Video Lecture in Engineering Fracture Mechanics, Prof. K.

Ramesh, IIT Madras 1

Module No. # 01, Lecture No. # 02: Spectacular Failures

Further details on Fatigue test

In the last class, we had seen what

is the brief outline on the course on
Engineering Fracture Mechanics.
Then, the concepts from strength
of materials were reviewed. The
idea that was emphasized was,
whatever the concept that you had
learned, are very useful, but the
knowledge you gained is limited.
And, when you want to
characterize the material, one of
the simplest test that was done
was, tension test.

The focus on the tension test was,

you have an elastic region,
followed by a large plastic zone, then only the material fails. But once people had moving
parts in their design, like locomotives started failing, people felt a simple tension test is not
sufficient to characterize the material completely. Something more needs to be done.

So, this prompted the use of designing a new test, what is known as a fatigue test.

And we had also looked at, the focus was to find out, apply a repeated loading on the
specimen and the test was confined to measuring only the number of cycles, when the test
specimen fails eventually. So, you do not record any other information in between. So, it is a
improvement from a simple tension test. In service condition, you have repeated loading.
So, you have been able to simulate a repeated loading and also perform a test until the
specimen breaks. So, you collect voluminous data and this data is presented, in a nice form,
for you to process.

(Refer Slide Time: 02:13)

And, you have a log-log plot of the

stress applied and the number of
cycles to failure.
Video Lecture in Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 2

(Refer Slide Time: 02:20)

In fact, arriving at, what should be

the way x axis to be plotted and y
axis to be plotted is not simple and
what you see here is, you have
scatter of data. This is an indication
of an experiment well performed.
So, in an experiment, normally, you
will have scatter of data and you
take your time to make a sketch of
this graph. Whatever the data that
you have got, you need to make
some sense out of it. You know, if
you look at the genesis of concept
of stress as well as strain, it started
from performing load elongation curves.

Then, instead of plotting as load elongation curve, if you plot it as stress versus strain, you
will have just one graph for a given material. So, if you look at any development in
engineering, collecting data is one aspect of it; reporting data and trying to find out a
meaning from the data, is equally challenging. So, in the case of a fatigue test, they decided
to plot a log-log graph between the life, in number of cycles and you have the y axis is,
whatever the amplitude of the alternating stress divided by the ultimate tensile strength.

(Refer Slide Time: 03:35)

And what you find here is you are

in a position to draw a line like this;
that gives the least expected life N
for a given alternating stress S.
And, you also have from a design
point of view, a convenient
parameter called the endurance
limit. So, what you find is, if you
load the specimen, only to the
endurance limit, it gives an
impression that, you will have
infinite life without any problem. It
is very convenient parameter from
design point of view. See, the
modern research tells, there is
nothing like endurance limit for any
of the material. Every material fails after sufficient number of cycles.

And you will also have to appreciate one more aspect, when you have got this endurance
limit, for which you have to perform a fatigue test, there are also attempts to find out an
empirical relation, involving the result of a tension test; and what you find here is, the
endurance limit is given for this rod steel as 0.5 times the ultimate tensile strength.

So, what you find here is the development of S-N curve for designing structures with infinite
life came into vogue then. So, that is an improvement. From tension test, you have
graduated to simulate what happens in actual service condition, if not completely, at least
Video Lecture in Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 3

very close to simulating alternating loads on the structure, because when you do a fatigue
test, you go for a sinusoidal loading.

If you really go to actual service condition, you will have variable amplitude loading, which
could be using Fourier series; you will be able to convert that, as a series of sinusoidal
loading. And, there are ways, fatigue itself is a vast subject, there are ways how to count the
number of cycles, in what way you have to go about, so on and so forth. But the important
aspect here is you have been able to improve your design methodology, by bringing in the
aspect of alternating load. Definitely this approach, to some extent, has recognized the
failure due to crack growth.

(Refer Slide Time: 06:23)

We have already seen how the test

was performed. The test was
performed only to record the
number of cycles. So, there was no
provision to monitor the growth of a
crack or predict the remaining life.
So, this we look at, from the point
of view of fracture mechanics.
Now, we know crack is very
important, it grows; growth has to
be understood properly. Then you
say that, you will have to find out
even the data pertaining to crack
growth, but considering the tension
test, definitely fatigue was an
improvement.

So, obviously, for damage tolerance approach to be effective, one needs to device more
comprehensive tests, that give information on crack growth behavior. So, this is very
important, without which you will not be able to do a damage tolerance approach. So,
anticipate that, we will have to have a new test, if we have to practice fracture mechanics.

Review on conventional design methodologies

(Refer Slide Time: 07:39)

And before we go into that, let us

look at what we have been doing,
in conventional design. The
conventional design approaches
are done with the premise that, the
material is an elastic continuum
without any material defects or
flaws. See, this is how you can
proceed, because even if you read
mechanics, you start from rigid
body mechanics, then, you
graduate to deformable solids.
When you go to deformable solids, what do you do? You dont directly take up non-linear
Video Lecture in Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 4

behavior, you idealize the deformation as small and it is also linear, it makes your life lot
more simple.

On the similar vein, when you have started modeling the material behavior, it was
convenient, purely out of convenience, you consider that, it is an elastic continuum. And
what are the design approaches you had? You have one approach, design based on
strength. There is another approach, design based on stiffness.

(Refer Slide Time: 08:53)

So, in design based on strength,

what do you do? You would
determine the cross-section and
choose a material such that, the
stress at any point in the structure
does not touch the yield stress of
the material. Because we have
already seen, in the case of
conventional design approach,
yielding was considered as a
failure. And I have also mentioned,
most of the mechanical and
aerospace components serve well
below the elastic limit.

So, its a good starting point. So,

you limit your stresses, such that, they are well below the yield stress. And one of the other
important aspects of conventional design approach is, you dont operate with the yield
strength to start with. You bring in a factor called factor of safety, because there may be
mistakes in calculating the service loads; there may be over loads; there may be material
defects. So, to take care of all these aspects, you bring in a concept of factor of safety.

(Refer Slide Time: 09:42)

And you have seen in several

designs, it ranges from 2 to 10. So,
this is always used to decide the
maximum stress allowed, to take
care of improper modeling of
service loads, material properties
as well as production methods. And
if you really look back, you know, if
you have this cable ropeway cars,
where you have very difficult
service condition and safety is very
very important and its also
exposed to gusts of wind and you
have factor of safety of the order of
10. On the other hand, if you come
to our Ambassador cars, this was
operating with the factor of safety of 4. On the other hand if you look at Maruti, its of very
thin sheet metal work and there is also a demand on fuel consumption reduction. So, they
Video Lecture in Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 5

want to bring down the dead weight. So, they bring down the factor of safety. When you
bring down the factor of safety, your analysis has to be more and more precise.

So, you have to have a tradeoff between cost for analysis and the gain you get out of it;
obviously, the aerospace structures, if you do the analysis the gain is more. For very
conventional day to day activities, you want to have a foolproof design methodology that
would give a reasonable values on the final cross-section. So that, there is a tradeoff
between detailed analysis and a simplified procedure. So, one of the design approaches, is
you do it design based on strength.

(Refer Slide Time: 11:44)

And you essentially determine the

cross-section. If you find the cross-
section, you cannot modify
because of certain restriction. You
can choose a better material to
withstand higher stresses.

(Refer Slide Time: 12:12)

The other design approach, what

you have is, designs based on
stiffness. You know, for many of
this mechanical and aerospace
components, the satisfactory
performance of the component
depends on deflection or
deformation, whatever that comes
across. So, you limit the values on
the deflection, so that, you find out
what is the geometry or what is the
cross-section that you have to do,
for a given design approach. And if
you really look at, when you have a
shaft design, people go for mild
steel and there it is actually
dictated by the deflection. If you go
by the strength calculation, you will
have a shaft which is of very small dimension, but from deflection point of view, it may
experience a larger deflection.

So, the deflection limits the size of the cross-section. So, you will go for a larger size of
cross-section. So, in the shaft, design is really dictated by the deflection. And also you find,
you use a low strength material for shaft. On the other hand, you have springs. Springs are
made of very high strength material. And if you really look at, you have a compression
spring or a tension spring. In actual practice, it is really experiencing a torsional load. So,
you design it for shear. So, you have to know, you have a done a course in design, you use
different materials for different applications. Shafts are made of only mild steel and when
you have a spring, it is always made of high strength steel.
Video Lecture in Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 6

So, the focus here is, when I have relative moving parts, the deflection or deformation plays
a very important role. Because you have to maintain close tolerance levels and usually
design based on stiffness.

And again it is emphasized, all these approaches to the design are done with the premise
that, the material is an elastic continuum without any material defects or flaws.

Its purely out of convenience. There is no other reason. If they had better mathematical
model and simplified approaches, they would have also incorporated what happens when
you have a flaw. Purely out of convenience, the initial design approaches were confined to
considering the medium as elastic continuum. And this worked well. You could do some kind
of engineering based on this, but what is the situation now?

(Refer Slide Time: 14:51)

See, you have to look at the

modern structural environment is
much more complex in terms of
operating temperatures,
aggressive environments and
types of loading, etc. See, now you
find people go to space and come
back to earth. So, they have to
pass through the barrier to earth
and very high temperatures are
developed. Its about order of 1600
degrees Centigrade. You never
even dream of that kind of
operating conditions earlier and
you have cryogenic engines. So,
you have demands that are far
different from what was initially thought of in conventional design approaches.

And in addition what you do? You also have requirement for optimization.

(Refer Slide Time: 15:50)

And this optimization requirements

have resulted in reduction in the
weight of structures, which in turn
leads to, enhanced operating
stress levels. So, what we want to
do is, we want to take out as much
as possible from the material that I
have put in for making a structure.
So, you want to have a tradeoff.
So, we want to have optimized
structures. So, that means, I
should do more analysis. When I
do more analysis, I must also go
closer to reality.
Video Lecture in Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 7

And obviously, if you ignore the presence of material defects or inherent flaws, it will
definitely lead to spectacular failures. In fact, thats what has happened.

(Refer Slide Time: 16:38)

So, finally, if you want to have a

successful design of engineering
structures, for long term life,
requires many things. It requires
understanding of the different
modes of failures and degradation
mechanisms. See, this was never
understood in conventional design.
Only in modern design, you also
look at, there are degradation
mechanisms. So, this
fundamentally, makes a difference
in damage tolerant approach. So,
you have to understand the
degradation mechanisms and what are they?

(Refer Slide Time: 17:14) Crack

growth due to service loads,
problems due to corrosion and
also special problem due to
material becoming brittle. One of
the important causes is hydrogen
embrittlement.

And you know, in fuel cells, one of

the major drawback is the structure
becomes brittle because of
hydrogen entrapment and you
have nuclear power for these days
and the major problem there is,
damage due to irradiation. So,
these are all degradation
mechanisms. Unless you understand the degradation mechanisms, you will not be able to
provide sufficient margins against these mechanisms. So, if you know that, you could
incorporate that in the design phase itself. So, the idea is, the modern design will have to
address the issue of degradation mechanisms. Before we really take up fracture mechanics,
let us look at the spectacular failures.
Video Lecture in Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 8

List of spectacular failures

Why they are called spectacular failures? You know, we are going to look at the key failures
that triggered the engineering community to sit back and evaluate the existing design
procedures are listed. They were very very important failures. So, one famous failure was
Boston molasses tank failure. This happened in 1919. In fact, I am going to give you a list of
4 such failures, over a span of about 70 years.

And, what was important in Boston molasses tank failure? And you all know, when you
travel in a train, if you cross the sugar cane factory, you have the smell of molasses. Its not
pleasant to smell. Okay. And certain failures are spectacular because people remember for
years, because of its notoriety.

(Refer Slide Time: 19:31)

The problem with molasses tank

failure was this left a characteristic
smell in parts of Boston, which
remained for several decades after
the disaster. So, people cannot,
even if they want to forget that
such a thing happened, the smell
will remind them. So, it becomes a
very important and you have to go
investigate, what happened, how
this happened, why it happened.
So, this is what made that
engineering community to think
and evolve new design
approaches. Another famous
failure was Liberty ship failure.
This is between 1942 to 1946.

(Refer Slide Time: 20:01)

Exigencies of World War 2 put a

pressure on speedy production of
ships leading to all-welded design.
Prior to that, people had ships
made of riveted design and it takes
quite a bit of time to fabricate a
ship. When they switched over to
welded design, they could quickly
make the ships. They also paid a
heavy price for it.
Video Lecture in Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 9

(Refer Slide Time: 20:35).

What happened was, the ships

failed; there was absolutely no
control on fracture. In fact, we
would see each one of these
disasters, one after another in
elaboration.

(Refer Slide Time: 20:50)

Right now, we only have a short

summary. Later, we will have, also
have a look at, in detail, some of
these failures. The another
disaster which we will look at is the
comet disaster. This happened in
1954. The importance of comet is,
this was the first commercial jet
liner put into service. Until then,
they were all only propeller planes.
Because its a jet liner, several
novel designs were put to litmus
test. And we will see, what was the
problem and how this was
diagnosed and solved. And very
recently, in 1988, you had Aloha
Airlines Boeing fuselage failure.
This really opened up the
accelerated ageing of structures,
due to corrosion. People ignored
corrosion totally in their design. So, we will go one failure after the other. We will look at
what happened in the Boston molasses failure.

Discussion on Boston molasses failure

(Refer Slide Time: 21:49)

So, what happened was without

warning. That is the key aspect.
See, you had seen a tension test;
you take a mild steel specimen or
a ductile specimen and pull it, you
have always noticed, you have a
cushioning; it goes through
extensive plastic deformation
before it fails. So, you naturally
conclude, when we make a
structure out of a ductile material, it
would give you some kind of a
Video Lecture in Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 10

warning before it fails. What happened in the case of molasses disaster? Without warning, in
January 1919, molasses surged over Boston and a frightful flood devastated a vast area of
the city.

You know, you can imagine, tsunami comes and you have sea water engulfs the country.
Now, we hear such disasters. And, there have also been volcanoes, which totally
submerged the cities. All those disasters have happened, where there are no one will
remain to tell you what happened. But in Boston molasses failure, very unusual, you see a
flood of molasses affecting the city. You had 2.3 million gallon Boston molasses tank, which
was only 3 years old at the time of failure. See, this is the pity. You know, normally when you
design a structure, you want it to come for 50 to 60 years. And, you dont expect a failure to
happen in just 3 years. And, the tank was very huge. It was 50 feet tall and 90 feet in
diameter.

(Refer Slide Time: 23:44)

Just to give you a size and this

also gives you, an idea of what is
the kind of devastation that
occurred and this figure, courtesy,
is from Boston library.

(Refer Slide Time: 23:58)

And really look at how the failure

happened. The tank came apart
with a thunderous cracking and an
estimated 14,000 tons of molasses
caused havoc. And, the molasses
travelled with a speed of 40 to 56
kilometers per hour. And, it was
not easy to clean. It took six
months to clean up the place. Not
only this, after cleaning, if there is
no smell, people would have
forgotten Boston molasses
disaster.

(Refer Slide Time: 24:38)

The pity was, the smell was

coming for decades. In fact, the
molasses actually continued to
creep out of the ground and cracks in the sidewalks for 30 years. So, you, even if you want
to forget, you cannot forget this. The smell remained for decades, a distinctive atmosphere
of Boston. And, this shows another picture of how the devastation was and this is again, the
courtesy goes to Boston public library.
Video Lecture in Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 11

(ReferSlide Time: 25:13).

Let us look at what happened. First

thing is, the tank was made of 15
millimeter thick steel plates. Very
thick, you know. 15 millimeter thick
is very very large. So, one should
really realize, I have made out of
very thick plate, why should I worry,
and what happened?

Before the explosion, the tank's

owner painted it brown. Even this is
looked at. When you look at the
failure analysis, people want to find
out, what kind of mistakes could
have happened. What is the color
of molasses? Molasses is brown.
Without knowing that, the owner of
the tank painted it brown. The
result is, even if there is a leak, it
will be very difficult to spot, there is
a leak in the tank. And the tank is
so huge. You cannot view it, unless
you have a contrast in color. And,
in fact, in modern design you have,
what is known as leak before
break. In all nuclear power
installations, it is mandatory; the
heat exchanger tubes have to go through L B B criteria. That is why, I want to emphasize, in
the case of Boston molasses failure, there was no scope for having a look at whether the
tank was leaking. That is one indication, that some reinforcement need to be done or
removes molasses, so that, you bring down the pressure build up and so on and so forth.

And this is very, very important. See, if you look at your Indane cooking gas, it is actually
odorless. People have added that smell, so that, if there is a leak, you would respond to it by
smell and go on to take corrective measures. Either, you go and open the windows, do not
switch on the light. Some corrective mechanism is possible, if you are alerted. So, in all
nuclear power installations, L B B criterion is very, very important. And, what happened?
The failure investigators found out, that the possible cause was a sudden temperature
change, as the temperature on the previous day was minus 17 degrees Centigrade and on
the day of occurrence, it was 4.5 degree Centigrade. So, there is a thermal shock. So,
thermal shock precipitated. And what they concluded? See, in conventional design,
whatever they dont understand, they put it as a factor of safety.
Video Lecture in Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 12

(Refer Slide Time: 28:05)

So, it happened in 1920. So, in

those days, people concluded
finally, the design was found to be
inadequate to withstand the
pressure created by expanding
molasses and the factor of safety
used was considered to be low. And
this is too trivial a conclusion,
because they were used to bring in
a factor of safety. So, they
concluded, it failed. So, something
has gone wrong. So, increase the
factor of safety. So, this is not going
to solve the problem, because you
are not really looking at the root cause of the problem. It is only a cosmetic changes in your
design approaches

Discussion on Liberty ship failure

(Refer Slide Time: 28:44)

And people really burned their

fingers in Liberty ships. And look
at, over 5000 Liberty ships and T-2
tankers were mass produced
during World War 2. See, this is
very important. It appears in the
history of ship building. Nowhere
you find, a same design was used
to produce such a large number of
ships. If the ships were successful,
no one would have worried. But
you find, the failures were
astronomical. You know, you have
to be in shame, because you
know, the whole scientific community was clueless, why such failures have happened. And
you look at the failures. Of the 5000 ships, thousand suffered significant failures, and in fact,
some of them broke into 2. Imagine a ship breaking into 2. It is made of ductile material and
it broke into 2 all of a sudden. See, that is the most dangerous part of it.

So, between 1942 and 46, thousands suffered. And, because of low temperature, is what
they were able to identify at that time. And, between 1942 to 52, they had also improved
some of the methodologies and 200 suffered serious fractures in this time frame. So,
definitely you will have to look at and find out what has happened. And observation was, the
failure rate of the welded ships was statistically astronomical in the North Atlantic, while
literally, nonexistent in the warm waters of the South Pacific.

So, they understood, yes, this is something to do with low temperature. Low temperature is
one cause and you also have welded ships. Because previously ships were made of riveted
joints and when they moved over from riveted joints to welding, first thing they observed
was, there was weight reduction; very quick in fabricating the ship; I was told that, within a
Video Lecture in Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 13

week they could fabricate one ship. This is how, they have achieved, because the war
exigencies required more and more fleet to be generated at a short time.

But the result was, you find, there were many failures, that needs to be looked at and the
cause needs to be ascertained.

(Refer Slide Time: 31:33)

And its also mentioned that, no

one know exactly how many ships
just disappeared in North Atlantic.
The engineering community was
clueless that low temperature can
cause brittle fracture of steel. In
fact, the navy formulated a
committee, a board, to go into the
details and find out what really
caused such failures, and come
out with recommendations. In fact,
I have access to that report, thanks
to the Project Liberty ship program
and I would read the foreword that
would be more revealing. I would
just read from that. See, the
foreword starts like this, early in the war, welded merchant vessels experienced difficulties
in the form of fractures, which could not be explained. The fractures, in many cases,
manifested themselves with explosive suddenness and exhibited a quality of brittleness,
which was not ordinarily associated with the behavior of a normally ductile material, such as
ship steel. So, they formulated a committee.

And, the committee was given a dictum, what they should do. And, thats also very
important. You know, its very nice to see, how people have looked at it. See, there are two
ways when a failure happens, you just condemn the scientists are so useless; whatever the
funding we give to them, goes down the drain and condemn them and go with your life; this
is one approach, which we commonly come across in India, because you know, when you
have GLSV failure, people only first laugh, they dont look at, there are genuine problems,
when you are dealing with modern technology.

So, people have to be sympathetic and provide proper way of analyzing data and come out
of the problem. And imagine, out of the 5000 ships, 1000 ships break. They had serious
fractures and some of them broke into 2. It is really a sort of a blot on the engineering
community which designed it. But what you will have to appreciate is, the fractures were
very systematically analyzed, we would also see some of that graphs, and useful data was
collected. In fact, it is the naval research board, which funded fracture mechanics research
initially.

So, you find fracture mechanics grew from such a study. And, the board was given this kind
of dictum, to investigate. That also I will read. The secretaries directive to the board, read in
part, as follows: make a complete investigation of the matter, hereby submitted and upon the
conclusion of its investigation, will report the facts, establish there by. And I would say, it is
really a scientific approach. When you face failures, investigate, find out what is the cause,
so that, you dont repeat it again; rather than ridiculing failures. Because failure has always
been a method for inventing new ideas. Without failures, nothing has happened. If you really
Video Lecture in Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 14

look at the bridge design, people built a suspension bridge. After a few months of operation,
it violently vibrated and the whole bridge collapsed. Then people realized resonance is very
important.

So, the key is, when you face failures, analyze it systematically. And, this was done a very
nice scientific analysis. And the dictum reads like this, if the facts establish the existence of
defects, in their designs of or in the methods being followed in the construction of such
merchant vessels, which in the opinion of the board adversely affect the sea worthiness
there off; the board will also submit its recommendations, as to the measures which should
be taken to correct such defects. So, its very clear. You know, people have realized, that
design is one aspect, the construction is another aspect and how well the construction
practices are enforced; because when you are doing welding, you know welding is a very
tricky manufacturing process; and you are talking about way back; in those days people
have not fully understood the welded defects. So, the board was very scientific in its
approach, open minded and carefully collected voluminous data and beautifully
characterized it.

(Refer Slide Time: 36:31)

And if you really look at, the whole

analysis started when you had this
T-2 tanker Schenectady broke into
two in the warm waters of, in the
cold waters, I should not say warm
waters, in the cold waters of Swan
island and this is the tanker. This
ship broke into two. You see the
fracture here and this figure is from
Project Liberty ship and what you
had was, the water temperature
was about 4.5 degree centigrade.
Then the problem was taken up
seriously. And, this shows the
fracture through the deck. You can
see how the entire deck has
fractured. Its a very serious
matter.

So, what you find is, these

calamities focus attention on the
fact that, normally ductile mild steel
can become brittle under certain
conditions. So, if you really go to
metallurgists, you know, they have
understood now, there is
something called ductile to brittle
transition. We will have a look at it.
There is nothing like a material is
always ductile. Above certain
temperatures, it may behave in a ductile fashion; below some temperature it may become
brittle; that kind of an understanding came about.
Video Lecture in Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 15

Ductile to brittle transition temperature and its relevance

(Refer Slide Time: 38:07)

And this is what you see in ductile
to brittle transition. And it is
reported that, fracture toughness,
which you would learn in this
course, this was found to vary for
ferritic steels over a small
temperature range. At low
temperature, steel is brittle and
fails by clevage. At high
temperature it is ductile and fails
by microvoid coalescence or
plastic collapse. And, this is how
you have, you can make a neat
sketch of it. So, you plot
temperatures versus fracture
toughness and the graph is like
this.

(Refer Slide Time: 38:07)

So, in this region, the material is

100 percent ductile; below this
temperature, which is labeled as
NDT, the expansion is nil ductility
transition temperature. So, you
find there is a transition zone,
from ductile to brittle and what
people have found is, I had
already mentioned that, there are
degradation mechanisms,
particularly in nuclear power
plants. People have to
understand, how to safely guard
the nuclear power plants; what
they found was, when the
structure is exposed to radiation,
there is a drastic change happens
on the nil ductility temperature.
And people have understood and
call it as a radiation shift.

So, what you find here is, the

whole graph shits to the right. So,
this is called radiation shift and
you find the nil ductility transition
temperature has increased. So,
that means, over a period of time,
the nuclear power installation
degrades because of exposure to radiation. This aspect needs to be taken into account at
the design phase itself; that is what is important.
Video Lecture in Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 16

So, the Liberty ship failure has shown you very important things; that there is a concept of
changing from ductile to brittle; there is a temperature, transition temperature and a related
concept is radiation introduces a sort of embrittlement. So, what they do is, they keep test
specimens in the reactor and take out periodically and test its ductility. It is done. That is
why you have robots are used and you have, because its all nuclear, you have exposure to
radiation. In fact, robotic technology got developed, purely when they have to handle things
related to nuclear installation, to start with.

So, they have to take out the material periodically, test specimens are kept inside the
reactor, take out periodically and conduct the complete test, without any human intervention.
It has to be done by robotic arms
and tools and so on and so forth.
And this is how you have to
monitor any of those installations.
So, what you find from a Liberty
ship failure is, a ductile material
can become brittle. So, that is the
important learning that you gained
from that. And it all started from the
failure of this ship, and this was
docked in the yard. You know this
is what is important; its not that
this failure has happened in heavy
sea.

(Refer Slide Time: 41:58)

Heavy sea, people would have ascribed this to very heavy loads, that is coming up because
of high waves, and or some bomb has been dropped on to the ship; some such explanation
you can give, but you dont expect a ship kept in the yard to break. So, that alerted, it is
something to do with the structural design and the environment. And if you really look at,
during the first half of the 20th century for typically high grade ship steel, the ductile to brittle
failure transition was only 10 degree centigrade. That is too high. Because the operating
temperature is far below this.
You know, you never had this
kind of an exposure earlier.

If they had stuck only with

riveted ships, cracks would have
formed, but they would have got
stopped in another rivet hole. In
a welded ship, what happened
was, there was nothing to arrest.
So, the crack has gone through
the full ship and fracture
occurred. They would not have
got alerted, transition in
temperature is a major cause.
And there is also data associated
with these failures.
Video Lecture in Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 17

(Refer Slide Time: 43:29)

And as I have mentioned it earlier, below the NDT, you find the ship broke like glass; that
means, without warning, suddenly the ship breaks into two. Its very dangerous; there is not
even time for you to escape out. And, in fact, such failures have happened and it has been
recorded and analyzed; what you will have to look at is, people have analyzed all those
failures.

(Refer Slide Time: 43:59)

And several such graphs are

available and what you find is, the
ship is designed; within the first
few years, you had so many
failures. This is something
unhealthy and this gives for a
period from 42 to 52 and you also
have another set of graphs, which
shows the condition at sea.

(Refer Slide Time: 44:26)

I think you need to make a copy of

this. Because the focus here is,
there were failures in heavy seas
about 154, in moderate sea
condition its about 20, when the
sea is calm, it is about 23. Thats
what alerted the scientists, to look
at critically, what has caused the
failure and if you find this, in this
kind of situation, the temperature is
8 degree centigrade below the
heavy sea temperature. So, this all
came about by looking at critically,
what has caused the failure. So,
from that data, you are able to
segregate and find out, even under
calm sea conditions, failure could
happen and it has happened.

(Refer Slide Time: 45:35)

And finally, you know, you dont have to feel that the design was very bad. You still have a
nice Liberty ship available. Out of the many ships fabricated, two still remain and it appears,
now, you can go for a jolly ride in this. They remember this, nostalgic memories and what is
recorded is, out of the 2751 ships built between 1941 and 45, only two remain afloat. And its
all named on the soldiers who fought in this war, S S John W Brown and another one is
Jeremiah and right now, you have special cruise aboard these ships are arranged, even
now. And if you really look at the modern development, you had seen the ship steels in the
middle of 20th century had this NDT as something like 10 degree centigrade. Now, it has
Video Lecture in Engineering Fracture Mechanics, Prof. K. Ramesh, IIT Madras 18

been brought down to something

like minus 4.5 degree
centigrade; this is one aspect of
it; that means, when there is a
failure, improve the material.

So, people have studied the

material and they have been
able to bring down this nil
ductility transition temperature.
So, the new steels are like that.
Another improvement is you
bring in in-built crack arresters
as part of your design. And what
do these crack arresters do?
They prevent easy crack
propagation. So, thats the way, you have to handle fracture. You have to find out some way
to improve your design and also salvage. When there is a crack, you do not allow it to
propagate easily.

So, in this class what we have looked at was, we started looking at the fatigue test and the
focus was, you only find out the number of cycles for the specimen to fail; you dont take any
note of how the crack propagates while you perform the test. Then, we looked at, what are
all spectacular failures. Of the four, we have been able to see two of them; one is the Boston
molasses failure; another is a liberty ship failure. And what I pointed out was, people were
not detracted by the failures. They took this as a challenge and the failures were
systematically analyzed to cull out relevant data to improve future design. Thank you.