ME-372 Mechanical Design II Creep Experimental Lab Page 1 of 8 S2007 GR

CREEP TEST
OBJECTIVE:
The fundamental objectives of this study are: 1. to perform the creep test on a polypropylene specimen and to observe the creep phenomenon. 2. to experimentally obtain the creep curve. 3. to derive the creep constants from experimental data.

APPARATUS REQUIRED:
SM106 creep measurement apparatus, Polypropylene specimens, Vernier, Weights (800gm), Weight hanger (100gm), Thermometer.

THEORY:
1. What is Creep? When a material like steel is plastically deformed at ambient temperatures its strength is increased due to work hardening. This work hardening effectively prevents any further deformation from taking place if the stress remains approximately constant. Annealing the deformed steel at an elevated temperature removes the work hardening and restores the steel to its original condition. However, if the steel is plastically deformed at an elevated temperature, then both work hardening and annealing take place simultaneously. A consequence of this is that steel under a constant stress at an elevated temperature will continuously deform with time, that is, it is said to "creep". Creep in steel is important only at elevated temperatures. In general, creep becomes significant at temperatures above about 0.4Tm where Tm is the absolute melting temperature. However, materials having low melting temperatures will exhibit creep at ambient temperatures. Good examples are lead and various types of plastic. For example, lead has a melting temperature of 326C (599K) and at 20C (293K, or about 0.5Tm) it exhibits similar creep characteristics to those of iron at 650C. The SM106 Creep Measurement Apparatus is a simple unit designed for demonstrating and investigating the creep characteristics of lead and polypropylene specimens at room temperature. A temperature module is provided to enable investigation of the effects of temperature on creep rate. 2. Creep in metals:

ME-372 Mechanical Design II Creep Experimental Lab Page 2 of 8 S2007 GR

A creep test is carried out by applying a constant load to a specimen and observing the increase in strain (or extension) with time. A typical extension - time curve is shown in Fig. 1.

Three regions can be readily identified on the curve: 1 to 2 Primary Creep -creep proceeds at a diminishing rate due to work hardening of the metal. 2 to 3 Secondary Creep - creep proceeds at a constant rate because a balance is achieved between the work hardening and annealing (thermal softening) processes. 3 to 4 Tertiary Creep - the creep rate increases due to necking of the specimen and the associated increase in local stress. Failure occurs at point 4. In terms of dislocation theory, dislocations are being generated continuously in the primary stage of creep. With increasing time, more and more dislocations are present and they produce increasing interference with each others movement, thus causing the creep rate to decrease. In the secondary stage, a situation arises where the number of dislocations being generated is exactly equal to the number of dislocations being annealed out. This dynamic equilibrium causes the metal to creep at a constant rate. Eventually, however, the creep rate increases and the specimen fails due to localized necking of the specimen (or component), void and micro crack formation at the grain boundaries, and various metallurgical effects such as coarsening of precipitates. When in service, an engineering component should never enter the tertiary stage of creep. It is therefore the secondary creep rate, which is of prime importance as a design criterion. Components, which are subject to creep, spend most of their lives in the secondary stage, so it follows that the metals or alloys chosen for such components should have as small a secondary creep rate as possible. In general it is the secondary creep rate, which determines the life of a given component.

ME-372 Mechanical Design II Creep Experimental Lab Page 3 of 8 S2007 GR

Secondary creep rate for a particular metal or alloy depends on several variables, the most important of which are stress and temperature. The most commonly used expression for relating secondary creep rate to stress and absolute temperature T has the form

= A n e RT

. (1)

where A and n are constants, E is the activation energy for creep in the metal and R is the universal gas constant (8.31 J/mol K). The equation shows that the creep rate is increased by raising either the stress or the temperature. Taking natural logarithms gives: ln = ln A + n ln E RT

.2

Thus, for tests at constant temperature and varying stress, the stress exponent n can be found by plotting ln against ln . Alternatively, if the stress is kept constant and the temperature varied, E can be determined by plotting In c against l/T. For the special case of lead, the stress exponent n has a value of about 10 for the relatively high levels of stress used in the SM106 apparatus, and the activation energy E is approximately 120 kJ/mol. The fact that the exponent n varies with stress demonstrates the inadequacy of simple laws for correlation of data over a wide range of stress levels. In practice, more complicated equations are used to correlate experimental data. For our purposes, however, it is sufficient to use the equation 1 since the resulting plots are very nearly linear for the stress levels normally obtained with the SM106 Apparatus. In this manual, the power law of equation 1 is used in the analysis of results. 2. Creep in plastics: Plastics also creep at ambient temperatures but, compared to lead, they are able to sustain much greater extensions before failure, the creep curves are similar in shape to those for metals, but the mechanism of deformation is quite different because of the difference in structure of the material. A polymer consists of long chain-like molecules in a tangled and coiled arrangement; creep occurs by chains untangling and slipping relative to one another. The creep rate is still dependent on stress and temperature but equation 1 can no longer be applied. The complex processes taking place during creep make it difficult to quote an equation that describes the creep behavior of all polymers. Many empirical equations have been proposed and one, which applies to some of the common engineering plastics, has the form = + m t k
.3

ME-372 Mechanical Design II Creep Experimental Lab Page 4 of 8 S2007 GR

where is the tensile creep strain after a time t, is the applied creep stress, is the instantaneous or initial strain produced on loading, and , m, k are constants for a given polymer. The elastic component of the initial strain can be calculated by dividing the creep stress by the tensile modulus of the polymer, which for polypropylene is 1250 N/mm2. In many polymers this initial strain is very small and can be ignored, so that in these cases = m t k 4

A plot of log against log t will therefore be linear and the slope will give the value of the exponent k. Values of k quoted in the literature range from 0.025 to 0.33. For polypropylene, k is in the range 0.1 to 0.2 and tends to increase with stress level. In cases where the stress exponent m is close to unity we have the situation where / is a constant as k 0, in other words the material is behaving in an elastic manner. Alternatively with high values of k, say k 1, then /t is a constant and the material is behaving as a viscous fluid. The value of k obtained from creep data is therefore a measure of the relative contribution of elastic and viscous deformation to the creep process. Finally, it should be noted that with polymer materials the primary creep stage, where the creep rate is decreasing, is largely recovered when the creep load is removed. This behavior is unlike that observed in most metallic systems, and the effect can be easily demonstrated using the SM106 apparatus by removing the load after the polymer has been creeping for 20 to 30 minutes and continuing to take strain readings. It will be found that the elastic strain is removed instantaneously, but that further recovery of strain takes place over a period of several minutes. This time dependent effect is due to recovery of the viscoelastic component of the creep strain. For the stress levels used in the SM106 apparatus (typically 19 N/mm2), approximately 40% of the creep strain is recovered after 5 minutes.

DESCRIPTION OF THE APPARATUS:

The SM106 Creep Apparatus uses a simple lever to apply a steady tensile load to a specimen (see Fig. 2). The specimen is held at each end by a plain stainless steel pin inserted through each loading stirrup. The pins are a close fit in the holes in the ends of the specimens and deformation of these holes during a test is negligible compared to the elongation of the specimen (i.e. less than 2% of the total elongation). Loads are applied by hanging weights on the end of the lever arm. A rest pin is provided to support the weight of the lever arm when loading the specimen prior to a test. A dial gauge measures the elongation of the specimen. The lever arm has a mechanical advantage of 8. The mass of the arm is 0.4 kg and that of the weight hanger 0.1 kg. The load on the specimen can be found by taking moments about the pivot. If the mass added to the weight hanger is m kg then:

ME-372 Mechanical Design II Creep Experimental Lab Page 5 of 8 S2007 GR

F = (2.2 + 8 m) g N

...5

Note that the mass m does NOT include the mass of the hanger, as this is included in the constant 2.2. The moment due to the specimen stirrup and pin is balanced by that of the weight hanger stirrup, so these do not appear in the equation.

Figure 2. SM106 apparatus and its components.

EXPERIMENTAL PROCEDURE:
Before starting the experiment a suitable load should be determined which will produce a complete creep curve in the time available. A suitable load can be determined from Fig. 3, which shows time to failure in terms of load and temperature. Normally, the load should be chosen to give a time of at least 15 minutes, but this can be reduced for simple demonstrations. For serious experiments it is recommended that the specimens should be labeled.

ME-372 Mechanical Design II Creep Experimental Lab Page 6 of 8 S2007 GR

Figure 3. Time to break specimen as a function of the specimen temperature for different hanger loadings (kg). Testing procedure for determination of the time exponent k: 1. Measure the cross-section and gauge length using a caliper. This is to enable subsequent calculation of stress and strain. 2. Raise the lever arm and support it by inserting the rest pin. 3. Slide the specimen into the stirrups and insert the load pins. (Be sure not to distort the specimen). 4. Remove the rest pin and gently lower the lever arm to take up any free movement, then zero the dial gauge. 5. Raise the lever arm and replace the rest pin. Hang the weight hanger on the end of the arm and add the required load (m=0.8 kg). 6. Record the ambient temperature and set a stopwatch or stop-clock to zero, ready to start the test. 7. Remove the rest pin and gently lower the lever arm until the specimen takes up the full load. At the same time start the stopwatch or clock. 8. Record the extension of the specimen at suitable intervals of time (normally every 30 second, but every minute is adequate for longer tests). The total loading period is 15 min.

ME-372 Mechanical Design II Creep Experimental Lab Page 7 of 8 S2007 GR

9. The load is removed towards the end of the loading phase and further readings are taken to show the recovery of creep strain. The recovery period is 5 min. 10. At the end of the test take another reading of temperature and record the load. Check that you have recorded the dimensions of the specimen. 11. Perform the experiment on a total of two specimens and use the average values to plot the graphs i) Elongation vs. Time and ii) ln vs. ln t. 12. The time exponent k is obtained from the slope of the curve plotted between ln and ln t.

RESULTS:
Determine the value of the time exponent k using the above described experimental procedure. Plot the graphs i) Elongation vs. Time and ii) ln vs. ln t. Tests can be conducted at different loads in order to determine the stress exponent m(not required for this lab) . Typical plots for polypropylene specimens are shown below:

Figure 4. Creep curve for Polypropylene showing creep recovery.

ME-372 Mechanical Design II Creep Experimental Lab Page 8 of 8 S2007 GR

Figure 5. Strain vs. Time on Logarithmic Scales - Polypropylene Specimen.

DISCUSSIONS:
A typical creep curve for a polypropylene specimen is shown in Fig. 4. This was obtained following the testing procedure. The initial part of the characteristic shows the three stages of creep. The elastic extension has been calculated by dividing the creep stress (18.8 N/mm 2) by the tensile modulus of polypropylene (1250 N/mm 2), which gives an elastic strain of 0.015. For a gauge length of 20 mm, this is equivalent to an extension of 0.3 mm. When the load was removed, this elastic extension was recovered immediately and this was followed by a more gradual recovery of creep strain. In this particular experiment the creep strain recovery was about 3 mm, in other words approximately 45% of the creep strain was recovered. As indicated in section 3, a value for the time exponent k in equation 4 can be estimated from a plot of log against log t, as shown in Fig. 4. In this particular Figure, has been expressed as % strain and the time expressed in minutes. It should be noted that equation 4 is only applicable for the primary and secondary stages of creep, so at the upper end of the graph the results begin to deviate from a straight line. From the initial part of the curve, the slope is found to be approximately 0.2 which is at the upper end of the range which would be expected for polypropylene. However, this result is for a higher stress and very much shorter creep time than would normally be used for material testing. The value of k will approach the more normally accepted value of 0.l if lower stresses and longer test times are used.

REFERENCES:
1. Engineering Materials and Their Applications, 4th ed., Richard A. Flinn & Paul K. Trojan; Houghton Mifflin Company, Boston, MA. 2. Mechanics of Materials, Ferdinand P. Beer & E. Russell Johnston, Jr; McGraw-Hill Book Company, New York.