 w%), the machinability has
improved considerably but unfortunately at the expense
of its corrosion resistance and its weldability. However,
over the years free machining grades with a low sulphur
content and thus a higher corrosion resistance have been
developed.

If stainless steel is heated up to 500°C - 800°C for a
longer period of time during welding, the chromium
might form chromium carbides with the carbon present in
the steel. This reduces chromium’s capability to maintain
the passive film and might lead to intergranular

corrosion
also referred to as sensitisation (see section 1.6.2).
If low carbon grades of stainless steel are used the risk of
sensitisation is reduced. Stainless steel with a low content

1. Design of pumps and motors

1.1 Pump construction, (10)

of carbon is referred to as EN 1.4306 (AISI 304L) or EN

1.4404 (AISI 316L). Both grades contain 0.03% of carbon
compared to 0.07% in the regular type of stainless steel
EN 1.4301 (AISI 304) and EN 1.4401 (AISI 316), see
illustration 1.6.17.

The stabilised grades EN 1.4571 (AISI 316Ti) contain a small
amount of titanium. Because titanium has a higher affinity
for carbon than chromium, the formation of chromium
carbides is minimised. The content of carbon is generally
low in modern stainless steel, and with the easy availability
of ‘L’ grades the use of stabilised grades has declined
markedly.

Ferritic-austenitic or duplex (magnetic)

Ferritic-austenitic (duplex) stainless steel is characterised
by high strength, good toughness, high corrosion resistance
and excellent resistance to stress corrosion cracking and
corrosion fatigue in particular.
Ferritic-austenitic stainless steel is typically used in
applications that require high strength, high corrosion
resistance and low susceptibility to stress corrosion cracking
or a combination of these properties. Stainless steel
EN 1.4462

is widely used for making pump shafts and pump

housings.

68

Section 1.6
Materials

Nickel alloys

Nickel base alloys are defined as alloys in which nickel
is present in greater proportion than any other alloying
element. The most important alloying constituents are
iron, chromium, copper, and molybdenum. The alloying
constituents make it possible to form a wide range of
alloy classes. Nickel and nickel alloys have the ability to
withstand a wide variety of severe operating conditions,
for instance corrosive environments, high temperatures,
high stresses or a combination of these factors.

HastelloysTM alloys are a line of commercial alloys containing
Ni, Mo, Cr, and Fe. Nickel base alloys, such as InconelTM
Alloy 625, HastelloysTM C-276 and C-22 are highly corrosion
resistant and not subject to pitting or crevice corrosion in
low velocity seawater, and do not suffer from erosion at
high velocity.

The price of nickel base alloy limits its use in certain
applications. Nickel alloys are available in both wrought
and cast grades. However, nickel alloys are more difficult
to cast than the common carbon steels and stainless steel
alloys. Nickel alloys are especially used for pump parts in
the chemical process industry.

Copper alloys

Pure copper has excellent thermal and electrical properties,
but is a very soft and ductile material.
Alloying additions result in different cast and wrought
materials, which are suitable for use in the production of
pumps, pipelines, fittings, pressure vessels and for many
marine, electrical and general engineering applications.

Brasses are the most widely used of the copper alloys
because of their low cost, their easy or inexpensive
fabrication and machining. However, they are inferior in
strength to bronzes and must not be used in environments
that cause dezincification (see selective corrosion).
Red brass, bronze and copper nickels in particular have,
compared to cast iron a high resistance to chlorides in
aggressive liquids, such as seawater. In such environments,
brass is unsuitable because of its tendency to dezincificate.
All copper alloys have poor resistance to alkaline liquids
(high pH), ammonia and sulfides and are sensitive to
erosion. Brass, red brass and bronze are widely used for
making bearings, impellers and pump housings.

1) Lead can be added as an alloying element to improve
the machinability.

2) Bronze can be alloyed with aluminium to increase strength.

Fig 1.6.18: Common types of copper alloys

69

Titanium

Pure titanium has a low density, is quite ductile and has a
relatively low strength. However, when a limited amount
of oxygen is added it will strengthen titanium and produce
the so-called commercial-pure grades. Additions of various
alloying elements, such as aluminium and vanadium
increase its strength significantly, at the expense of
ductility. The aluminium and vanadium alloyed titanium
(Ti-6Al-4V) is the “workhorse” alloy of the titanium
industry. It is used in many aerospace engine and airframe
components. Because titanium is a high-price material, it
is not yet a material which is often used for making pump
components.

Titanium is a very reactive material. As it is the case for
stainless steel, titanium’s corrosion resistance depends on
the formation of an oxide film. However, the oxide film
is more protective than that on stainless steel. Therefore,
titanium performs much better than stainless steel in
aggressive liquids, such as seawater, wet chlorine or
organic chlorides, that cause pitting and crevice corrosion.

1. Design of pumps and motors

1.1 Pump construction, (10)

Aluminium

Pure aluminium is a light and soft metal with a density
of about a third of that of steel. Pure aluminium has a
high electrical and thermal conductivity. The most common
alloying elements are silicon (silumin), magnesium, iron and
copper. Silicon increases the material’s castability, copper
increases its machinability and magnesium increases its
corrosion resistance and strength.

The obvious advantages of aluminium are that the material
naturally generates a protective oxide film and is highly
corrosion resistant if it is exposed to the atmosphere.
Treatment, such as anodising, can further improve

this
property. Aluminium alloys are widely used in structures
where a high strength to weight ratio is important, such
as in the transportation industry. For example, the use of
aluminium in vehicles and aircrafts reduces weight and
energy consumption.

On the other hand, the disadvantage of aluminium is that
it is not stable at low or high pH and in chloride-containing
environments. This property makes aluminium unsuitable for
exposure to aqueous solutions especially under conditions
with high flow. This is further emphasised by the fact that
aluminium is a reactive metal, i.e. has a low position in the
galvanic series (see galvanic corrosion) and may easily suffer
from galvanic corrosion if coupled to nobler metals and
alloys.

Designation Major alloying element

1000-series Unalloyed (pure) >99% Al

2000-series

Copper is the principal alloying element, though other

elements (magnesium) may be specified

3000-series Manganese is the principal alloying element

4000-series

Silicon is the principal alloying element

5000-series

Magnesium is the principal alloying element

6000-series

Magnesium and silicon are principal alloying elements

7000-series

Zinc is the principal alloying element, but other elements,
such as copper, magnesium, chromium, and zirconium
may be specified

8000-series

Other elements (including tin and some lithium

compositions)

CP: commercial pure (titanium content above 99.5%)

Fig 1.6.19: Major alloying elements of aluminium

Fig 1.6.20: Titanium grades and alloy characteristics

70

Section 1.6
Materials

1.6.4 Ceramics

Ceramic materials are defined as inorganic, non-metallic
materials, which are typically crystalline in nature. They
are composed of metallic and non-metallic elements.
Common technical ceramics are aluminium oxide (alumina

- Al2O3), silicon carbide (SiC), tungsten carbide (WC), and
silicon nitride (Si3N4).

Ceramics are suitable for applications which require high
thermal stability, high strength, high wear resistance, and
high corrosion resistance. The disadvantage of ceramics is
the low ductility and high tendency for brittle fractures.
Ceramics are mainly used for making bearings and seal
faces for shaft seals.

1.6.5 Plastics

Some plastics are derived from natural substances, such as
plants, but most types are man-made. These are known as
synthetic plastics. Most synthetic plastics come from crude
oil, but coal and natural gas are also used.
There are two main types of plastics: Thermoplastics and
thermosets (thermosetting plastics). The thermoplastics
are the most common kind of plastic used worldwide.
Plastics often contain additives, which transfer certain
additional properties to the material. Furthermore,
plastics can be reinforced with fiberglass or other fibres.
These plastics together with additives and fibres are also
referred to as composites.

Examples of additives found in plastics

• Inorganic fillers for mechanical reinforcement

• Chemical stabilisers, e.g. antioxidants

• Plasticisers

• Flame retardants

Thermoplastics

Thermoplastic polymers consist of long polymer molecules
that are not linked to each other, i.e. have no cross-links.
They are often supplied as granules and heated to permit
fabrication by methods, such as moulding or extrusion.
A wide range is available, from low-cost commodity plastics
(e.g. PE, PP, PVC) to high cost engineering thermoplastics
(e.g. PEEK) and chemical resistant fluoropolymers (e.g.
PTFE, PVDF). PTFE is one of the few thermoplastics, which
is not melt-processable. Thermoplastics are widely

used
for making pump housings or for lining of pipes and pump
housings.

Thermosets

Thermosets harden permanently when heated, as cross­linking hinders bending and rotations. Cross-linking is
achieved during fabrication using chemicals, heat, or
radiation; this process is called curing or vulcanization.
Thermosets are harder, more dimensionally stable, and
more brittle than thermoplastics and cannot be remelted.
Important thermosets include epoxies, polyesters,
and polyurethanes. Thermosets are among other things
used for surface coatings.

PP
PE
PVC
PEEK
PVDF
PTFE*

Abbreviation

Polymer name

Polypropylene
Polyethylene
Polyvinylchloride
Polyetheretherketone
Polyvinylidene fluoride
Polytetrafluoroethylen

e

*

Trade name: Teflon®

Linear polymer chains

Thermoplastics

Elastomers

Thermosets

Branched polymer chains

Weakly cross-linked polymer chains

Strongly cross-linked polymer chains

Fig 1.6.22: Different types of polymers

Fig 1.6.21: Overview of polymer names

71

1. Design of pumps and motors

1.1 Pump construction, (10)

1.6.6 Rubber

The term rubber includes both natural rubber and
synthetic rubber. Rubbers (or elastomers) are flexible long­chain polymers that can be stretched easily to several times
their unstretched length and which rapidly return to their
original dimensions when the applied stress is released.
Rubbers are cross-linked (vulcanized), but have a low cross­link density, see figure 1.6.22. The cross-link is the key to
the elastic, or rubbery, properties of these materials. The
elasticity provides resiliency in sealing applications. Different
components in a pump are made of rubber, e.g. gaskets
and O-rings (see section 1.3 on shaft seals). In this section we
will present the different kinds of rubber qualities and their
main properties as regards temperature and resistance to
different kinds of liquid groups.

At temperatures up to about 100°C nitrile rubber is an
inexpensive material that has a high resistance to oil and
fuel. Different grades exist - the higher the acrylonitrile
(ACN) content, the higher the oil resistance, but the poorer
is the low temperature flexibility. Nitrile rubbers have high
resilience and high wear resistance but only moderate
strength. Further, the rubber has limited weathering
resistance and poor solvent resistance. It can generally be
used down to about -30°C, but certain grades can operate
at lower temperatures.

Ethylene propylene has excellent water resistance which
is maintained to high temperatures approximately 120­140°C. The rubber type has good resistance to acids, strong
alkalis and highly polar fluids, such as methanol and
acetone. However, it has very poor resistance to mineral
oil and fuel.

Fluoroelastomers (FKM)

Fluoroelastomers cover a whole family of rubbers designed
to withstand oil, fuel and a wide range of chemicals
including non-polar solvents. offers excellent

resistance to high temperature operation (up to 200°C

depending on the grade) in air and different types of oil.

rubbers have limited resistance to steam, hot water,
methanol, and other highly polar fluids. Further, this type
of rubber has poor resistance to amines, strong alkalis
and many freons. There are standard and special grades

- the latter have special properties, such as improved low­temperature or chemical resistance.

Silicone rubbers have outstanding properties, such as low
compression set in a wide range of temperatures (from

-60°C to 200°C in air), excellent electrical insulation and
are non-toxic. Silicone rubbers are resistant to water,
some acids and oxidizing chemicals. Concentrated acids,
alkalines, and solvents should not be used with silicone
rubbers. In general, these types of rubber have poor
resistance to oil and fuel. However, the resistance of FMQ
silicone rubber to oil and fuel is better than that of the
silicone rubber types MQ, VMQ, and PMQ.

Perfluoroelastomers have very high chemical resistance,
almost comparable to that of PTFE (polytetrafluorethylene,
e.g. TeflonR). They can be used up to high temperatures,
but their disadvantages are difficult processing, very high
cost and limited use at low temperatures.

Common types of copper alloys

NBR

Abbreviation

Nitrile rubber

EPDM, EPM

Ethylene propylene rubber

FKM

Fluoroelastomers Viton

R

Siloprene

R

Buna-N

R

FFKM

Pe

rfluoroelastomers

Chemra

z

R

Kalrez

R

MQ, VMQ, PMQ, FMQ

Silicone rubber

Common nam

e

Examples of
trade name

Nordel

R

Fig 1.6.23: Rubber types

72

Section 1.6
Materials

1.6.7 Coatings

Protective coating – metallic, non-metallic (inorganic)
or organic – is a common method of corrosion control. The
main function of coatings is (aside from galvanic coatings,
such as zinc) to provide an effective barrier between the
metal (substrate) and its environment. They allow the use
of normal steel or aluminium instead of more expensive
materials. In the following section we will examine
the possibilities of preventing corrosion by means of
different coatings:
Metallic and non-metallic (inorganic) coatings and organic
coatings.

Metallic coatings

Metallic coatings less noble than the substrate

Zinc coatings are commonly used for the protection of
steel structures against atmospheric corrosion. Zinc has
two functions: it acts as a barrier coating and it provides
galvanic protection. Should an exposed area of

steel
occur, the zinc surface preferentially corrodes at a slow
rate and protects the steel. The preferential protection
is referred to as cathodic protection. When damage is
small, the protective corrosion products of zinc will fill the
exposed area and stop the attack.

Metallic coatings nobler than the substrate

Electroplating of nickel and chromium coatings on steel
are examples of metallic coatings that are nobler than
the substrate. Unlike galvanic coatings where the coating
corrodes near areas where the base metal is exposed,
any void or damage in a barrier coating can lead to an
immediate base metal attack.

To protect the base steel,

zinc coating sacrifices itself

slowly by galvanic ac

tion.

Steel coated with a more noble

metal, such as nickel, corrodes

more rapidly if the coating

is damaged.

Fig 1.6.24: Galvanic vs. barrier corrosion protection

73

Paints

As mentioned above, paints are an important class of
organic coating. Figure 1.6.25 shows several types of
organic coatings. A typical paint formulation contains
polymeric binders, solvents, pigments and additives. For
environmental reasons, organic solvents are increasingly
being replaced by water or simply eliminated, e.g powder
coating. Painted metal structures usually involve two or
more layers of coating applied on a primary coating, which
is in direct contact with the metal.

1. Design of pumps and motors

1.1 Pump construction, (10)

<

Non-metallic coatings
(inorganic coatings)

Conversion coatings are an important category of non­metallic coatings (inorganic).

Conversion coatings

Conversion coatings are formed by a

controlled corrosion

reaction of the substrate in an oxidised

solution.
Well-known examples of conversion coatings are anodising
or chromating of aluminium, and phosphate treatment of
steel. Anodising is mainly used for surface protection of
aluminium, while chromating and phosphating are usually
used for pre-treatment in connection with painting.
Besides improving paint adhesion, it helps to prevent the
spreading of rust under layers of paint.

Organic coatings

Organic coatings contain organic compounds and are
available in a wide range of different types. Organic coatings
are applied to the metal by methods of spraying, dipping,
brushing, lining or electro-coating (paint applied by means
of electric current) and they may or may not require
heat-curing. Both thermoplastic coatings, such as
polyamide,

polypropylene, polyethylene, PVDF and PTFE and

elastomer

coatings are applied to metal substrates to
combine the mechanical properties of metal with the
chemical resistance of plastics but paints are by far the
most widely used organic coating.

Physical states of common organic coatings

Resin Solvent- Water- Powder Two comp.
type

based based coating liquid

Ac

rylic X X X

Alky

d X X

Epox

y X X X X
Polyester X X X
Polyurethane X X X X
Vi

nyl X X X

Fig 1.6.25: Physical states of common organic coatings

74

Section 1.6
Materials

Chapter 2. Installation and performance reading

Section 2.1: Pump installation

2.1.1 New installation

2.1.2 Existing installation

2.1.3 Pipe flow for single-pump installation

2.1.4 Limitation of noise and vibrations

2.1.5 Sound level (L)

Section 2.2: Pump performance

2.2.1 Hydraulic terms

2.2.2 Electrical terms

2.2.3 Liquid properties

Correct advice and selection of pump type for an installation
has larger implication than what meets the eye. The
larger the pumps, the greater the costs with respect
to investment, installation, commissioning, running
and maintenance – basically the life cycle cost (LCC). An
extensive product portfolio combined with competent
advice and after-sales service is the foundation of a correct
selection. The following analysis, recommendations and
hints are general for any installation, but to a greater
extent relevant for medium-sized and large installations.
We will present our recommendations for two types of
installation: New and existing installations.

2.1.1 New installation

• If the pipework has not been planned yet, you can base

the selection of pump type on other primary selection
criteria, e.g. efficiency, investment costs or life cycle
costs (LCC). This will not be covered in this
section. However, the general guidelines, which are
presented, also apply for pipework that has not yet
been planned.

• If the pipework has already been planned, the selection

of pump is equivalent to replacing a pump in an existing

installation.

76

2.1.2 Existing installation – replacement

The following five steps will help you make an optimum
pump selection for an existing installation:

Pre-investigation of the installation should include
the following considerations:

• Basic pipe flow – pipes in and out of the building, e.g.

from ground, along floor or from ceiling

• Specific pipework at the point of installation, e.g.

in-line or end-suction, dimensions, manifolds

• Space available – width, depth and height

• Accessibility in connection with installation and

maintenance, for instance doorways

• Availability of lifting equipment or alternatively

accessibility of such equipment

• Floor type, e.g. solid or suspended floor with basement

• Existing foundation

• Existing electric installation

Previous pump installation

• Pump make, type, specifications including old duty

point, shaft seal, materials, gaskets, controlling

• History, e.g. lifetime, maintenance

Future requirements

• Desired improvements and benefits

• New selection criteria including duty points and

operating times, temperature, pressure, liquid
specifications

• Supplier criteria, e.g. availability of spare parts

Advisory

• Major changes might be beneficial in a long or short

term or both and must be documented, e.g. installation

savings, life cycle costs (LCC), reduction on environmental
impact like noise and vibrations and accessibility in
connection with maintenance

Selection

• Must be based on a customer-agreed list of priorities

For the selection of the correct pump type and advice on
installation, two main areas are important: Pipe flow and
limitation of noise and vibrations. These two areas will
be dealt with on the following pages.

Section 2.1
Pump installation

2.1.3 Pipe flow for single-pump installation

Figure 2.1.1 is based on single-pump installation. In parallel installations accessibility plays a
major role for how good a pump choice is.

The evaluation criterion is as simple pipework as possible, hence as few bends as possible.

Pipework

To the pump:

Along floor

Best choice

Best choice

Best choice

Best choice

Best choice

Best choice

Best choice

Good choice

Good choice

Good choice

Good choice

Good choice

Good choice

Good choice

Good choiceGood choice

Good choiceGood choice

Least good choice Least good choice

Least good choice

Least good choice

Good choice

Best choice

Best choice

Best choice Best choice

Best choice

Best choice

Not applicable

From ground

A. In-line close-coupled

(horizontal or vertical
mounting)

From ceiling

Wall­mounted

From the pump:

Along floor

Along floor

To ground

To ceiling

To ground

To

ceiling

Wall­mounted

Along floor

To

ceiling

Pump type

To ground

C. End-suction long-coupled

(only horizontal mounting)

B. End-suction close- coupled

(horizontal or vertical
mounting)

Scores:

Best choice

Good choice

Least good choice

Not applicable

Fig. 2.1.1 Pipework and pump type

77

Accessibility plays a major role in how well a specific pump
choice is in connection with installation of several pumps
in parallel. The accessibility is not always easy for in-line
pumps installed in parallel because of the pipework, see
figure 2.1.2. As it appears from figure 2.1.3, end-suction
pumps installed in parallel provide easier accessibility.

2.1.4 Limitation of noise and vibrations

To achieve optimum operation and minimise noise and
vibration, it may be necessary to consider vibration
dampening of the pump in certain cases. Generally, this
should always be considered in the case of pumps with
motors above 7.5 kW. Smaller motor sizes, however, may
also cause undesirable noise and vibration. Noise and
vibration are generated by the rotation in motor and
pump and by the flow in pipes and fittings. The effect
on the environment depends on correct installation and
the state of the entire system. Below we will present 3
different ways of limiting noise and vibration in a pump
installation: Foundation, dampeners and expansion joint.

Foundation

Floor constructions can be divided into two types: Solid
floor and suspended floor.

Solid – minimum risk of noise due to bad
transmission of vibrations, see figure 2.1.4.

Suspended – risk of floor amplifying the noise.
Basement can act as a resonance box,
see figure 2.1.5.

The pump should be installed on a plane and rigid surface.
Four basic ways of installation exist for the two types
of floor construction: Floor, plinth, floating plinth and
foundation suspended on vibration dampeners.

Fig. 2.1.3:
3 end-suction pumps in parallel; easier maintenace access
because of pipework

Fig. 2.1.4: Solid floor construction

Fig. 2.1.5: Suspended floor construction

Fig. 2.1.2:
3 in-line pumps in parallel; limited maintenance
access because of pipework

Floor

Solid ground

Floor

Wall

Ground floor

Basement

Floor

Solid ground

78

Section 2.1
Pump installation

Floor
Direct mounting on floor, hence direct vibration
transmission, see figure 2.1.6.

Plinth
Poured directly on concrete floor, hence as floor, see
figure 2.1.7.

Floating plinth
Resting on a dead material, e.g. sand, hence
reduced risk of vibration transmission, see figure 2.1.8.

Foundation suspended on vibration dampeners
Optimum solution with controlled vibration
transmission, see figure 2.1.9.

As a rule of thumb, the weight of a concrete foundation
should be 1.5 x the pump weight. This weight is needed to
get the dampeners to work efficiently at low pump speed.

Fig. 2.1.6: Floor

Floor Base plate Pump unit

Fig. 2.1.10: The same
foundation rules go
for vertical in-line
pumps

Fig. 2.1.7: Plinth

Floor Plinth Base plate Pump unit

Fig. 2.1.8:
Floating plinth

Floor Sand Plinth Base plate Pump unit

Fig. 2.1.9: Foundation
suspended on
vibration dampeners

Floor

Vibration dampeners Foundation Base plate Pump unit

Pump unit

Foundation

Vibration
dampeners

Floor

79

Dampeners

The selection of the right vibration dampener requires
the following data:

• Forces acting on the dampener

• Motor speed considering speed control, if any

• Required dampening in % (suggested value is 70%)

The determination of the right dampener varies from
installation to installation but a wrong selection of
dampener may increase the vibration level in certain cases.
The supplier should therefore size vibration dampeners.

Pumps installed with vibration dampeners must always
have expansion joints fitted at both the suction and the
discharge side. This is important in order to avoid that the
pump hangs in the flanges.

Expansion joints

Expansion joints are installed to:

• absorb expansions/contractions in the pipework

caused by changing liquid temperature

• reduce mechanical strains in connection with pressure

waves in the pipework

• isolate mechanical noise in the pipework (not for metal

bellows expansion joints)

Expansion joints must not be installed to compensate for
inaccuracies in the pipework, such as centre displacement
or misalignment of flanges.

Expansion joints are fitted at a distance of minimum 1 to

1.5 . DN diameter from the pump on the suction side as well
as on the discharge side. This prevents the development of
turbulence in the expansion joints, resulting in better suction
conditions and a minimum pressure loss on the pressure
side. At high water velocities (>5 m/s) it is best to install
larger expansion joints corresponding to the pipework.

Fig. 2.1.11: Installation with expansion joints, vibration dampeners and
fixed pipework

Base plate

Pump unit

Vibration
dampeners

Floor

Expansion
joint

Foundation

80

Section 2.1
Pump installation

Figures 2.1.12-2.1.14 show examples of rubber bellows
expansion joints with or without tie bars.

Expansion joints with tie bars can be used to minimise the
forces caused by the expansion joints. Expansion joints
with tie bars are recommended for sizes larger than DN 100.
An expansion joint without tie bars will exert force on
the pump flanges. These forces affect the pump and the
pipework.

The pipes must be fixed so that they do not stress the
expansion joints and the pump, see figure 2.1.11. The fix
points should always be placed as close to the expansion
joints as possible. Follow the expansion joint supplier’s
instructions.

At temperatures above 100°C combined with a high pressure,
metal bellows expansion joints are often preferred due to
the risk of rupture.

2.1.5 Sound level (L)

The sound level in a system is measured in decibel (dB).
Noise is unwanted sound. The level of noise can be
measured in the following three ways:

1. Pressure – Lp : The pressure of the air waves

2. Power – LW : The power of the sound

3. Intensity - LI: The power per m2 (will not be
covered in this book)

It is not possible to compare the three values directly, but it
is possible to calculate between them based on standards.
The rule of thumb is:

Fig. 2.1.14: Metal
bellows expansion
joints with tie bars

Fig. 2.1.12: Rubber bellows
expansion joints with tie bars

Fig. 2.1.13: Rubber
bellows expansion
joints without tie
bars

Smaller pumps, e.g. 1.5 kW: Lw = LP + 11 dB

Larger pumps, e.g. 110 kW: Lw = LP + 16 dB

120

100

80

60

40

20

20 50 100 200 1 2 5 10 20kHz500Hz

0

Frequency
kHz

Pain threshold

Lp (dB)

Threshold of hearing

Speech

Music

Fig. 2.1.15: Threshold of hearing vs. frequency

81

The EU Machine Directive prescribes that sound levels
have to be indicated as pressure when they are below 85
dB(A) and as power when they exceed 85 dB(A).

Noise is subjective and depends on a person´s ability
to hear, e.g. young vs. old person. Therefore, the above­mentioned measurements get weight according to
the sensibility of a standard ear, see figure 2.1.15. The
weighting is known as A-weighting (dB(A)), expressed as
e.g: LpA, and the measurements are adjusted depending on
frequency. In some cases it increases and in other cases it
decreases, see figure 2.1.16. Other weightings are known
as B and C but they are used for other purposes, which we
do not cover in this book.

In case of two or more pumps in operation, the sound level
can be calculated. If it is pumps with the same sound level
the total sound level can be calculated adding the value
from figure 2.1.17, e.g. 2 . pumps is Lp + 3 dB, 3 . pumps is
Lp + 5 dB. If the pumps have different sound level, values
from figure 2.1.18 can be added.

Indications of sound level should normally be stated as
free field conditions over reflecting surface, meaning the
sound level on a hard floor with no walls. Guaranteeing
values in a specific room in a specific pipe system is
difficult because these values are beyond the reach of the
manufacturer. Certain conditions could have a negative
impact (increasing sound level) or a positive impact
on the sound level. Recommendations to installation
and foundation can be given to eliminate or reduce the
negative impact.

dB (A)

10

0

10

100 1000

-10

-20

-30

-40

-50

-60

-70

-80

10000 Hz

4 8 12 16 20 24

5

10

15

2 4 6 8 10

1

2

2.5

1.5

0.5

3

Fig. 2.1.16 A-weighting curve

Fig. 2.1.17 Increase of the total sound pressure level with
equal sources

Fig. 2.1.18 Increase of the total sound pressure level with
different sources

Experience values:

Rise of Perceived as:

+ 3dB Just noticeable
+ 5dB Clearly noticeable
+10dB Twice as loud

82

Section 2.1
Pump installation

Section 2.2
Pump performance

When you examine a pump, there are several things
you have to check. Besides checking the pump’s physical
condition, for instance if it is rusty or makes abnormal
noise, you have to know a number of values in order to
be able to tell if the pump performs as it is supposed
to. On the next pages, we will present three groups of
values you may need to focus on when you examine a
pump’s performance: Hydraulic terms, electrical terms,
mechanical terms and liquid properties.

2.2.1 Hydraulic terms

When you want to examine pump performance, there are
a number of values you need to know. In this section, we
will present the most important hydraulic terms: Flow,
pressure and head.

Flow

Flow is the amount of liquid that passes through a pump
within a certain period of time. When we deal with
performance reading, we distinguish between two flow
parameters: Volume flow and mass flow.

Volume flow (Q)

Volume flow is what we can read from a pump curve or
put in another way, a pump can move a volume per unit
of time (measured in m3/h) no matter the density of the
liquid. When we deal with e.g. water supply, volume flow
is the most important parameter, because we need the
pump to deliver a certain volume, e.g. of drinking water
or water for irrigation.Throughout this book the term flow
refers to volume flow.

Mass flow (Qm)

Mass flow is the mass, which a pump moves per unit of
time and is measured in kg/s. The liquid temperature has an
influence on how big a mass flow the pump can move per unit
of time since the liquid density changes with the temperature.
In connection with heating, cooling and air-condition systems,
the mass flow is an essential value to know, because the mass
is the carrier of energy (see Heat capacity).

Fig. 2.2.1: Calculation examples

Examples Unit

Water

Volume flow Q 10m3/h

Density 998 943kg/m

3

Mass flow Q

m

9980 9403kg/h

2.77 2.62kg/s

at 20˚C at 120˚C

Q

m

Q

m

ρ

. Q

ρ

Q

= =

;

83

Pressure (p)

Pressure is a measure of force per unit area. We
distinguish between static pressure, dynamic pressure
and total pressure. The total pressure is the sum of the
static pressure and the dynamic pressure:

Static pressure

The static pressure p

sta

is the pressure, which is measured
with a pressure gauge placed perpendicular to the flow or
in a non-moving liquid, see figure 2.2.2.

Dynamic pressure

The dynamic pressure p

dyn

is caused by liquid velocity.
Dynamic pressure cannot be measured by a normal pressure
gauge, but is calculated by the following formula:

1
2

where:
ρ is the density of the liquid in [kg/m3]
v is the velocity of the liquid in [m/s]

Dynamic pressure can be converted into static pressure
by reducing the liquid velocity and vice versa. Figure 2.2.3
shows a part of a system where the pipe diameter increases
from D1 to D2, resulting in a decrease in liquid speed from v1
to v2. Assuming that there is no friction loss in the system,
the sum of the static pressure and the dynamic pressure is
constant throughout the horizontal pipe.

1
2

1
2

So, an increase in pipe diameter, as the one shown in
figure 2.2.3 results in an increase in the static head which
is measured with the pressure gauge p2.

In most pumping systems, the dynamic pressure p

dyn

has
a minor impact on the total pressure. For example, if the
velocity of a water flow is 4.5 m/s, the dynamic pressure is
around 0.1 bar, which is considered insignificant in many
pumping systems. Later on in this chapter, we will discuss
dynamic pressure in connection with determining the
head of a pump.

Fig. 2.2.2: How to determine the static pressure p

sta

, the dynamic

pressure p

dyn

and the total pressure p

tot

Fig. 2.2.3: The static pressure increases if the liquid velocity is reduced.
The figure applies for a system with insignificant friction loss

D

2

D

1

p

sta

p

tot

p

dyn

A

P

B

p

1

p

2

v

1

v

2

p

sta

p

tot

p

dyn

p

tot

p

sta

p

sta

p

tot

Q

84

Section 2.2
Pump performance

10.2 m

10.6 m

12.75 m

2

4

6

8

10

12

H(m)

Q

Duty point for brine at 20˚C

Duty point for water at 20˚C

Duty point for water at 95˚C

Duty point for diesel at 20˚C

7.3 m

10.2 m

10.6 m

12.75 m

1 bar 1 bar 1 bar 1 bar

Brine at 20°C
1300 kg/m

3

1 bar = 7.3 m

Water at 20°C
997 kg/m

3

1 bar = 10.2 m

Water at 95°C
960 kg/m

3

1 bar = 10.6 m

Diesel oil at 20°C
800 kg/m

3

1 bar = 12.75 m

Fig. 2.2.5: Pumping four different liquids at 1 bar at the discharge side
of the pump results in four different heads (m), hence four different
duty points

Measuring pressure

Pressure is measured in e.g. Pa (N/m²), bar (105 Pa) or PSI
(Ib/in²). When we deal with pressure it is important to know
the point of reference for the pressure measurement. Two
types of pressure are essential in connection with pressure
measurement: Absolute pressure and gauge pressure:

Absolute pressure (p

abs

)

Absolute pressure is defined as the pressure above
absolute vacuum, 0 atm, that is the absolute zero for
pressure. Usually, the value “absolute pressure” is used in
cavitation calculations.

Gauge pressure

Gauge pressure, often referred to as overpressure, is the
pressure, which is higher than the normal atmospheric
pressure (1 atm). Normally, pressure p is stated as gauge
pressure, because most sensor and pressure gauge
measurements measure the pressure difference between
the system and the atmosphere. Throughout this book the
term pressure refers to gauge pressure.

Head (H)

The head of a pump is an expression of how high the pump
can lift a liquid. Head is measured in meter (m) and is
independent on the liquid density. The following formula
shows the relation between pressure (p) and head (H):

1
2

1
2

where :
H is the head in [m]
p is the pressure in [Pa = N/m2]
ρ is the liquid density in [kg/m3]
g is the acceleration of gravity in [m/s2]

Normally, pressure p is measured in [bar], which equals
10

5

Pa. However, other pressure units are used as well, see

figure 2.2.4.

The relation between pressure and head is shown in figure

2.2.5 where a pump handles four different liquids. The
head of the pump depends on the type of liquid. As it
appears from the figure, the pumping of different liquids
results in different heads and hence different duty points.

7.3 m

10.2 m

10.6 m

12.75 m

1 Pa = 1 N/m

2

10

-5

1 9.87 . 10

-4

750

.

10

-5

1.02

.

10

-4

750

1.02

.

10

-5

1

10

5

0.987 10.21.02

760

1.013

1.013 . 10

5

1 10.331.033

736

0.981

0.981 . 10

5

0.968 101

73.6

0.0981

0.981 . 10

4

* Physical atmosphere ** Theoretical atmosphere

0.0968 10.1

Pa bar

Conversion table for pressure units

atm* at** mH2O mmHg

1 bar

1 atm

1 at = 1 kp/cm

3

1 m H2O

2

4

6

8

10

12

H(m)

1 bar 1 bar 1 bar 1 bar

Q

Duty point for brine at 20˚C

Duty point for water at 20˚C

Duty point for water at 95˚C

Duty point for diesel at 20˚C

Fig. 2.2.4:

Conversion table for pressure units

85

Fig. 2.2.6: Standard end-suction pump with dimension difference on
suction and discharge ports

How to determine the head

The pump head is determined by reading the pressure on
the flanges of the pump p2, p1 and then convert the values
into head – see figure 2.2.6. However, if a geodetic difference
in head is present between the two measuring points, as it
is the case in figure 2.2.6, it is necessary to compensate for
the difference. Furthermore, if the port dimensions of the
two measuring points differ from one another the actual
head has to be corrected for this as well.

The actual pump head H is calculated by the following
formula:

1
2

1
2

where :
H is the actual pump head in [m]
p is the pressure at the flanges in [Pa = N/m2]
ρ is the liquid density in [kg/m3]
g is the acceleration of gravity in [m/s2]
h is the geodetic height in [m]
v is the liquid velocity in [m/s]

The liquid velocity v is calculated by the following
formula:

1
2

1
2

where:
v is the velocity in [m/s]
Q is the volume flow in [m3/s]
D is the port diameter in [m]

When combining these two formulas, head H depends
on the following factors: The pressure measurements
p1 and p2, the difference in geodetic height between the
measuring points (h2-h1), the flow through the pump Q
and the diameter of the two ports D1 and D2 .

1
2

1
2

The correction due to the difference in port diameter is
caused by the difference in the dynamic pressure. Instead of
calculating the correction from the formula, the contribution
can be read in a nomogram, see appendix F.

h2h

1

v

1

p

1D1

D

2

v

2

p

2

86

Section 2.2
Pump performance

Calculation example

A pump of the same type as the one shown in figure 2.2.7
is installed in a system with the following data:

Q = 240 m3/h
p1 = 0.5 bar
p2 = 1.1 bar
Liquid: Water at 200C

Suction port diameter D1 = 150 mm
Discharge port diameter D2 = 125 mm.
The difference in height between the two ports where the
pressure gauges are installed is h2-h1 = 355 mm.

We are now able to calculate the head of the pump:

As it appears from the calculation, the pressure difference
measured by pressure gauges is about 1.1 m lower than
what the pump is actually performing. This calls for an
explanation. First, the deviation is caused by the difference
in height between the pressure gauges (0.36 m) and
second, it is caused by the difference in port dimensions,
which in this case is 0.77 m.

Fig. 2.2.7: Standard end-suction pump with different
dimensions of suction and discharge ports (Example)

1
2

1
2

1
2

1
2

1
2

1
2

h2h

1

v

1

p

1D1

D

2

v

2

p

2

h2 - h1 = 355 mm

v1 = 3.77 m/s

2

p1 = 0.5 bar

D1 = 150 mm

D

2

= 125 mm

v

2

= 5.43 m/s

2

p

2

= 1.1 bar

87

Fig.2.2.8: The system pressure H

sta

in a closed system

has to be higher than the physical height of the installation

If the pressure gauges are placed at the same geodetic
height, or if a differential pressure gauge is used for the
measurement, it is not necessary to compensate for the
difference in height (h2-h1). In connection with in-line
pumps, where inlet and outlet are placed at the same
level, the two ports often have the same diameter. For
these types of pumps a simplified formula is used to
determine the head:

1
2

1
2

Differential pressure (∆p)

The differential pressure is the pressure difference
between the pressures measured at two points, e.g. the
pressure drops across valves in a system. Differential
pressure is measured in the same units as pressure.

System pressure

The system pressure is the static pressure, which is present
at a point in the system, when the pumps are not running.
System pressure is important to consider when you deal
with a closed system. The system pressure in (m) H

sta

in
the lowest point must always be higher than the height of
the system in order to ensure that the system is filled with
liquid and can be vented properly.

h

Dry cooler

Chiller

H

syst

> h

H

syst

88

Section 2.2
Pump performance

NPSH

Q

H

Fig.: 2.2.12: NPSH-curve

Cavitation and NPSH

Cavitation occurs somewhere in a pump when the pressure
is lower than the vapour pressure of the liquid, see figures

2.2.9 and 2.2.10.

When the pressure on the suction side drops below
the vapour pressure of the liquid, (figure 2.2.10 yellow
dot). small vapour bubbles form. These bubbles collapse
(implode) when the pressure rises (figure 2.2.10 red dot)
and releases shock waves. Consequently, impellers can
be damaged by the energy released. The rate of damage
to the impeller depends on the properties of the material.
Stainless steel is more resistant to cavitation than bronze,
and bronze is more resistant to cavitation than cast iron,
see section 1.6.3.

Cavitation decreases flow (Q) and head (H), which leads
to reduced pump performance, see figure 2.2.11. Damage
due to cavitation is often only detected when the pump
is dismantled. Furthermore, cavitation results in increased
noise and vibrations, which can consequently damage
bearings, shaft seals and weldings.

Calculation of the risk of cavitation

To avoid cavitation, the following formula is used to
calculate the maximum suction head:

h

max

– Maximum suction head

Hb – Atmospheric pressure at the pump site; this is the
theoretical maximum suction lift, see figure 2.2.13

Hf – Friction loss in the suction pipe

NPSH = Net Positive Suction Head (is to be read at the
NPSH curve at the highest operational flow), see figure

2.2.12.

b

Pressure [Pa]

Impeller inlet Impeller outlet

a

p

p

1

Vapour pressure

p

Fig.: 2.2.9: Implosion of cavitation bubbles on the back of impeller vanes

Q

NPSH

H

Q

H

Curve when
pump cavitates

Fig.: 2.2.10: Development of pressure through a centrifugal pump

Fig.: 2.2.11: Pump curve when pump cavitates

a = Front of impeller vanes
b = Back of impeller vanes

a

b

a = Front of impeller vanes
b = Back of impeller vanes

89

Imploding vapour bubbles

The NPSH value indicates to what extent the pump is not
able to create absolute vacuum, that is to lift a full water
column 10.33 m above sea level, see figure 2.2.13.

NPSH can either be named NPSHr (required) or NPSHa
(available).

NPSH

required

The required suction head for the pump

NPSH

available

The available suction head in the system

The NPSH value of a pump is determined by testing
according to ISO 9906 and is made in the following
way. The suction head is reduced while the flow is kept
at a constant level. When the differential pressure has
decreased by 3%, the pressure at the pump’s suction side
is read; and the NPSH value of the pump is defined. The
testing is repeated at different flows, which form the basis
of the NPSH curve in the end.

Put in another way: When the NPSH curve is reached, the
level of cavitation is so high that the head of the pump has
decreased by 3%.

Hv – Vapour pressure of the liquid; for more information
concerning vapour pressure of water, go to appendix D.

Hs – Safety factor. Hs depends on the situation and normally
varies between 0.5 m and 1 m and for liquids containing
gas up to 2 m, see figure 2.2.15.

2.2.2 Electrical terms

When you want to examine a pump performance, you need
to know a range of values. In this section we will present
the most important electrical values: Power consumption,
voltage, current and power factor.

Liquid with air

Q [m3/s]

H [m]

NPSH

Vented liquid

Fig.: 2.2.15: Typical NPSH-curve for liquid containing gas

NPSH

H

b

H

f

h

H

v

20

15

12
10

8,0

6,0
5,0

4,0

3,0

2,0

1,0
0,8

0,6

0,4
0,3

0,2

0,1

1,5

120

110

90

100

80

70

60

50

40

30

20

10

0

Hv
(m)tm(˚C )

150

130

140

25

35

45
40

30

20

15

12
10

8,0

6,0
5,0

4,0

3,0

2,0

1,0
0,8

0,6

0,4
0,3

0,2

0,1

1,5

120

110

90

100

80

70

60

50

40

30

20

10

0

Hv
(m)

tm

(˚C )

150

130

140

25

35

45
40

30

Fig.: 2.2.14: System with indication of the
different values that are important in
connection with suction calculations

Height above

sea level

(m)

0

500

1000

2000

1.013 10.33

0.935

100

9.73

0.899

0.795

9.16

8.10

99

96

93

Barometric

pressure

pb (bar)

Water

column

Hb(m)

Boiling point

of water

(˚C)

Fig.: 2.2.13: Barometric pressure above sea level

90

Section 2.2
Pump performance

Power consumption (P)

Pumps are made of several components, see figure 2.2.16.
The power consumption of the different components is
designated in the following way:

P1 The power input from the mains or, put in another

way, the amount of power the consumer has to
pay for

P2 The power input to the pump or the power output

from the motor. Often referred to as shaft power

PH Hydraulic power – the power that the pump

transfers to the liquid in the shape of flow
and head

For the most common pump types, the term power
consumption normally refers to P2. Power is measured
in W, kW.

Efficiency in connection with pumps normally only covers
the efficiency of the pump part ηP. A pump’s efficiency is
determined by several factors, for instance the shape of
the pump housing, the impeller and diffuser design and
the surface roughness. For typical pump units consisting of
both pump and electric motor, the total efficiency ηT also
includes the efficiency of the motor:

If a frequency converter is included as well, the efficiency
of the entire unit also has to incorporate the efficiency of
the frequency converter:

P

1

P

2

P

H

Fig. 2.2.16: Pump unit with indication of different power
consumption levels

91

Voltage (U)

Like pressure drives flow through a hydraulic system,
voltage drives a current (I) through an electrical circuit.
Voltage is measured in volts (V) and can be either direct
current (DC), e.g. 1.5 V battery – or alternating current (AC),
e.g. electricity supply for houses, etc. Normally, pumps are
supplied with AC voltage supply.
The layout of AC mains supply differs from one country to
another. However, the most common layout is four wires
with three phases (L1, L2, L3) and a neutral (N). Besides
these four wires, a protective earth connection (PE) is
added to the system as well, see figure 2.2.17.
For a 3x400 V/230 V mains supply, the voltage between
any two of the phases (L1, L2, L3) is 400 V. The voltage
between one of the phases and neutral (N) is 230 V.
The ratio between the phase-phase voltage and the phase­neutral voltage is determined by the formula on your
right:

Current is the flow of electricity and is measured in ampere
(A). The amount of current in an electrical circuit depends
on the supplied voltage and the resistance/ impedance in
the electrical circuit.

Power (P) and power factor (cosϕ or PF)

Power consumption is indeed of high importance when
it comes to pumps. For pumps with standard AC motors,
the power input is found by measuring the input voltage
and input current and by reading the value cosϕ on the
motor/pump nameplate. cosϕ is the phase angle between
voltage and current. cosϕ is also referred to as power factor
(PF). The power consumption P1 can be calculated by the
formulas shown on your right depending on whether the
motor is a single-phase or a three-phase motor.

L

1

L

2

L

3

N

PE

400V Three-phase supply

230V Single-phase supply

Fig. 2.2.17: Mains supply, e.g. 3 x 400 V

}

}

AC single-phase motor, e.g. 1 x 230 V

AC three-phase motor, e.g. 3 x 400 V

92

Section 2.2
Pump performance

The voltage between any two phases (L1, L2, L3)
is for a 3x400 V/230 V mains supply, 400 V.
The voltage between one of the phases and
neutral (N) is 230V. The ratio between the
phase-phase voltage and the phase-neutral
voltage is:

2.2.3 Liquid properties

When you are making your system calculations, there are
three properties you might need to know about the liquid
in order to make the right calculations: Liquid temperature,
density, and heat capacity.

The liquid temperature is measured in °C (Celcius), K
(Kelvin), or °F (Fahrenheit). °C and K are actually the
same unit but 0°C is the freezing point of water and 0K
is the absolute zero, that is -273.15°C - the lowest possible
temperature. The calculation between Fahrenheit and
Celcius is: °F = °C . 1.8 + 32, hence the freezing point of
water is 0°C and 32°F and the boiling point is 100°C and
212°F.

The density is measured in kg/m3 or kg/dm3.

The heat capacity tells us how much additional energy a
liquid can contain per mass when it is heated. Liquid heat
capacity depends on temperature, see figure 2.2.18. This is
used in systems for transporting energy, e.g. heating, air­con and cooling. Mixed liquids, e.g. glycol and water for
air-con have a lower heat capacity than pure water hence
higher flow is required to transport the same amount of
energy.

93

-40 -20 0 20

60

80

40 100 120˚C

2.0

2.4

2.8

3.2

3.6

4.0

4.4

0.5

0.6

0.7

0.8

0.9

1.0

kJ/kgK kcal/kgK

0% pure water

20%

34%

44%

52%

Fig. 2.2.18: Heat capacity vs. temperature for ethylene glycol

Chapter 3. System hydraulic

Section 3.1: System characteristics

3.1.1 Single resistances

3.1.2 Closed and open systems

Section 3.2: Pumps connected in series and parallel

3.2.1 Pumps in parallel

3.2.2 Pumps connected in series

Section 3.1
System characteristics

Previously, in section 1.1.2 we discussed the basic
characteristics of pump performance curves. In this
chapter we will examine the pump performance curve at
different operating conditions as well as a typical system
characteristic. Finally, we will focus on the interaction
between a pump and a system.

A system’s characteristic describes the relation between the
flow Q and head H in the system. The system characteristic
depends on the type of system in question. We distinguish
between two types: Closed and open systems.

• Closed systems

are circulating systems like heating or air-conditioning
systems, where the pump has to overcome the friction
losses in the pipes, fittings, valves, etc. in the system.

• Open systems

are liquid transport systems like water supply systems. In
such systems the pump has to deal with both the static
head and overcome the friction losses in the pipes and
components.

When the system characteristic is drawn in the same
system of co-ordinates as the pump curve, the duty point
of the pump can be determined as the point of intersection
of the two curves, see figure.3.1.1.

Open and closed systems consist of resistances (valves,
pipes, heat exchanger, etc.) connected in series or
parallel, which altogether affect the system characteristic.
Therefore, before we continue our discussion on open
and closed systems, we will briefly describe how these
resistances affect the system characteristic.

Fig. 3.1.1: The point of intersection between the pump curve and the system
characteristic is the duty point of the pump

96

3.1.1 Single resistances

Every component in a system constitutes a resistance
against the liquid flow which leads to a head loss across
every single component in the system. The following
formula is used to calculate the head loss ∆H:

∆H = k . Q

2

k is a constant, which depends on the component in
question and Q is the flow through the component. As it
appears from the formula, the head loss is proportional to
the flow in second power. So, if it is possible to lower the
flow in a system, a substantial reduction in the pressure
loss occurs.

Resistances connected in series

The total head loss in a system consisting of several
components connected in series is the sum of head losses
that each component represents. Figure 3.1.2 shows a
system consisting of a valve and a heat exchanger. If we
do not consider the head loss in the piping between the
two components, the total head loss ∆H

tot

is calculated by

adding the two head losses:

∆H

tot

= ∆H1 + ∆H

2

Furthermore, figure 3.1.2 shows how the resulting curve
will look and what the duty point will be if the system
is a closed system with only these two components. As
it appears from the figure, the resulting characteristic
is found by adding the individual head losses ∆H at a
given flow Q. Likewise, the figure shows that the more
resistance in the system, the steeper the resulting system
curve will be.

Fig. 3.1.2: The head loss for two components connected in series
is the sum of the two individual head losses

97

Resistances connected in parallel

Contrary to connecting components in series, connecting
components in parallel result in a more flat system
characteristic. The reason is that components installed
in parallel reduce the total resistance in the system and
thereby the head loss.
The differential pressure across the components connected
in parallel is always the same. The resulting system
characteristic is defined by adding all the components’
individual flow rate for a specific ∆H. Figure 3.1.3 shows
a system with a valve and a heat exchanger connected in
parallel.

The resulting flow can be calculated by the following
formula for a head loss equivalent to ∆H

Q

tot

= Q 1 + Q

2

3.1.2 Closed and open systems

As mentioned previously, pump systems are split into two
types of basic systems: Closed and open systems. In this
section, we will examine the basic characteristics of these
systems.

Closed systems

Typically, closed systems are systems, which transport
heat energy in heating systems, air-conditioning systems,
process cooling systems, etc. A common feature of these
types of closed systems is that the liquid is circulated and
is the carrier of heat energy. Heat energy is in fact what the
system has to transport.

Closed systems are characterised as systems with pumps
that only have to overcome the sum of friction losses,
which are generated by all the components. Figure 3.1.4
shows a schematic drawing of a closed system where
a pump has to circulate water from a heater through a
control valve to a heat exchanger.

Fig. 3.1.3: Components connected in parallel reduce the resistance in
the system and result in a more flat system characteristic

Fig. 3.1.4: Schematic drawing of a closed system

98

Section 3.1
System characteristics

All these components together with the pipes and fittings
result in a system characteristic as the one shown in figure

3.1.5. The required pressure in a closed system (which the
system curve illustrates) is a parabola starting in the point
(Q,H) = (0,0) and is calculated by the following formula:

H = k . Q

2

As the formula and curve indicate, the pressure loss is
approaching zero when the flow drops.

Open systems

Open systems are systems, where the pump is used to
transport liquid from one point to another, e.g. water
supply systems, irrigation systems, industrial process
systems. In such systems the pump has to deal with
both the geodetic head of the liquid and to overcome the
friction losses in the pipes and the system components.

We distinguish between two types of open system:

• Open systems where the total required geodetic lift
is positive.

• Open systems where the total required geodetic lift
is negative.

Open system with positive geodetic head Figure

3.1.6 shows a typical open system with positive geodetic
lift. A pump has to transport water from a break tank at
ground level up to a roof tank on the top of a building.
First of all, the pump has to provide a head higher than
the geodetic head of the water (h). Secondly, the pump
has to provide the necessary head to overcome the total
friction loss between the two tanks in piping, fittings,
valves, etc. (Hf). The pressure loss depends on the amount
of flow, see figure 3.1.7.

Fig. 3.1.5: The system characteristic for a closed system is a
parabola starting in point (0,0)

Fig. 3.1.6: Open system with positive geodetic lift

Fig. 3.1.7: System characteristic together with the pump performance
curve for the open system in figure 3.1.6

Q

Q

1

99

The figure shows that in an open system no water flows
if the maximum head (H

max

) of the pump is lower than
the geodetic head (h). Only when H > h, water will start
to flow from the break tank to the roof tank. The system
curve also shows that the lower the flow rate, the lower
the friction loss (Hf) and consequently the lower the power
consumption of the pump.

So, the flow (Q1) and the pump size have to match the
need for the specific system. This is in fact a rule-of­thumb for liquid transport systems: A larger flow leads
to a higher pressure loss, whereas a smaller flow leads to
a smaller pressure loss and consequently a lower energy
consumption.

Open system with negative geodetic lift

A typical example of an open system with negative
required head is a pressure booster system, e.g. in a water
supply system. The geodetic head (h) from the water tank
brings water to the consumer - the water flows, although
the pump is not running. The difference in height between
the liquid level in the tank and the altitude of the water
outlet (h) results in a flow equivalent to Qo. However, the
head is insufficient to ensure the required flow (Q1) to the
consumer. Therefore, the pump has to boost the head to
the level (H1) in order to compensate for the friction loss
(Hf) in the system. The system is shown in figure 3.1.8
and the system characteristic together with the pump
performance curve are shown in figure 3.1.9.

The resulting system characteristic is a parabolic curve
starting at the H-axes in the point (0,-h).

The flow in the system depends on the liquid level in the
tank. If we reduce the water level in the tank the height (h)
is reduced. This results in a modified system characteristic
and a reduced flow in the system, see figure 3.1.9.

Fig. 3.1.8: Open system with negative geodetic lift

Fig. 3.1.9: System characteristic together with the pump
performance curve for the open system in figure 3.1.8

100

Section 3.1
System characteristics