Practical Testing
Using a uniaxial tensile test, drop tower impact test, and dynamic mechanical analysis to gather material response data at a range of strain rates and temperatures.
A group project completed during 4th year of University.
Teammates: Fang Te Fong, Afdhal Zofran bin Mohd Amin, Nur Amanda binti Mustapha Kamal
Industrial Partner: Crux Product Design
Uniaxial Tensile Test
The uniaxial tensile test strains the test specimen at strain rates within the quasistatic range at room temperature. This is necessary to firmly establish the baseline behaviour of each material and to later validate the use of the time-temperature superposition principle.
Test Images of Polypropylene (PP)
For this test, the sample was extended at a rate of 5mm/minute. This corresponds to a strain rate of 0.001 per second. Five specimens of each material were tested at this strain rate until failure or reaching the maximum possible crosshead displacement. The test specimen is a standard wishbone shape, following the ASTM D638-14 testing standard. Throughout each test, the strain is measured by a video extensometer. Due to the low thickness of the specimens (2mm), physically attached stress measuring devices would influence the material's behaviour by introducing local strains.
Test Images of Polyethylene Terephthalate (PET)
For this test, the sample was extended at a rate of 500mm/minute. This corresponds to a strain rate of 0.1 per second. Five specimens of each material were also tested at this strain rate.
Example Results
The results obtained for the PET uniaxial tensile test. It can be seen that the Young's modulus is significantly lower when the material is strained at a higher rate. The pattern holds across all four tested polymers, confirming the importance of fully understanding this relationship.
The Young's modulus was consistent across the repetitions of each test, but the yield strain and stress were variable. This is most likely due to defects introduced in specimen manufacturing.
The stress data is noisy and has to be smoothed before it can be used for material model calibration.
Drop Tower Impact Test
The drop tower impact test strains the specimen within the 'high rate' region, between approximately 40 and 100 per second. This establishes the behaviour of each material at the rates most likely to cause unexpected issues in medical devices and gives another set of data points to validate the use of the time-temperature superposition principle.
The test works by dropping a heavy mass (impacter) onto a thin, flat test specimen, and, throughout the impact, measuring the force the specimen applies to the impactor. This can be used to calculate the stress on the specimen during impact. Since the Young's modulus was needed, the data describing the strain on the specimen during impact also needed to be gathered. Strain gauges were used for this, along with a 50000Hz data logger. Four specimens of each material were used. Their dimensions were 150mm x 100mm x 4mm following the ASTM D7137M17 testing standard.
Plate Stress Approximation
To obtain the stress from the load data, the specimen is modelled as a rectangular plate, fixed on all edges, with a uniform load applied in a circle at its centre.
The approximation is taken from Young, Budynas, and Sadegh's book: Roark's Formulas for Stress and Strain.
Example Results
The results obtained for the three repetitions of the PP drop tower impact test. For this material, the behaviour was relatively consistent across the three repetitions. For all four polymers, the Young's modulus obtained from the drop tower impact test is significantly higher than from the uniaxial tensile test.
The noise is more significant than the uniaxial tensile test, and the smoothing necessary to use the data for material model calibration could introduce errors.
Dynamic Mechanical Analysis
Dynamic mechanical analysis (DMA) was the only test method that measured the material's change in behaviour in response to temperature. It also directly tests within the 'medium' to 'quasi-static' strain rate ranges.
DMA Methodology
DMA works by measuring the force needed to oscillate a test specimen at a range of frequencies. From this, the storage and loss modulus for those frequencies can be obtained. There are multiple methods of fixing the test specimen, but the single cantilever method shown was used for this project.
The test apparatus is contained in a hood that can be either heated or cooled, giving material behaviour for a range of temperatures too (for this test, -60 to 160 degrees C).
Each specimen was oscillated at a range of frequencies between 0.03Hz and 100Hz to an amplitude of 15 micrometers. This corresponds to strain rates between 0.006 and 15 per second.
Example Results
The results obtained from dynamic mechanical analysis of PET. The temperature range was chosen to capture the change in behaviour across the material's glass transition temperature (70 degrees C). This temperature is signified by a peak in the loss modulus.
The graph shows how different forcing frequencies (proportional to strain rates) affect the storage and loss modulus, and how that relationship is affected by temperature. The storage modulus is always higher at higher strain rates. The loss modulus, when above the glass transition temperature, is also higher at higher strain rates, but the trend reverses when below the glass transition temperature.
Time-Temperature Superposition
Through time-temperature superposition (TTS), DMA can give information on a far wider range of strain rates than what is physically tested. The principle is that testing a polymer at a higher temperature is equivalent to testing it at a lower frequency. The same applies in reverse: testing at a lower temperature is equivalent to a higher frequency. This means by heating or cooling the testing chamber, data can be gathered for simulated strain rates beyond the physical capabilities of the machine.
The limits of this method are determined by the temperature range. The upper limit on temperature (and the lower limit on strain rate) comes from each polymer's melting point. The lower limit on temperature (and the upper limit on strain rate) comes from the machine's cooling capabilities. Unfortunately, the liquid nitrogen cooling accessory needed to reach temperatures below -60C was not available.
Example TTS Results
PET Master Curve
Shows the master curve for PET at 30 degrees C, generated through time-temperature superposition. It can be seen that the range of simulated frequencies tested is extremely large. The master curve for each material is compared to the data from the tensile and impact tests, and used to calibrate the strain rate-sensitive components of the material models.
Combined Results
Combined results for Polyoxymethylene (POM)
For Polyoxymethylene, shows the master curve at 20 degrees C, the uniaxial tests for extension rates of 5mm/min and 500mm/min, and the drop tower impact tests. There is a large disparity in the values for the Young's modulus generated by the uniaxial tests and those generated by DMA and TTS, while the drop tower tests and DMA results suggest broadly similar values. This suggests either significant experimental error or an incorrect application of time-temperature superposition.
These combined results are fed into the constitutive model as material parameters.