Pumps - Nptel Iit Kanpur
Pumps - Nptel Iit Kanpur
Pumps - Nptel Iit Kanpur
Rotodynamic Pumps
A rotodynamic pump is a device where mechanical energy is transferred from the rotor to the fluid by
the principle of fluid motion through it. The energy of the fluid can be sensed from the pressur and
velocity of the fluid at the delivery end of the pump. Therefore, it is essentially a turbine in reverse.
Like turbines, pumps are classified according to the main direction of fluid path through them like (i)
radial flow or centrifugal, (ii) axial flow and (iii) mixed flow types.
Centrifugal Pumps
The pumps employing centrifugal effects for increasing fluid pressure have been in use for more than
a century.The centrifugal pump, by its principle, is converse of the Francis turbine. The flow is radially
outward, and the hence the fluid gains in centrifugal head while flowing through it. Because of certain
inherent advantages,such as compactness, smooth and uniform flow, low initial cost and high
efficiency even at low heads, centrifugal pumps are used in almost all pumping systems. However,
before considering the operation of a pump in detail, a general pumping system is discussed as
follows.
The word pumping, referred to a hydraulic system commonly implies to convey liquid from a low to a
high reservoir. Such a pumping system, in general, is shown in Fig. 33.1. At any point in the system,
the elevation or potential head is measured from a fixed reference datum line. The total head at any
point comprises pressure head, velocity head and elevation head. For the lower reservoir, the total
head at the free surface is and is equal to the elevation of the free surface above the datum line
since the velocity and static pressure at A are zero. Similarly the total head at the free surface in the
higher reservoir is ( ) and is equal to the elevation of the free surface of the reservoir above
the reference datum.
The variation of total head as the liquid flows through the system is shown in Fig. 33.2. The liquid
enters the intake pipe causing a head loss for which the total energy line drops to
point B corresponding to a location just after the entrance to intake pipe. The total head at B can be
written as
As the fluid flows from the intake to the inlet flange of the pump at elevation the total head drops
further to the point C (Figure 33.2) due to pipe friction and other losses equivalent to . The fluid
then enters the pump and gains energy imparted by the moving rotor of the pump. This raises the
total head of the fluid to a point D (Figure 33.2) at the pump outlet (Figure 33.1).
In course of flow from the pump outlet to the upper reservoir, friction and other losses account for a
total head loss or down to a point E . At E an exit loss occurs when the liquid enters the
upper reservoir, bringing the total heat at point F (Figure 33.2) to that at the free surface of the upper
reservoir. If the total heads are measured at the inlet and outlet flanges respectively, as done in a
standard pump test, then
Figure 33.1 A general pumping system
where and are the velocities in suction and delivery pipes respectively.
(33.1)
The head developed H is termed as manometric head . If the pipes connected to inlet and outlet of
the pump are of same diameter, and therefore the head developed or manometric head H is
simply the gain in piezometric pressure head across the pump which could have been recorded by a
manometer connected between the inlet and outlet flanges of the pump. In practice, ( ) is so
small in comparison to that it is ignored. It is therefore not surprising o find that the
static pressure head across the pump is often used to describe the total head developed by the pump.
The vertical distance between the two levels in the reservoirs is known as static head or static lift.
Relationship between , the static head and H , the head developed can be found out by applying
Bernoulli's equation between A and C and between D and F (Figure 33.1) as follows:
(33.2)
Between D and F ,
(33.3)
substituting from Eq. (33.2) into Eq. (33.3), and then with the help of Eq. (33.1),
we can write
(33.4)
Therefore, we have, the total head developed by the pump = static head + sum of all the losses.
The simplest from of a centrifugal pump is shown in Figure 33.3. It consists of three important parts:
(i) the rotor, usually called as impeller, (ii) the volute casing and (iii) the diffuser ring. The impeller is a
rotating solid disc with curved blades standing out vertically from the face of the disc. The impeller
may be single sided (Figure 33.4a) or doublesided (Figure 33.4b). A double sided impeller has a
relatively small flow capacity.
The tips of the blades are sometimes covered by another flat disc to give shrouded blades (Figure
33.4c), otherwise the blade tips are left open and the casing of the pump itself forms the solid outer
wall of the blade passages. The advantage of the shrouded blade is that flow is prevented from
leaking across the blade tips from one passage to another.
(a) Single sided impeller (b) Double sided impeller (c) Shrouded impeller
Figure 33.4 Types of impellers in a centrifugal pump
As the impeller rotates, the fluid is drawn into the blade passage at the impeller eye, the centre of the
impeller. The inlet pipe is axial and therefore fluid enters the impeller with very little whirl or tangential
component of velocity and flows outwards in the direction of the blades. The fluid receives energy
from the impeller while flowing through it and is discharged with increased pressure and velocity into
the casing. To convert the kinetic energy or fluid at the impeller outlet gradually into pressure energy,
diffuser blades mounted on a diffuser ring are used.
The stationary blade passages so formed have an increasing cross-sectional area which reduces the
flow velocity and hence increases the static pressure of the fluid. Finally, the fluid moves from the
diffuser blades into the volute casing which is a passage of gradually increasing cross-section and
also serves to reduce the velocity of fluid and to convert some of the velocity head into static head.
Sometimes pumps have only volute casing without any diffuser.
Figure 34.1 shows an impeller of a centrifugal pump with the velocity triangles drawn at inlet and
outlet. The blades are curved between the inlet and outlet radius. A particle of fluid moves along the
broken curve shown in Figure 34.1.
Figure 34.1 Velocity triangles for centrifugal pump Impeller
Let be the angle made by the blade at inlet, with the tangent to the inlet radius, while is the
blade angle with the tangent at outlet. and are the absolute velocities of fluid at inlet an outlet
respectively, while and are the relative velocities (with respect to blade velocity) at inlet and
outlet respectively. Therefore,
A centrifugal pump rarely has any sort of guide vanes at inlet. The fluid therefore approaches the
impeller without appreciable whirl and so the inlet angle of the blades is designed to produce a right-
angled velocity triangle at inlet (as shown in Fig. 34.1). At conditions other than those for which the
impeller was designed, the direction of relative velocity does not coincide with that of a blade.
Consequently, the fluid changes direction abruptly on entering the impeller. In addition, the eddies
give rise to some back flow into the inlet pipe, thus causing fluid to have some whirl before entering
the impeller. However, considering the operation under design conditions, the inlet whirl velocity
and accordingly the inlet angular momentum of the fluid entering the impeller is set to zero. Therefore,
Eq. (34.1) can be written as
We see from this equation that the work done is independent of the inlet radius. The difference in total
head across the pump known as manometric head, is always less than the quantity
because of the energy dissipated in eddies due to friction.
The ratio of manometric head H and the work head imparted by the rotor on the fluid
(usually known as Euler head) is termed as manometric efficiency . It represents the effectiveness
of the pump in increasing the total energy of the fluid from the energy given to it by the impeller.
Therefore, we can write
(34.3)
(34.4)
where, Q is the volume flow rate of the fluid through the pump, and P is the shaft power, i.e. the input
power to the shaft. The energy required at the shaft exceeds because of friction in the
bearings and other mechanical parts. Thus a mechanical efficiency is defined as
(34.5)
so that
(34.6)
Slip Factor
Under certain circumstances, the angle at which the fluid leaves the impeller may not be the same as
the actual blade angle. This is due to a phenomenon known as fluid slip, which finally results in a
reduction in the tangential component of fluid velocity at impeller outlet. One possible
explanation for slip is given as follows.
In course of flow through the impeller passage, there occurs a difference in pressure and velocity
between the leading and trailing faces of the impeller blades. On the leading face of a blade there is
relatively a high pressure and low velocity, while on the trailing face, the pressure is lower and hence
the velocity is higher. This results in a circulation around the blade and a non-uniform velocity
distribution at any radius. The mean direction of flow at outlet, under this situation, changes from the
blade angle at outlet to a different angle as shown in Figure 34.2 Therefore the tangential
34.2, and the difference is defined as the slip. The slip factor is defined as
Figure 34.2 Slip and velocity in the impeller blade passage of a centrifugal pump
With the application of slip factor , the work head imparted to the fluid (Euler head)
becomes . The typical values of slip factor lie in the region of 0.9.
Mechanical friction power loss due to friction between the fixed and rotating parts in the bearing
and stuffing boxes.
Disc friction power loss due to friction between the rotating faces of the impeller (or disc) and the
liquid.
Leakage and recirculation power loss. This is due to loss of liquid from the pump and recirculation of
the liquid in the impeller. The pressure difference between impeller tip and eye can cause a
recirculation of a small volume of liquid, thus reducing the flow rate at outlet of the impeller as shown
in Fig. (34.3).
theoretical head developed. Therefore we can write for theoretical head developed as
(35.1)
(35.2)
where Q is rate of flow at impeller outlet and A is the flow area at the periphery of the impeller. The
blade speed at outlet can be expressed in terms of rotational speed of the impeller N as
Using this relation and the relation given by Eq. (35.2), the expression of theoretical head developed
can be written from Eq. (35.1) as
(35.3)
where, and
For a given impeller running at a constant rotational speed. and are constants, and therefore
head and discharge bears a linear relationship as shown by Eq. (35.3). This linear variation of
with Q is plotted as curveI in Fig. 35.1.
If slip is taken into account, the theoretical head will be reduced to . Moreover the slip
will increase with the increase in flow rate Q . The effect of slip in head-discharge relationship is
shown by the curve II in Fig. 35.1. The loss due to slip can occur in both a real and an ideal fluid, but
in a real fluid the shock losses at entry to the blades, and the friction losses in the flow passages have
to be considered. At the design point the shock losses are zero since the fluid moves tangentially onto
the blade, but on either side of the design point the head loss due to shock increases according to the
relation
(35.4)
Figure 35.1 Head-discharge characteristics of a centrifugal pump
where is the off design flow rate and is a constant. The losses due to friction can usually be
expressed as
(35.5)
where, is a constant.
Equation (35.5) and (35.4) are also shown in Fig. 35.1 (curves III and IV) as the characteristics of
losses in a centrifugal pump. By subtracting the sum of the losses from the head in consideration of
the slip, at any flow rate (by subtracting the sum of ordinates of the curves III and IV from the ordinate
of the curve II at all values of the abscissa), we get the curve V which represents the relationship of
the actual head with the flow rate, and is known as head-discharge characteristic curve of the pump.
The head-discharge characteristic of a centrifugal pump depends (among other things) on the outlet
angle of the impeller blades which in turn depends on blade settings. Three types of blade settings
are possible (i) the forward facing for which the blade curvature is in the direction of rotation and,
therefore, (Fig. 35.2a), (ii) radial, when (Fig. 35.2b), and (iii) backward facing for
which the blade curvature is in a direction opposite to that of the impeller rotation and
therefore, (Fig. 35.2c). The outlet velocity triangles for all the cases are also shown in Figs.
35.2a, 35.2b, 35.2c. From the geometry of any triangle, the relationship between and can
be written as.
In case of forward facing blade, and hence cot is negative and therefore is more
blade, and Therefore the sign of , the constant in the theoretical head-
discharge relationship given by the Eq. (35.3), depends accordingly on the type of blade setting as
follows:
With the incorporation of above conditions, the relationship of head and discharge for three cases are
shown in Figure 35.3. These curves ultimately revert to their more recognized shapes as the actual
head-discharge characteristics respectively after consideration of all the losses as explained earlier
Figure 35.4.
For both radial and forward facing blades, the power is rising monotonically as the flow rate is
increased. In the case of backward facing blades, the maximum efficiency occurs in the region of
maximum power. If, for some reasons, Qincreases beyond there occurs a decrease in power.
Therefore the motor used to drive the pump at part load, but rated at the design point, may be safely
used at the maximum power. This is known as self-limiting characteristic. In case of radial and
forward-facing blades, if the pump motor is rated for maximum power, then it will be under utilized
most of the time, resulting in an increased cost for the extra rating. Whereas, if a smaller motor is
employed, rated at the design point, then if Q increases above the motor will be overloaded and
may fail. It, therefore, becomes more difficult to decide on a choice of motor in these later cases
(radial and forward-facing blades).
Figure 35.3 Theoretical head-discharge characteristic curves of a
centrifugal pump for different blade settings
round the circumference. The combination of uniform radial velocity with the free vortex (
=constant) gives a pattern of spiral streamlines which should be matched by the shape of the volute.
This is the most important feature of the design of a pump. At maximum efficiency, about 10 percent
of the head generated by the impeller is usually lost in the volute.
Vanned Diffuser
A vanned diffuser, as shown in Fig. 36.1, converts the outlet kinetic energy from impeller to pressure
energy of the fluid in a shorter length and with a higher efficiency. This is very advantageous where
the size of the pump is important. A ring of diffuser vanes surrounds the impeller at the outlet. The
fluid leaving the impeller first flows through a vaneless space before entering the diffuser vanes. The
divergence angle of the diffuser passage is of the order of 8-10 which ensures no boundary layer
separation. The optimum number of vanes are fixed by a compromise between the diffusion and the
frictional loss. The greater the number of vanes, the better is the diffusion (rise in static pressure by
the reduction in flow velocity) but greater is the frictional loss. The number of diffuser vanes should
have no common factor with the number of impeller vanes to prevent resonant vibration.
Cavitation is likely to occur at the inlet to the pump, since the pressure there is the minimum and is
lower than the atmospheric pressure by an amount that equals the vertical height above which the
pump is situated from the supply reservoir (known as sump) plus the velocity head and frictional
losses in the suction pipe. Applying the Bernoulli's equation between the surface of the liquid in the
sump and the entry to the impeller, we have
(36.1)
where, is the pressure at the impeller inlet and is the pressure at the liquid surface in the
sump which is usually the atmospheric pressure, Z1 is the vertical height of the impeller inlet from the
liquid surface in the sump, is the loss of head in the suction pipe. Strainers and non-return valves
are commonly fitted to intake pipes. The term must therefore include the losses occurring past
these devices, in addition to losses caused by pipe friction and by bends in the pipe.
In the similar way as described in case of a reaction turbine, the net positive suction head 'NPSH' in
case of a pump is defined as the available suction head (inclusive of both static and dynamic heads)
at pump inlet above the head corresponding to vapor pressure.
Therefore,
(36.2)
The Thomas cavitation parameter s and critical cavitation parameter are defined accordingly (as
done in case of reaction turbine) as
(36.3)
and (36.4)
In order that s should be as large as possible, z must be as small as possible. In some installations, it
may even be necessary to set the pump below the liquid level at the sump (i.e. with a negative vale
of z ) to avoid cavitation.
The Figure 37.1 shows an axial flow pump. The flow is the same at inlet and outlet. an axial flow
pumps develops low head but have high capacity. the maximum head for such pump is of the order of
20m.The section through the blade at X-X (Figure 37.1) is shown with inlet and outlet velocity
triangles in Figure 37.2.
Figure 37.1 A propeller of an axial flow pump
The blade has an aerofoil section. The fluid does not impinge tangentially to the blade at inlet, rather
the blade is inclined at an angle of incidence
(i) to the relative velocity at the inlet . If we consider the conditions at a mean radius then
For maximum energy transfer , , i.e .Again , from the outlet velocity triangle,
(37.1)
For constant energy transfer over the entire span of the blade from hub to tip , the right hand side of
Equation (37.1) has to be same for all values of . It is obvious that increases with radius ,
therefore an equal increase in must take place, and since is constant then
must increase. Therefore , the blade must be twisted as the radius changes.
The design point of a hydraulic pump corresponds to a situation where the overall efficiency of
operation is maximum. However the exact operating point of a pump, in practice, is determined from
the matching of pump characteristic with the headloss-flow, characteristic of the external system (i.e.
pipe network, valve and so on) to which the pump is connected.
Let us consider the pump and the piping system as shown in Fig. 15.18. Since the flow is highly
turbulent, the losses in pipe system are proportional to the square of flow velocities and can,
therefore, be expressed in terms of constant loss coefficients. Therefore, the losses in both the
suction and delivery sides can be written as
(37.2a)
(37.2b)
where, is the loss of head in suction side and is the loss of head in delivery side and f is the
Darcy's friction factor, and are the lengths and diameters of the suction and delivery
pipes respectively, while and are accordingly the average flow velocities. The first terms in
Eqs. (37.1a) and (37.1b) represent the ordinary friction loss (loss due to friction between fluid ad the
pipe wall), while the second terms represent the sum of all the minor losses through the loss
coefficients and which include losses due to valves and pipe bends, entry and exit losses,
etc. Therefore the total head the pump has to develop in order to supply the fluid from the lower to
upper reservoir is
(37.3)
Now flow rate through the system is proportional to flow velocity. Therefore resistance to flow in the
form of losses is proportional to the square of the flow rate and is usually written as
(37.4)
= system resistance =
where K is a constant which includes, the lengths and diameters of the pipes and the various loss
coefficients. System resistance as expressed by Eq. (37.4), is a measure of the loss of head at any
particular flow rate through the system. If any parameter in the system is changed, such as adjusting
a valve opening, or inserting a new bend, etc., then K will change. Therefore, total head of Eq. (37.2)
becomes,
(37.5)
The head H can be considered as the total opposing head of the pumping system that must be
overcome for the fluid to be pumped from the lower to the upper reservoir.
The Eq. (37.4) is the equation for system characteristic, and while plotted on H-Q plane (Figure 37.3),
represents the system characteristic curve. The point of intersection between the system
characteristic and the pump characteristic on H-Q plane is the operating point which may or may not
lie at the design point that corresponds to maximum efficiency of the pump. The closeness of the
operating and design points depends on how good an estimate of the expected system losses has
been made. It should be noted that if there is no rise in static head of the liquid (for example pumping
in a horizontal pipeline between two reservoirs at the same elevation), is zero and the system
curve passes through the origin.
Figure 37.3 H-Q Characteristics of pump and system
For points A, B and C , the corresponding heads and flows at a new speed are found as follows:
(37.7)
Applying Eqs. (37.6) and (37.7) to points A, B and C the corresponding points and are
found and then the characteristic curve can be drawn at the new speed
Thus,
and
which gives
or (37.8)
Equation (37.8) implies that all corresponding or similar points on Head-Discharge characteristic
curves at different speeds lie on a parabola passing through the origin. If the static lift becomes
zero, then the curve for system characteristic and the locus of similar operating points will be the
same parabola passing through the origin. This means that, in case of zero static life, for an
operating point at speed , it is only necessary to apply the similarity laws directly to find the
corresponding operating point at the new speed since it will lie on the system curve itself (Figure
37.4).
and
or, (38.1)
When the head or flow rate of a single pump is not sufficient for a application, pumps are combined in
series or in parallel to meet the desired requirements. Pumps are combined in series to obtain an
increase in head or in parallel for an increase in flow rate. The combined pumps need not be of the
same design. Figures 38.1 and 38.2 show the combined H-Q characteristic for the cases of identical
pumps connected in series and parallel respectively. It is found that the operating point changes in
both cases. Fig. 38.3 shows the combined characteristic of two different pumps connected in series
and parallel.
The concept of specific speed for a pump is same as that for a turbine. However, the quantities of
interest are N, Hand Q rather than N, H and P like in case of a turbine.
For pump
(38.2)
Figure 38.3 Two different pumps connected in series and parallel
The effect of the shape of rotor on specific speed is also similar to that for turbines. That is, radial flow
(centrifugal) impellers have the lower values of compared to those of axial-flow designs. The
impeller, however, is not the entire pump and, in particular, the shape of volute may appreciably affect
the specific speed. Nevertheless, in general, centrifugal pumps are best suited for providing high
heads at moderate rates of flow as compared to axial flow pumps which are suitable for large rates of
flow at low heads. Similar to turbines, the higher is the specific speed, the more compact is the
machine for given requirements. For multistage pumps, the specific speed refers to a single stage.