3B4 Lead Creep

Eamonn Slattery

12313584

Group 1 Lab Date: 24/03/2015

Abstract
This purpose of this experiment is to examine the material properties of lead with respect to creep.
A standardised test was carried out on a test piece of lead with varying loads and several constants
were extrapolated from the data to accurately represent the mechanical properties of lead
mathematically. During the test it was found that lead will quickly undergo creep after its initial
displacement under a load at room temperature. With increasing load it was found that the creep
rate increased and the steady state creep period (the length under which a material experiences
creep before failure) was shortened.

Introduction
Creep is the slow, continuous deformation of a solid over time where the strain becomes dependent
on stress, temperature and time as opposed to just stress.[1]
For most metals and ceramics, room temperature is considered to be low where their strain can
be represented as:
= ()
However, for some materials such as lead, room temperature is considered hot as it is ~300K
which is half the melting point (600K). This means that at room temperature lead behaves like a
material that has been heated up to a point where it becomes soft. At this softened state a material
is susceptible to creep and its strain depends on the
variables mentioned above:
= (, , )

To be more specific about when a material is susceptible to

creep other than just being considered hot or cold
there is a general rule for defining the temperatures at
which materials will begin to creep. Generally:
> 0.3 0.4
> 0.4 0.5

Figure 1: Bulb filament failed due to creep. The

high temperature caused the tungsten filament
to undergo creep under its own weight

So for a material such as lead room temperature is over 0.5

times the melting temp which means that it is very much in
the creep range. Other materials which can undergo creep
at room temperature are tin, polyester and acrylic.[1]

Creep is an important phenomenon for many areas of engineering such as forensic, design and all
the other core disciplines involving the mechanical use of materials. When designing a system an
engineer must ascertain the operating temperature that the material will work under and deduce
whether that material will be subject to creep deformation. If the material is found to then be
susceptible to creep, the designer must calculate the amount of time that the material can
withstand under the applied load before it will fail due to creep. This time range is known as the
steady state creep and can vary from seconds to years depending on the material and applied load.

As creep failure is dependent on 3 variables,

temperature, time and stress engineers generally
plan work cycles and temperature limits to control
creep.
An example would be a preheating tower and kiln
in a cement processing plant. The inside of the
pipework is lined with ceramic firebricks to
insulate the steel outer from the intense heat,
without this insulation the heat (while not high
enough to melt the steel) would increase the
creep rate so high in the structure that it would
fail under its own weight if it continued operating.
Temperatures are also monitored continuously to
ensure that it stays at a level low enough to
prevent the onset of creep deformation.

Another example is the fire coating that steel

girders are given in skyscrapers. This coating is
applied to the steel even though a fire would not
reach a heat high enough to melt the steel
structure. A fire would however be enough to heat the steel to above its creeping temperature and
the massive self-weight load of a skyscrapers steel lattice would cause the structure to fail after a
short period of steady state creep.
Figure 4: Preheating Tower and Kiln, similar to the one
located in Platin, Drogheda. The extreme temperatures
require a strict maintenance regime to minimise the risk of
creep failure.

Figure 2: Steel with insulation sprayed onto its surface.

Figure 3: After a building fire this steel structure failed under creep.
Whereas the burned wooden beams were left supporting the steel.

Theory
Creep happens in a material when an elevated temperature promotes diffusion whilst the material is
subjected to a tensile load that is lower than the materials yield strength. The diffusion allows the
atoms inside the material to rearrange when a stress is applied, this in turn allows the dislocations in
the material to move past obstacles in their path and as a result the material begins to deform even
though the applied load is under its yield strength.

Figure 5: Close up of crystal grain structure (Creative

Commons)

When the load is applied the atoms near the loaded

dislocation treat the dislocation like an area of high
concentration and begin to diffuse towards areas of
lower stress. As these atoms diffuse away the
obstruction blocking the dislocations movement is
dissolved gradually as more and more of its atoms
diffuse and eventually the dislocation is free to move
in the direction of the applied force. This process is
called Climb and since it requires diffusion it can
only happen at temperatures above 0.3-0.4 times its
melting temperature i.e. the creep temperature.[1]

The process of creep happens over three stages, primary creep, secondary (or steady-state) creep
and tertiary creep. The initial primary creep is the elastic deformationthat the material undergoes
when the load is applied, there is initially a high rate of increasing strain followed by by a gradual
drop in strain rate. After this the material begins to exhibit steady state creep and the rate of
increasing slope begins to level off towards zero before gradually increasing over a period of time
() . Finally at the tertiary creep stage the strain rate accelerates quickly to the point of failure in
the material
Both the primary annd tertiary stages are
not of much importance to Engineers as
the first stage is mainly elastic deformation
and at the final stage the material is
considered to have all but failed. The
important region is the steady state creep,
as this gives an Engineer the information
he needs to work with and design for. By
knowing how long a material will undergo
creep for when subjected to a certain load
the designer can plan tolerence limits or
work cycles to his designs to ensure that a
product will not fail when used correctly.
In order to analyse the steady state creep behaviour of a material, we use a standardised test on a
specimen with precise dimensions and constant temperatures in order to plot the steady state creep
rate over time.

By then plotting the Log of the steady state creep against the Log of the stress we can establish that:
=
Where A = the creep, and n = the creep exponent.
This sort of creep is called the power-law creep
The test equipment used a lever with a mechanical advantagein order to impart a tensile load on the
test material. Taking into account the beams self weight, the nagers self weight and the mechanical
advantage we found the stress equation to be:
=

(( 8) + (0.4 4))

The Log of that stress value is then plotted against the the Log of the steady state strain for three
different weights and a line is fitted. The equation of the line is thus:
= +
Where LogA is the intercept and n is the slope. We can then use these two values as the material
constants for lead in the previous equation =

Materials & Method

Rig
The setup for this experiment was a standardised lead test piece inserted into tecquipment machine.
The tecquipment machine is a basic rig that uses an
8:1 mechanical advantage lever arm to pull a test
piece apart between two pins. A hanger at the end
of the lever allows variable sized weights to be
hung from the rig.
At the other end of the hanger, above the test
piece and lever arm, the LVDT is placed with the
probe touching the top surface of the arm. As the
test piece creeps the lever arm moves upwards and
the transducer registers the movement on an
external computer.
The computer reads in the data from the
transducer and updates a sample every second.
After the piece is loaded in the machine the
computer program is activated and runs a test every second. After the piece has failed the program
is stopped and an output of each displacement per second is output.
Figure 6: The tecquipment machine with a test piece just
after failure

Test Piece
The test piece is a cut piece of lead sheet 2mm thick. The thin section
is 5mm wide and 19mm long between both chamfers. The chamfers
remove stress concentrations and ensure that the test works over
the 19mm long straight section.

Procedure
1. The pre-fabricated test pieces are carefully loaded into the tecquipment machine over the
two dowels. Care is taken not to bend the pieces as lead is soft and easily deformed
plastically, which would affect the results.
2. A 1kg weight is carefully loaded onto the hanger at the other end of the rig, care is taken to
ensure the initial loading on the test piece is gradual as even a dynamic load could fail the
piece.
3. The computer program is started and begins to sample the displacement read from the
transducer every second.
4. Eventually the test piece fails under creep, the computer program is stopped, the piece is
removed and the rig is reloaded with a new test piece.
5. The procedure is then repeated with 1.2 and 1.4 kg weights.
6. The results from the experiments are post edited in excel and graphs are used to calculate
the material constants of lead.

Results
The data from the experiment was entered into a excel spreadsheet and the data was created into
three graphs representing 1, 1.2 and 1.4kgs respectively.

Strain vs Time 1.0kg load

Strain vs Time
0.14
0.12

Strain (E)

0.1
0.08
0.06
0.04
0.02

1
16
31
46
61
76
91
106
121
136
151
166
181
196
211
226
241
256
271
286
301
316
331
346
361
376

0
Time (s)

The graphs map the displacement of the test piece over time, the initial spike is the loading strain
and elastic deformation of the material. It then levels off somewhat to the steady state phase where
we can then get a value for the steady state strain rate. The graph above was the first test at 1kg and
the steady state strain rate was found to be 0.084s-1

Strain vs Time 1.2Kg Load

Strain vs Time
0.6

Strain (E)

0.5
0.4
0.3
0.2
0.1

1
9
17
25
33
41
49
57
65
73
81
89
97
105
113
121
129
137
145
153
161
169
177
185
193
201
209

0
Time (s)

The graph above is the second test at 1.2kg weight. The strain curve is very much like the theoretical
image shown in figure 6. The steady state creep ranges from around 17 up to 160 seconds. After 160
seconds the tertiary phase begins and the piece fails completely approximately 10 seconds later.

Strain vs Time 1.4Kg Load

Strain vs Time
0.5
0.4

Strain (E)

0.3
0.2
0.1
0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57
-0.1

Time (s)

The 3rd graph is the loading test now with a 1.4kg weight. Again the curve is as expected however
the time to failure from start is much more rapid (~50s) the steady state phase in this test only lasts
approx. 30 seconds before tertiary phase and ultimately failure happens.
After the three tests were ran the steady state
strain rate for each case was found to be as
found on the chart to the right:

Mass
1.0kg
1.2kg
1.4kg

Steady State Strain Rate

0.000526
0.001053
0.007368

Strain Rate vs Time

Using the previous values for strain against time we then calculate strain rate vs time. Which is the
rate of change of strain (the slope). Using these graphs we can accurately locate the transitions
between either ends of the steady state region, as the steady state region has very little slope.

Strain Rate vs Time 1.0Kg

Strain Rate vs Time
0.016
0.014

Strain Rate

0.012
0.01
0.008
0.006
0.004
0.002
-0.002

1
16
31
46
61
76
91
106
121
136
151
166
181
196
211
226
241
256
271
286
301
316
331
346
361
376

0
Time (s)

The first graph indicates the transition from primary to secondary creep quite well. It also reinforces
that there was no tertiary phase during the first test which will be explained further in the
discussion.

Strain Rate vs Time 1.2kg

Strain Rate vs Time
0.1

Strain Rate

0.08
0.06
0.04
0.02

1
9
17
25
33
41
49
57
65
73
81
89
97
105
113
121
129
137
145
153
161
169
177
185
193
201
209

0
Time

The second test had good indications of both primary -> steady state and steady state > tertiary
creep. The slope is shown to be near zero from 15 to 160 seconds.

Strain Rate vs Time 1.4kg

Strain Rate vs Time
0.1

Strain Rate

0.08
0.06
0.04
0.02
0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57
-0.02

Time

The final test also showed two clear peaks marking the transition between creep phases. The peaks
in this case are less abrupt due to the short overall time of the experiment, there was only 25% of
the amount of samples as the second test. Slope was near zero for approximately 30 seconds.
LogStrain vs LogStress
stress

steady state strain

9417600
10987200
12556800

0.000526
0.001053
0.007368

log strain steady

state
-3.279014256
-2.977571629
-2.132650383

log stress
6.97394024
7.04088703
7.09887898

In order to find the material constants for lead we finally plotted the Log of the strain against the Log
of the stress. The stress was calculated using the formula from the theory section:
=

(( 8) + (0.4 4))

The strain at steady state was found from where the strain rate was non-changing and the logs were
taken of both results.
Sample calculations for 1kg case:

((1 9.81 8) + (0.4 9.81 4))

=
(0.005 0.002)

The results were placed on a scatter plot and a linear fit line was fitted. The equation of the line was
then extrapolated and used to find the ultimate values of A and n.

LogSteadyStateStrain vs LogStress
0
-0.5 6.96

6.98

7.02

7.04

7.06

7.08

7.1

7.12

LOG STRAIN

-1
-1.5

y = 9.056x - 66.531
R = 0.9078

-2
-2.5
-3
-3.5
-4

LOG STRESS
LogStrain vs LogStress

Linear (LogStrain vs LogStress)

The equation of the line for the fitted line was found to be y = 9.056x - 66.531
Which we can then use the equation:
= +
To find the values of A and n:
A = 2.94410-67
N = 9.0056
Questions
Estimate the maximum applied stress which the material can withstand without significant creep,
defined in two possible ways:
(a) Less than 1% per year.
=
= 0.01
0.01
= 2.9441067 = 3.16 1010
A = 2.94410-67
N = 9.0056
3.16 1010 = 2.944 1067 9.0056
9.0056 =

3.16 1010
2.944 1067

3.16 1010
= .
2.944 10 67
(b) A total time to failure of more than 10 years.
9.0056

0.471053
=
= 1.5 109
10 ()
3.15569 108

9.0056

1.5 109
= .
2.944 1067

Discussion
Strain vs Time
The graphs were produced and appeared exactly as expected with the exception of the 1kg load. For
the other two the steady state creep region was reasonably visible with the curve also indicating
areas of primary and tertiary stress
The first graph may have been recorded in error or there may have been an error in how we set up
the apparatus. The data ends at what appears mid-way through the steady state creep. This may
have been an error in the recording system or an error in setting up the rig. Another possibility is
that the data was set to stop recording automatically when no new data is read after a certain
amount of time. Its possible that the intermittent data was updating so slow that the system may
have thought the material had failed, although as the last reading was a different value to its
previous value this may not have been the case.
Strain Rate vs Time
These graphs give a good indication of the transition of phases and the length of the steady state
creep. The two spikes in the graph represent area of high rate of change in strain and the trough in
between represents the steady state creep where there is an almost zero rate of change.
The first graph further reinforced that an error caused the data recording to end prematurely as
there is no second spike in the graph representing the transition from steady state to tertiary creep.
LogStrain vs LogStress
The scatter plot produced a linear trend line which was used to get a line equation from and then
values of A and n. The value of n was found to be 9.0056 which is within the expected range of 3-10.
After getting the values of our material constants we were then able to answer the two questions
provided. The answers were found to be around 2-3Mpa which is around 20-30% the stress that was
imparted on the material during the experiment. This indicates that at a lower force of 2.1MPa the
lead would not undergo any significant creep and would be able to deform under creep by more
than 10 years before the test would have failed.
Overall the experiment proved to confirm most of the theory regarding creep in materials. The
primary creep was clearly visible on the graph and showed us the material behaving elastically
before the onset of creep. The steady state creep was most visible on the second set of graphs
(strain rate vs time) as a long trough of near zero data between two peaks. And the tertiary creep
was visible as a sharp spike on the strain rate graphs and a sudden rising curve on the strain graphs,
this clearly indicated how the material had begun to fail and was quickly pulled apart in tension after
the dislocations had completely failed in the material.
There was several sources of error that could have affected the results however and some of the
most important are listed as followed:
We did not measure the test pieces before the experiment and have no idea if they had been
machined correctly to the exact tolerances required, it would have been prudent to measure the
dimensions of the piece before starting in order to eliminate this source of error.

If the pieces were machined correctly we have no idea then the conditions they were stored in or
how much they were handled with before the experiment, lead is an extremely soft metal at room
temperature and a person holding the piece could easily have deformed it plastically in their hands.
This would have significantly changed the materials properties under test conditions.
The LVDT was not calibrated before the experiment and as such we have no idea of its accuracy
during the test. Again it would have been wise to check when the device was last calibrated in order
to remove this source of possible error.
Finally the first graph recorded incomplete, this could have been a software error or possibly a
problem with the LVDT but there is no way to know for sure. The process of carrying out the
experiment under 3 different loads minimised this error however and we were still able to get good
data. Had we only carried out the experiment once, the results would have been inconclusive.

Conclusion
The aim of this experiment was to investigate into the phenomenon of creep in materials and
specifically the behaviour of lead, a material that is at its creep temperature at room temp, when
under a load that is well below its yield strength.
At each load the lead was found to undergo creep and eventually fail. For each increasing load the
steady state was shortened and the piece failed sooner.
After analysing the data the material constants of lead were calculated and used to calculate two
design loads for using the test pieces.
This experiment has shown the importance of considering the possibility of creep in future
engineering designs and has also shown how experiments can be used to come up with design
calculations which can be used to minimise the dangers of creep.

References
1.

Jones, A., Engineering Materials 1. 3rd ed. 2005: Elsevier.