Fluid Dynamics And Turbomachines.

Dr Dhiman Chatterjee.
Department Of Mechanical Engineering.
Indian Institute Of Technology Madras.
Part C.
Module-2.
Lecture-6.
Pumps.
–
Good afternoon. I welcome you all for week 7 lecture in fluid dynamics and Turbomachines.
In the last 2 weeks we have discussed about various aspects of Turbomachines, mainly the
fundamentals. We have derived Euler’s energy equation which is the heart of these
Turbomachines energy transfer which talks about the energy transfer between the fluid and
the blade or either from the fluid to the machine or from the machine to the fluid. But though
we have referred to different Turbomachines, we have not gone into details of each of these.

From today onward, that is in this week and the following week, we will take up specific
Turbomachine applications and we will talk about in details each of these machines. Today I
will discuss on pumps. And we already know, from our earlier discussion, the classification
of pumps based on specific speed.

(Refer Slide Time: 1:23)

Just recollect that though there are different ways of classifying pumps, we have talked about
the use of shape number, we have also talked about the use of specific speed in which the
dimensional terms come into picture. Here we will use the definition of specific speed as we
have defined earlier which is used mostly in the pump industries. So we say that N is in
revolutions per minute rpm, V dot in metre cube per second and H in meters. I again want to
reaffirm that if we use a definition the specific speed of pumps, we need to be very careful
about the units because Nq has a unit which is rpm in this case.

And we have also seen that as specific speed increases like we see from A to E, the shape of
the impeller changes. At very low specific speeds, that is in case of A we see that D2 by D1,
that is the outlet diameter to the inlet diameter is large. I will give the values when we talk in
the next slide. When specific speed increases, what do we find? We find that the outlet
diameter reduces. And you may say that we will still like to have a radial flow pump. So in B
we see a radial flow pump with moderate values of D2 by D1 ratios.

And in C, we find that D2 by D1 has reduced further. One more aspect you will see which is
important is that the inlet edge which was straight in case of A or in B has now been curved.
This is because as you reduce the diameter D2, then what happens, the blade length which is
given in this simple radial flow case is this length will also reduce. If the blades length
reduces, then what happens, the guidance to the flow will be improper. And hence in order to
extend the guidance further, in order to increase the length of the blade little more, what is
done is the inlet edge has been drawn almost to the eye of the impeller and the edge is made
curved so that the blade length can be somewhat compensated, though the D2 by D1 has
reduced.

With further increase in the specific speed, such an adjustment is also not possible. And what
we get is a mixed flow impeller and you can see that the flow direction inside the impeller
makes an angle θ and that θ is less than 90 degrees and more than zero. So it is somewhere in
between, the axial and radial. If you increase the specific speed further, then what happens is
D1 becomes equal to D2 and the flow becomes purely axial. So if you are given the task of
designing a pump, you, we should first ask ourselves what is the specific speed.
(Refer Slide Time: 5:10)

We need to have some charts to know what range the specific speed falls and what should be
the impeller shape A, B, C, D or E and this is the guideline which is given for the type of the
impellers that we can choose for different values of specific speed. And we see that between
11 and 38 we have radial impellers with load D1 by D2, note that while discussing the last
slide I talked in terms of D2 by D1, here it is given in terms of D1 by D2 and hence low D1
by D2 is being mentioned.

Then we have radial impellers with medium D1 by D2 followed by radial impellers with high
D1 by D2 and then we have the mixed flow impellers and finally the axial flow impellers.
Also you may be inquisitive to know that why do not we have any values less than 10. What
am I supposed to do if I get Nq for some application less than 10? Well the first thing we can
try is to use multi-staging, that is we can connect 2 impellers in series and stick to the same
shaft.

The fluid from one impeller gets energy from this impeller and then it goes to the 2 nd one and
finally it is collected and taken to the downstream of the system. So this is possible but
sometimes this does not become economical and in these cases recollect what we have
discussed about positive displacement pumps. We have stated that positive displacement
pumps are useful when there is a large pressure rise accompanied with a small volume flow
rate. And hence you can think of Nq, that is specific speed will be lower when head is large
because head is in the denominator and hence you can also use a positive displacement pump.
I have given a reciprocating pump here just as an example but what I mean is this is a
positive displacement pump. Thus when we have Nq less than 10, it is many times advisable,
economically desirable to have a positive displacement machine and not a Turbomachines.
So in this course we will restrict ourselves to Nq greater than 10.

(Refer Slide Time: 7:46)

Next we have talk about generalized description of shape number or specific speed. What is
the necessity? This is because we know that the pump, just now as we have discussed, can be
multi-staged. We have shown 2 stages but it can have more than 2 stages. How do I then
define the total head developed by the pump and the head that is developed by an individual
impeller? Or we can have a volume the flow requirement which is higher, in that case as you
can recall we have discussed earlier we can have a double suction pump.

I have already discussed the double suction pump. So how do I describe the specific speed
or the shape number in these 2 cases. One, multistage pump and second double suction pump.
Because our definition of Nq given so far does not tell me what H is, what is V dot.It says
that H is the head developed by the pump and V dot is a flow rate through the pump. So we
need to refine this definition of specific speed and shape number. I will give an example to
tell how this can be done.

Many times what happens is that in a multistage pump we make identical impellers and if the
total head to be developed is H, then what we do is, we divide the head development
requirement per stage and have that much number of stages. And hence we should write the
definition of specific speed as N square root V dot by H by S to the power three fourth. Or in
symbolically we can write it as N multiplied by square root of V dot divided by H by S to the
power three fourth.

Here H by S refers to the head developed per stage, where H is the total head developed by
the pump and S is the number of stages. If we have a double suction pump, then we have to
consider the volume flow rate. And hence we can say that we have to talk about the volume
flow rate per entry. So what we can do is, we can now define the most generalized form of
the specific speed or the shape number, I am talking about specific speed particularly as Nq is
equal to N times square root of V dot by E where E is the number of entry whole divided by
H by S to the power three fourths.

If you have a particular case when it is a single entry, single end suction pump let us say and
we have a single stage, then E and S becomes 1 and we get back to the formula given at the
top of the slide, that is Nq equal to N square root of V dot whole divided by H the power
three fourth. So now you understand that as the case can be, we have to take care of the
number of stages and the number of entries and as the specific speed changes, not only the
impeller size changes, we can also talk about the other way the impeller can be different.

(Refer Slide Time: 11:28)

As I have already discussed, let us just recollect what we have discussed in the earlier
lectures. We have talked about semi-open and a closed impeller. And you can also see that
time we had not introduced the meridional views, so you can see in the semi-open impeller,
the front shroud is missing and this is also described in the meridional view. You can contrast
the semi-open impeller with the closed impeller and you can see in this case of closed
impeller, we have both the front shroud and the rear shroud and both the shrouds are hatched
here. Whereas in the case of semi-open impeller, only the rear shroud is visible, because the
front shroud is not present.

(Refer Slide Time: 12:16)

But that is not the only difference as far as the impellers are concerned. That is, impellers can
be long or short, radial or mixed or axial, can be semi-open or closed but we can also talk in
terms of the individual blade shapes. Let us take the end of low specific speed where we have
a radial flow impeller and D2 by D1 is very large or let me say it, in the terms of D1 by D2 is
very small, D1 is the impeller inlet diameter. In these cases the blades are found to have a
single curve. Imagine that you have a paper in your hand and you bend the paper and look at
it --- this forms a single curvature blade.

So this is an individual blade shown which you can relate to the plate that I have shown here
(Refer Slide Time: 11:28). Imagine that I have just taken one of these blades and am showing
here. So you see that it has one curvature which is bend here, whereas the higher specific
speed radial flow impellers, you will see blade with double curvature or twist will be visible
and you can see the picture given here. How to identify it, you can see that when you look
from the top, here we will see only this edge. In this case you can see a part of the surface is
visible on the other side and the twist is clear. So we called these blades as played by double
curvature.
(Refer Slide Time: 13:56)

Now this curvature of blades is very important. It plays a very important role in blade specific
work, and pressure rise inside the pump and hence in the degree of reaction. You can also
appreciate the plate curvature is very important from our earlier discussions because we know
that W bl infinity W bl has a term which depends on C2u. So let us look at that. Let us
assume as I have said in many cases earlier that the inlet whirl or the pre-whirl centrifugal
pump can be neglected. We can assume that the flow is entering in such a way that C1 equal
to C1m and C1u equal to 0.

And this is the inlet velocity triangles. U1, W1 and C1 corresponds to the let U2, W2 and C2
correspondence the outlet. I once again remind all of you that this subscript 1 refers to the
low-pressure side in case of pump 1 refers to the low-pressure side which is the inlet, the
subscript 2 refers to the high pressure side, in case of pump the high pressure side is the
outlet. And hence we can write W bl infinity at this point as I am not worried about the slip
and the losses that we have talked about earlier.

We will say we have an idealized vane congruent flow and you can recall that we have used
the notation bl for blade and infinity for vane congruent flow and hence W bl infinity refers
to the blade specific work under the vane congruent flow, that is with idealized assumptions.
We will relax these assumptions and we will take into consideration actual specific work later
on today. So when we write Wbl infinity, we write U2 C2u and - U1 C1u and hence by our
assumption, C1 equal to C1m and hence there is no whirl component which goes off.
Now look at the outer velocity triangles. This is the C2u and this is U2, this is C2m, hence
what happens is this distance is nothing but U 2 - C2m cot β2. And hence once we replace
that U C2u by U2 - C2m cot β2. We get in terms of U2 and C2 m. Now C2m or for that
matter Cm is related with the volume flow rate. For a radial flow pump as I have shown in the
last lecture that V dot is nothing but pi D2 b2 times C2m or C2m can be written in terms of
volume flow rate divided by the effective flow area letter pi D2 b2.

Since it is a vane congruent flow, I do not have to worry about the blockage because of the
blade thickness. If in real life you have a blade thickness, then that can also be accommodated
suitably as we have discussed in the last lecture. And for an axial flow, we can write that V
dot is pi by 4 tip diameter square - hub diameter square multiplied by C2m and we get C2m is
V dot by pi by 4 Dt square - Dh square. What am I trying to say from this is that Cm is
directly proportional to V dot for a given pump.

(Refer Slide Time: 18:05)

I can write that Cm or C2m is nothing but some proportionality constant which depends on
the area times the volume flow rate. And if we write W bl infinity equal to U2 square - U2
C2m cot β2. Essentially what we have done is we have removed the bracket. So we can write
that now U2 square - U2 K area times V dot cot β2 as we have just discussed that the C2m is
related with the volume flow rate with some area related constant, which is constant for a
given pump.

And hence if I say I have a pump of a given size which means my Karea gets fixed and if I
also say the pump runs at a constant rotational speed, then I should be able to write W bl
infinity as A - BV dot cot β2. Note here A is the function of the size as well as the rotational
speed, B is a function of the size and the rotational speed, size comes in 2 ways, from through
U2 as well as from K-area. And hence we get W bl infinity equal to A - BV dot cot β2. And 3
cases may arise depending on the values of β2.

I may say that I have a value of β2 less than 90 degrees. I may say it will have it equal to 90
degrees or greater than 90 degrees. If it is less than 90 degrees, then I get that W bl infinity is
A - B prime V dot. Why am I writing B prime? Because I am taking B cot β2 because in this
case β2 is fixed. The blade has a definite curvature. β2 recall is the impeller blade angle at the
exit. β is the blade angle, 2 is the exit or the impeller outlet in case of pump. So β2 is the
blade angle at the impeller exit as we have discussed in the last class.

So we say that it is less than 90 degree, means that whatever be the value, it is fixed. So I can
write B cot β2 as B prime times V dot. And we can say that I get a plot is V dot increase as,
then W bl infinity decreases linearly. I can also have β2 equal to 90 degree. If β2 is equal to
90 degree, then this term does not contribute and you get W bl infinity as A and you see that
this is the W bl infinity equal to A or it does depend only on the speed and the size of the
pump.

We can also have β2 greater than 90 degree which gives you, now greater than 90 degree cot
beta 2 is negative, we have the - sign here, so - B cot β2 can be written as + B prime V dot.
And hence we see that with increase of volume flow rate W bl infinity increases linearly.
Thus we may be tempted to say that I would like to have a β2 greater than 90 degrees because
it gives me higher head for a given volume flow rate. That is if I draw a vertical line through
here, we will find that at a given volume flow rate W bl infinity is more for greater than 90
degree case.
(Refer Slide Time: 21:59)

Of course I have assumed that A is same in all the 3 cases. This is what we would like to
investigate now. We will do it soon but let us look at the actual scenario in which the blade
specific work W bl infinity we have said, how does it lead to the specific work of the pump
W. So this red Line shows the linear curve for speed constant and β2 less than 90 degrees. I
have taken a case where β2 is less than 90 degrees. Then we have to also consider that the
pump will not have infinite number of blades, vane congruent assumption breaks down, and
we will have slip.

And once we have the slip, we get the W bl, which is related with W bl infinity if you
recollect and we get this blue line. Then we have not still considered the losses. And once we
consider the losses we can say that the loss will be minimum at the design point. And at off
design point, the loss will increase, why will it increase, it will increase because not only the
other losses increase but also the new contribution comes from the shock or the incidence
loss as we have discussed.

The shock or the incidence loss is minimum at the design point, nominally it can be taken to
be zero at the design point and it increases on either side of the design point, that is at flow
rates higher than the design flow rate and at flow rates lower than the design flow rate. And
hence we get a point where you have the minimum losses, this vertical hatched lines in black
means the loss and you see the loss is minimum at this point and increases both ways.

And this loss includes both frictional loss as well as the shock loss as we have discussed. This
was shown in detail for one blade angle which is less than 90 degrees. However if we have a
blade which is more than 90 degree or equal to 90 degree, we have to draw different curves. I
want to re-emphasize that the speed has been kept constant. If you remember, in the earlier
derivation we had said that for a given rotational speed, for a given pump and hence
whenever you do pump performance test, these are called pump performance, we have to
mention that the speed is constant.

So now what we see is that this curve I am now reproducing here, so for β2b less than 90
degree we get a curve as shown, β2B greater than 90 degree or β2b equal to 90 degree can
give you a situation which I will take up little later. It is an interesting situation but not so
happy situation for a pump user, so called the unstable curve. I will talk about
thosecharacteristics little later. And we have the β2b greater than 90 degree as well.

(Refer Slide Time: 24:56)

So let us make some of the observations, whatever we have learned so far. As β2b is
increased, C2u increases as C2u is U2 - C2m cot β2 and we know that cot β2 value which is
nothing but cos β2 by sin β2 will now have a value going on reducing and hence this
contribution goes down and we will have C2u value increases. Of course, beyond 90 degree
this becomes negative and we get increase further because of the sign change. So specific
work increases, just now we saw for a given size of pump and speed of rotation.

We have not yet explored whether that is good and desirable, that we will take up soon. For a
given specific work, you may say that why should I keep the pump size constant, I do not
need to do that. So in that case what I can say is that I will keep the specific work constant
and reduce the size while keeping the speed of rotation constant and of course we have
assumed that the inlet blade angle is not changed. So let us consider the case that if we have
U 2 C2u which is my W bl infinity, I want to keep the product of U2 C2u constant.

And if specific work is constant, then what happens? Then we see that if C2u increases, U2
has to reduce. If speed of rotation is kept constant, it leads to the fact that diameter has to
reduce and we get very compact blade. So you see this is the relative diagram where we can
show that the blade, when the angle is less than 90 degree, the size is bigger than this radial
tip blade. I will talk about it when β2b 90 degree and finally if β2b is greater than 90 degree
we see that the for the same specific work, the size has reduced.

That is simply because, I again repeat, the mathematical relationship also will tell you that W
bl infinity is nothing but U2 C2u and C2u changes and hence to keep W bl infinity constant
as U2 has to reduce.

(Refer Slide Time: 27:29)

So let us look at this for an axial flow machine. Axial flow pumps we have already discussed
and the flow takes place from 1 to 2 and what we see here is that we see that from left to right
the blade angle increases. In the left side we have β2b less than 90 degree, in the right one we
have β2b greater than 90 degrees. And we can say as the blade angle becomes greater than 90
degree, what has happened is the C2u component has increased tremendously. C2 has
increased, C2u has increased tremendously.

You look at the leftmost velocity triangle and the rightmost one. You see that C2u is much
smaller here compared to the C2u over on the right-hand side. So what we are assuming right
now is that β1b will be kept constant, the speed will be kept constant as we have discussed
already, these are very important because if we start changing these, the entire derivation
becomes more difficult. Nevertheless we have the same relationships, the same effect can be
observed. We are trying to keep it same, particularly the β1b because we do not want to
understand the effect being compounded by the β1b.

And then Cm is unchanged. So in all these cases, we are keeping the meridional component
of the velocity, absolute velocity to be constant.

(Refer Slide Time: 29:06)

Similarly we can talk about in case of the radial flow pumps. In the case of radial flow
pumps, it is slightly more inward, let us try to understand that. We have discussed direction
of rotation in case of axial flow pumps. How did we define it? We said the direction of
rotation in case of pumps is dictated by the blade. The blade makes the fluid rotate and hence
the direction of rotation in an axial flow pump which is basically the blade shape of having an
aerofoil from the suction surface towards the pressure surface.

In case of pumps I have the direction of rotation, let us say, fixed as shown in each of these
figures. I can have a β2b which is less than 90 degree, I can show you that the arrows are also
shown so the flow leaves the blades at this point, we have W2 C2u in each case and we have,
the names are given for these blades as β2b less than 90 degree, we say the blade is backward
facing vane or backward facing blade. Why do we have such a name for these blades? That is
because you imagine that you are standing on these vanes and where is the centre of
curvature of this vane?
It is somewhere here. That is if I say that I have a vane where you are standing here, you are
here and the centre of curvature is here, then what happens, you are standing here looking at
the centre of curvature, you are looking against the direction of rotation. You are looking
backward to see the centre of the curvature of the blades. Whereas you take this one, in this
case you are again sitting on the vanes but the centre of curvature is somewhere near which
means you are looking in the direction as the direction of rotation of the blades.

So this one will be considered as the forward facing vane. If you look at it more attentively,
the 2nd one is also the same. In this case you have the observer here and the centre of
curvature here which is again in the direction of rotation. However for the case B, we give it a
special name to ensure the fact is kept in our mind that β2b is 90 degree and hence we say
that β2b is 90 degree is going to be the case of radial tipped vane. So you understand that
broadly speaking I should have classified it as backward facing and the forward facing vane.

But though radial tipped vane is essentially a forward facing vane, we call it a radial tipped
vane to make us feel comfortable or aware that this β2b is 90 degrees. Okay, so there are 3
possibilities, both in axial as well as radial flow pumps. Which one should we go for, which
one is normally the practice, let us try to understand that? Before giving the answer, let us do
a little more of the work.

(Refer Slide Time: 32:22)

So let us talk about the velocity triangles and like I have discussed, I have taken C1 equal to
Cm1 and I have taken the C1u to be zero. I have also considered an axial flow machine.
Please note that the derivation that I am trying to show you to bring out the effect of the blade
angle need not be restricted to axial flow. You can do the same derivation for radial flow
pump as well but the calculations will be more cumbersome. So I feel that to bring out the
physics, to bring out the significance of the blade curvature, it suffices if we take the axial
flow pump.

And also there is the nice assumption of C1u equal to 0. So in this case I know that U2 is
equal to U1 and hence I can write that W bl infinity is U times C2u. And which we have
already done. We can say that we have already discussed while discussing the degree of
reaction that a portion of the energy is used to increase the pressure in pump and the other
part of course goes for the dynamic head. So we can say that Wp infinity, this is my way of
saying that we are talking about the component of energy transfer which goes into the
pressure, p here signifies the pressure.

And infinity to remind us again that we are restricting ourselves to vane congruent flow, ideal
flow, lossless flow and we can also say like some books of will say that W bl infinity st for
static. So whatever notation we follow, we are essentially talking about the static pressure
change. And we know the static pressure change is nothing but the total work transfer W bl
infinity - C2 - square - C1 square the whole divided by 2. So let us write that, we write the
Wp infinity is equal to U C2u - C2 square - C1 square by 2.

Which gives us that Wp infinity is equal to half C2u multiplied by twice U - C2u. This result
is important. What it says, if C2u is 0 which is a trivial case because if C1u is zero and C2u is
zero, there is no power transfer, there is no energy transfer. So that is a trivial case which
cannot be accepted. But anyhow, mathematically if C2u is zero, then there is no pressure rise
but C2u if it is equal to twice the U, then also you get no pressure rise. What does it mean?
When there is no pressure rise we are talking about an impulse stage when there is no
pressure change.

So what will happen is, in that case, we will see that Wp infinity is zero at C2u equal to 0 and
at C2u equal to 2U. If we subtract the static pressure change from the total, we get what is
known as W bl infinity dyn, dyn stands for dynamic which is half C2 square. So what will
happen? W bl infinity goes with C2u as linear and this blue shaded region, this one gives me
the limit of W bl infinity or the static pressure change, you can see that this area is zero at
C2u equal to 0 and C2u equal to 2U which is the impulse case.
And W bl infinity dynamic is maximum at this point where C2u equal to twice U. So what
we see, if we increase blade angle β2, then what we increase is C2u and if we increase C2u
further and further, a point will come where there is no pressure change. Is it good, is it
desirable, though your W bl infinity has increased? The answer is no. Why?.

(Refer Slide Time: 36:32)

To go into that we scale that high kinetic energy at the impeller exit, in case of pumps needs
to be converted in the stationary diffuser system. What is the objective of a pump? To lift
water, to raise water, let us say from the ground level to the top of a multistoried building, a
high-rise building. So this means that the fluid should have enough pressure which can take it
to the top of the building. Now we have a situation when the impeller exit, there is the
extreme limit, in the extreme limit when C2u equal to 2U there is no static pressure, it has a
very high content of kinetic energy.

So if we have to convert this into pressure then we need to have to diffuse the flow and in the
stationary diffuser there will be losses. If the kinetic energy is high, the velocities are high,
then the losses are more and hence when we try to convert this into pressure rise, we will get
less energy conversion. And hence because of these high frictional losses the efficiency will
be low. So if we take this example, it is not just for pumps, it is also valid for fans, blowers,
etc.

So let us take where we can say this is useful, the high blade angle, is it at all useful in any
case or not? Let us take 2 examples. One of a fan or a blower which is used in mines or any
such heavy industries. So now we need a high volume flow rate, we also need in order to pay
less money for this electricity consumption, the blower to be more efficient. So in such a
scenario, it is prudent to use the β2 value which is less than 90 degree or in other words
backward facing vanes.

However imagine that you are in one of these shopping malls where the shops are actually
paying for the space occupied, i.e. per square feet area or per square metre area as the case
you want has to be paid. So in that case we really cannot afford to have a large space being
occupied by fans. There are wind curtains where you see as you enter you get the pleasant
wind. So in such cases what is required is the space should be small, the size of the fan
should be small and we have already seen that for the same work output the size can be small
if we take β2 to be large.

And so it may be possible in some cases to use a fan with a forward facing vane or a pump
with a forward facing vane, you if you come out with some example but it is generally not the
chosen practice. In practice, we use a backward facing vane for the pumps and the fans. In the
case of pumps, most of the times we use a backward facing vane and not a forward facing
vane. Now the next thing is the degree of reaction. This is the degree of reaction as you have
studied in the last class that U2 square - U1 square +, note that it is W1 square - W2 square
whole divided by the total energy transfer which is W bl infinity. And, since we have
assumed that it is an axial flow pump, U2 is equal to U1 and so this term drops off. And we
can write the W1 square from the velocity triangle is nothing but C1 square + U square, W2
square is nothing but C1 square because I have assume that C1 equal to Cm1 equal to Cm2,
so it becomes Cm square + Wu square which is nothing but C1 square cot β2, this angle is β2
and this is Cm1 equal to C1.

And we have C2 square equal to C1 square + U - U1 cot β2. That is I have assumed that Cm
remains constant and that value is given as C1. So when we derive this, we can say that it
becomes W1 square - W2 square whole divided by C2 square - C1 square + W1 square - W2
square. I can substitute these relationships which we have obtained here and get the
relationship in terms of U, C1 and β2. And we get that R becomes equal to U + C1 cot β2 by
2U.

(Refer Slide Time: 42:01)

And now I have assumed a special case. This is not required but just to do a small calculation
and see how the degree of reaction varies, I can say that Cm2 equal to 0.5U and hence I can
write the W bl infinity is U square whole multiplied by one - half cot β2 and R is nothing but
one + half cot β2. I choose this as an example to show how the degree of reaction varies. I
have shown that the degree of reaction actually goes to 0 at an angle slightly more than 150
degrees.

And you see the degree of reaction reduces, as β2 value increasing. So this is not desirable
and hence we will try to avoid it and we try to give the outlet angles smaller than 50 degrees
or so or close to that also. So then we can say that as β2b increases, blades become more and
more curved and final results in highly cambered impulse profile.
(Refer Slide Time: 43:11)

Usual values of impeller blade angles at the suction side β1b is given as, for pumps it is
between 16 to 20 degrees, we have to also consider something called cavitation. I will talk
about cavitation and cavitation in pumps towards the end of this week. And for radial
compressors we can have it between 28 to 34 degrees, we do not need it at this stage and for
fans and blowers. So these are the typical values we normally use, I give it in one place so
that you can have a comparison. So typically we are talking about for pumps is between 16 to
20 degrees.

(Refer Slide Time: 43:50)

We have discussed so far the effects of the blade angles on the pressure rise inside the pump
impeller in terms of degree of reaction and we have also discussed the blade specific work
that can be obtained and that how that varies with blade angle. So let us look at the
significance of the impeller blades on the performance of the pump. Let us review this point,
we said that as the blade angle increases, the impeller becomes more and more curved, the
degree of reaction reduces and in fact we have shown that at higher values the pump can
become impulse type.

But as we have discussed in the last lecture the impulse type of pump is not possible, is not
feasible, does not make sense because we want to increase the pressure rate and hence this is
not a very good practice. Hence in all practical applications, the impeller blade angle at the
exit is not greater than 90 degrees. Or in other words we talk about the backward facing vane
as a desirable choice of blade. Thank you.