Dynamics of Machines Prof. Amitabha Ghosh Department of Mechanical Engineering Indian Institute of Technology, Kanpur
Dynamics of Machines Prof. Amitabha Ghosh Department of Mechanical Engineering Indian Institute of Technology, Kanpur
Dynamics of Machines Prof. Amitabha Ghosh Department of Mechanical Engineering Indian Institute of Technology, Kanpur
Module - 11 Lecture - 4
So far we have been discussing the effect of excitation, either in the form of the dynamic
force or dynamic moment or some periodic motion to the support of a system. Primary
objectives were, during these studies, what is the response of the system which we
measured or which we defined through the displacement of the main body, which
represents the systems inertia. We will extend our studies little bit in the direction, which
will be extremely useful while designing a system. What we will do now is that try to
analyse, that if we keep a system which is supported by kind of or maybe a combination
of springs and dashpots. If the ground has some motion, what will be the motion
transmitted to the main body?
It is something similar, what we have been doing in our vehicle example, where the road
was wavy and the wheel was moving up and down. The driver along with the car was
also moving up and down, so that is one thing, that means the motion of the support or
the ground gets transmitted to the main body of the system. Similarly, we should also
investigate the situation where a mechanical system is generating some dynamic force,
due to some unbalanced present. Then, how much of that unbalanced force which is
produced in this system, gets transmitted through the combination of springs and
dashpots which we put into the ground. Therefore, what we will try to do? We will try to
design our support, with the help of a combination of springs and viscous dashpots, in
such a way that the system is isolated from the environment. That means, either the
forces or moments generated in the machine when it runs, gets transmitted to the floor
and that way it may affect other precision setups, or if a precision setup is being mounted,
it should be run is such a way, that all the disturbance which are there in the ground, they
do not get transmitted through this. This topic is generally termed as transmissibility and
isolation.
(Refer Slide Time: 02:57)
The main objective of this study will be, how to decide the parameters like stiffness and
dashpot, viscous dashpot, damping coefficient in a manner, so that the system can be
isolated from the environment.
This is the system we have been studying so far, a simple spring mass dashpot system.
Now in this case, we have already solved the problem that when the ground has a motion
like Y cosine omega t, then x is given by (Refer Slide Time: 05:08). This also oscillates
with a magnitude of this. Therefore, we define transmissibility by this quantity, this is the
displacement transmissibility that is X by Y, which is nothing but (Refer Slide Time:
06:06). We have already seen this, we have also plotted this earlier. Same thing it is
actually defined as the transmissibility and it is like this.
(Refer Slide Time: 06:41)
We have already seen transmissibility, if we plot against frequency ratio, zeta increasing
and both at r is equal to 0 and r is equal to 0.2, transmissibility is always 1. That means,
whatever disturbance I am giving, the same amount is being produced here. But if we can
design the system in such a way that r is large enough, we are operating somewhere here.
Then maybe transmissibility will be a small fraction, maybe only 10% of this motion,
which is being given to the ground, is being transmitted.
Let us find out how much force is being transmitted. Force is being transmitted only
through the element like spring and dashpot. So the vector sum of the force at any instant
on the dashpot and the spring is the force, which is transmitted through the ground. We
will again use our convenient rotating force. The spring force is represented by this
rotating vector, the damper force is represented by this rotating vector leading the spring
force by 90 degrees. Of course, inertia force is represented by this rotating vector and all
these three, we know that the whole thing is rotating with fixed rpm omega and the real
values at any instant, as we have been mentioning again and again are actually projection
on this some reference line. If this is omega t, then this will be phi. That means, the
displacement x lags the exciting force F0 y in angle phi. Now, what is the force which is
being transmitted to the ground FT?
We have seen that FT is the vector sum of the spring force and the dashpot force, that
means it will be represented by this rotating vector and at any instant, this projection will
be the instantaneous value. Therefore, let us see how much is the transmissibility? FT is
nothing but, square root of kX square plus the dashpot force divided by F0. F0 is again
this square plus this squared, square root. How much is this? This is kX minus m omega
square X, this square, plus this is again (Refer Slide Time: 13:07). This is the ratio of the
magnitude of the transmitted force, with magnitude of the generated force. If we take X
common outside, it goes out and it becomes (Refer Slide Time: 13:24) and you definitely
have guessed by now, that we are going to get the same, if we divide by k square or take
k square common outside, then it becomes 1 plus c omega by k is 2 zeta r. This is 1
minus m omega square by k is omega square by omega m square that is r square and this
is like this. (Refer Slide Time: 14:15). The expression for transmissibility, both from the
point of view of transmission of motion, vibration, or transmission of dynamic force
which gives the same expression we find.
Therefore, it is a very important result of the study of vibration is that, we can isolate
systems and machines from the environment, so that neither this unwanted disturbance
come to our system, when we are a handling very precision experiments, or the ground
does not get too much of disturbance from the machine on which I am working, suppose
a very heavy forcing machine or machine tool, which is generating a large force, then a
smaller amount is really going to transmitted to the ground. So, the ground will have
lesser vibration produced by this transmitted force.
This technique of isolating a system from the environment is called isolation. The
quantity which dictates to what degree one system has been isolated is called the
transmissibility and for simple systems like this, transmissibility behavior is like this.
Expression is like that, best will be perhaps if we solve a problem.
This is a vertical engine, is supported on a concrete block. This concrete block again in
turn is resting on a system of springs, with total resultant stiffness of k and system of
dashpot, total resultant damping coefficient c.
The unbalanced force generated in the engine is given by this expression (Refer Slide
Time: 17:14), where n is the rpm of the engine. This is a simple harmonic force, we are
assuming that is a primary unbalanced force. That is why it is 2 pi n by 60 t. That means
it is omega into t. That means, it is connected with the rotational speed of the engine and
therefore it is clear that it is the primary key. The magnitude of the force is also
increasing with force, obviously because the centrifugal force or unbalanced forces will
always be increasing in magnitude, not only in frequency when the rpm increases. This
tells us the approximate representation of the generated force.
Now, we say that at 1000 rpm, the magnitude of the transmitted force, which is FT is
equal to this 100 Newton. What will be the transmitted force at 1500 rpm? If the question
is this, that given the information is how in rpm of the engine, the magnitude of the
transmitted force is 100 Newton. Then, if the engine runs at 1500 rpm, what will be the
transmitted force? Now, we have been given the transmission ratio or transmissibility at n
is equal to 1000 is equal to 100 is the transmitted. How much is generated at 1000 rpm?
At 1000 rpm the magnitude of the generated force will be 100, so 100, equal to 1. Now,
what does it mean from this characteristics? The transmissibility can be 1 only at r is
equal to 0. That means the frequency 0, which is not possible because the engine is
running at 1000 rpm. The only other possibility is that when r is equal to root 2,
transmissibility will be n equal to 1000 rpm corresponds to r is equal to root 2.
We can find out the r, that means, omegan is equal to actual operational frequency omega
by root 2 is the natural frequency. This we know how much is the omega. It is 2 pi into
1000 rpm divided by root 2 into 60. This gives me the radian per second is the natural
frequency.
Therefore, let us find out r prime. r prime will be omega by omegan is equal to, how
much is omega at 1500 rpm? This will be 2 pi into 1500 divided by 60, divided by
omegan. Omega n is this much (Refer Slide Time: 23:04). It will be 15 root 2 or r prime
square equal to 4.5. Now, one point which was missed, when we mentioned this as such,
if we just tell r prime, we will not be able to solve the problem, because c is not known.
Therefore, the only possibility will be in this case, we can find out the transmitted forces
with 100 rpm if the damper is disconnected.
In the first experiment, it was not necessary to disconnect the damper, because at r is
equal to root 2, the effect of damping coefficient or damping factor is not present. It is
irrespective of that the value of zeta which is 1. But when you deviate from this, then the
transmission which depends on the damping coefficient information on which is needed,
but in this case we have not given that information. Therefore, the problem is actually,
what we can find out is that, when the damper is disconnected?
(Refer Slide Time: 24:51)
Similarly, for mounting very accurate equipment or instrument and such devices, when
they are also isolated, so that disturbances present in the ground do not come to the
instrument. This brings more or less to the end of our discussion on single degree
freedom damped systems. Only thing what we have not done, but as mentioned earlier
that handling problem, where a system is subjected to non harmonic but periodic forces
or disturbances. We can split the non harmonic, but periodic disturbance into a series of
harmonic disturbance. As the systems which we are discussing are always linear, we find
out the response of the system, because of each harmonic. The result will be the sum total
of the FT.
One, very common problem faced by mechanical engineering is the shafts carrying discs.
A shaft like this is rotating with a speed omega. Now, whether there is any potential
problem, the dynamics of rotating bodies itself is a subject. We will not be able to discuss
it there, but the simplest possible case we would like to take up, because it has lot of
similarities in all kinds of analysis we have been doing.
What are the possibilities, when we mount the disc on the shaft, the center of the disc of
the CG will not be exactly at the axis of the shaft. If this is the shaft, the CG maybe
somewhere here, center of the shaft we call it as o prime. Again because of the
centrifugal force which is generated due to these rotated body and which is unbalanced,
the shaft will also get bent and the axis that means, real axis rotation if we join this. That
means shaft will be bent, this is the point about which the whole thing will rotate, we will
take up the simplest possible case where the deflection of the shaft, then the location of
the G and the center of the disc, so that the relative position remains same. That means,
the bent shaft and the disc, they all rotate like a rigid body. This particular type of
situation is called forward synchronise precision. It occurs very often, let us see what we
can do about them.
Enlarging this figure, this is o, this is rho, this is e, this is G and this point is o prime.
Forces are k rho, c omega, this angle is theta and these also let us consider if this is the
rotation from time t is equal to 0, that means when o prime G is parallel to this direction,
that is the reference line. So if t is 0 there, this angle which makes now with that direction
is omega t. Remember, theta and omega, though they are not same, but the whole thing is
moving like a rigid body. Therefore, we should keep in mind that theta dot is also equal
to omega. If we split everything into two components, x and y, the coordinate of the
center of mass of the disc is the xG and yG and its accelerations will be xG two dot and yG
two dot this component of acceleration.
Therefore, equation of motion we can write that mass of the disc xG two dot must be
equal to the total force in the x direction acting on it. How much is that? One is c omega
rho. What will be the component in this direction? This angle is theta, this is omega t so
omega t minus theta, therefore this is 90 degrees minus theta, because this is 90 degrees.
Since this is theta, so it is 90 degrees minus theta. Therefore, it will be sine theta minus
the component k rho cosine theta. Similarly, in the acceleration component in the y
direction is equal to minus c omega rho (Refer Slide Time: 38:24).
We can also represent xG and yG. This xG is equal to xoprime plus e cosine omega t and yG
yoprime plus e. This will give me xG two dot is equal to xoprime two dot minus omega square
cosine omega t and yG two dot equal to yoprime two dot minus omega square e sine into
omega t.
(Refer Slide Time: 40:00)
Substituting in these equations we get (Refer Slide Time: 40:00), c omega rho is nothing
but velocity of this point. If we multiply by sine theta, what happens, it is the x
component of velocity of o prime point. It is nothing but c into xoprime dot.
Its velocity is in this direction, velocity is omega into rho. It has two components, this
component is nothing but xoprime dot and c comes there. Similarly, the other one will be yo
prime dot in this direction, when it is omega rho e, c omega rho cosine. So, this is this
(Refer Slide Time: 41:40) minus k and rho cosine theta, this is rho and this is theta, rho
cosine theta is nothing but x component of this point o prime (Refer Slide Time: 41:59).
The other equation becomes, I am substituting yG two dot by yoprime two dot minus omega
square e sine omega t and that is equal to omega rho cosine theta. Here we have found it
is omega rho, omega rho cosine theta is y component. The positive direction is still, so
minus sign will remain and rho sine theta is yoprime, which is minus k yoprime. This can be
written like this mxo prime two dot plus cxo prime dot plus kxoprime is equal to, I take this to
the other side (Refer Slide Time: 43:26). This equation becomes (Refer Slide Time:
43:34). Now, we remember our solutions to this form of equation.
We can always write x oprime is equal to, now F0 we have solved this problem mx two dot
plus cx dot plus kx is equal to F0 cosine omega t. Now F 0 is being replaced by this, so
what we will get here, the final solution F0 by k, I think that was the numerator (Refer
Slide Time: 44:26), where zeta of course is here, this cosine. We have to keep in mind r is
equal to omega, that is the rotational speed by omega n, omega n is the circular frequency
of natural lateral vibration of this, which we can keep it as (Refer Slide Time: 45:30) and
zeta is again, we know zeta is c by 2 square root of km.
One thing is very clear, if we want to find out rho square, this is xo prime, this is yoprime,
rho square is equal to xoprime square plus yoprime square. This will give us, rho equal to
(Refer Slide Time: 46:15).
So, we find that the characteristic let us see how it is going to be. For omega is equal to 0,
this rho will be equal to 0, because numerator is 0. As it increases when omega becomes
very large, then r is very large you can ignore this and since this is r to the power f and
this is r to the power two, so we can ignore this. After square root it becomes r square,
that means omega square by omegan square that is, k by n. so it will be e.
If we plot an e rho, when r is equal to 1, suppose zeta is extremely small we can ignore
this. So, denominator is 1 by r square approximately. When r is 0, omega r is equal to
omega by omegan, so omega is 0, means r is 0. It starts here, because rho is 0 at omega is
equal to 0, and then it shoots up when r tends to 1 and then comes down, becomes e. Of
course it is the magnitude I am talking about, if zeta is present, if zeta is increasing; at a
very high speed, the deflection of the shaft will be equal to the deflection of the area of
the center of mass from the center of the actual shaft or access of the shaft. However,
when omega is nearer to omegan, there is a possibility that the vibration that means the
deflection of the shaft will be very large, which is something like resonance what we had.
(Refer Slide Time: 49:46)
That sometimes is so large that the shaft maybe permanently damaged. This particular
phenomenon, when it happens that means at omega is equal to omega n. That means if the
shaft is like this, then the equivalent stiffness of the shaft here is k and mass of the disc is
km. If this is equal to omega square or omega is the rotational speed of the shaft, then
again something like resonance takes place. This phenomenon is called whirling of shaft.
Sometimes when it is found that the critical speed is low and the operational speed is
high. For example, like the alternators in thermal power plants. The shaft rotates at a
higher speed than the critical speed. In such cases to avoid whirling and damage what is
done, the shaft is quickly accelerated though the critical speed, so that it does not get
enough time to build up large vibration and large deformation of the shaft. This is the
simplest situation is very similar to what we get in case of a simple spring mass system
behavior, when some damping is there. We have not gone into the details of finding the
phase angle and all those things. It can be easily found out without any problem.
But the main thing what one has to keep in mind is that the shaft has certain space at
which, the system resonance causing very large deformation of the shaft from its nominal
state position and these speeds are called critical speeds. The design should be such that
they should be avoided. At very large rpm, when you operate the deformation of the
shaft, will be the amount of error which was there while fabricating the system, that is the
e or eccentricity of the center of mass.
I think with this we more or less complete most of the things, which you are supposed to
take care of in a simple system, which is spring mass and dashpot parameters. It is now
important because a very large number, in most of the systems case, it is not possible to
express or represent the system with the help of a simple mass. Therefore it is absolutely
essential for us to now investigate, what happens when the numbers of masses are more
or the degree of freedom of the systems become more or the system becomes more
complicated. Subsequently you will devote your attention to systems which higher degree
of complexity, in the sense that the number of degrees of freedom will be more.