Comparing Real World Measurements with Finite Element Results Report

This report presents a comprehensive analysis comparing experimental tensile testing results with finite element analysis (FEA) simulations. The study involved tensile tests on plastic, aluminum, and brass specimens, where force-displacement curves and material properties like elastic modulus, yield strength, and tensile strength were obtained. A Solidworks FEA model was created to replicate the tensile behavior, with appropriate boundary conditions and material properties. The report details the experimental setup, including test-piece dimensions and loading procedures, and the FEA model setup, including material definitions and boundary conditions. Results from both experimental testing and FEA simulations, such as stress-strain plots, failure loads, and stress distributions, are presented and compared. The discussion highlights the correlation between experimental and FEA results, the impact of linear vs. non-linear material properties, and the importance of mesh convergence for accurate stress predictions. The study concludes that the FEA model accurately simulates the strain behavior when failure load is applied, while stress values can be improved with non-linear material properties. The report serves to demonstrate the reliability and accuracy of FEA in assessing the performance of materials and designs.

ASSIGNMENT
CENTRE FOR ENGINEERING AND INDUSTRIAL DESIGN
APPLIED COMPUTATIONAL MODELLING
ASSIGNMENT 2
COMPARING REAL WORLD MEASUREMENTS WITH FINITE ELEMENT RESULTS
INTRODUCTION
Finite element modelling is reliable, accurate and handy way to assess performance of simple as well as
complex parts. Computer models can be made with the help of near-shape CADs of the parts to be
assessed, their functional requirements can be modelled as boundary conditions and appropriate loads.
There will be a few assumptions made in these models, despite which if the assumptions are reasonable,
the behaviour simulated by finite element methods can help performance, make improvements, and get
the design to be improved as per the constraints. Digital models need material inputs to replicate their
behaviour which can be captured through simple testing such as tensile testing, compression testing and
shear testing.
In the current study, simple tensile testing was performed on standard tensile specimen made using 3
different materials. The obtained results were analysed to obtain effective stress-strain results. A finite
element model was made in Solidworks to replicate the tensile behaviour. Simplification assumptions
were made in the model to simulate the results. A comparison of experimental and numerical results were
made to understand fidelity of numerical methods.
METHOD
1. Laboratory tests
a. Description of test-piece used for testing
The test-piece used for experimental testing is shown in figure 1. The overall length of the
sample is 90 mm. Test-piece has threads on both the holding region which are metric thread

2
M12 x 1.75. The center region or gauge region has a diameter of 3.3 mm. The length of
threaded regions are 31.7 mm for top and 19 mm for bottom.
Figure 1. Dimensions of tensile coupon used for testing
b. Test-piece mounting and loading details
To mount a tensile sample into the machine, the short threaded end of the tensile specimen (19
mm) is inserted into the threaded hole in the top of the load cell. We screw the sample inside
the hole of load cell until the top edge of short threaded section is in-line with top of the load
cell. After that we lower the load bar so that longer threaded section of tensile bar goes through
the center of the load bar. We lower the load bar until the bottom edge of longer threaded
section meets with bottom of load bar. After holding the sample, we screw the sample nut onto
the threaded section until sample is tightly held into the place.
c. Description of the recorded results
Force (N) vs. displacement (mm) curve was recorded over the time for tensile testing. After
mounting the samples, the knob is rotated slowly to move the loading head upwards which
pulls the sample hence force exerted by it. The force vs. displacement curves are used to
calculate engineering stress vs. engineering strain curves for which gauge region diameter is
used to calculate the area. The length of the gauge region is 35mm. The obtained plots are as
shown below:
Figure 2. Force-displacement & true stress-true strain plots for Plastic sample

3
Figure 3. Force-displacement & true stress-true strain plots for Aluminium sample
Figure 4. Force-displacement & true stress-true strain plots for Brass sample
Table 1. Calculated material properties for 3 materials derived from experimental data
Material Name Plastic Aluminum Brass
Elastic Modulus (GPa) 2.5 59 65
Yield Strength (MPa) 30 330 280
Tensile Strength (MPa) 45 550 450
Elongation (%) 3% 9.3% 13.1%
2. Finite element model
a. Brief description of simulation software used
Solidworks CAD and Simulation package has been used to make the CAD of tensile coupons
and simulate the tensile test to obtain the stress distribution and strain plot.

4
b. Model description
A static study is used in Solidworks to mimic the experimental test. The short end of the
sample which is fixed inside the machine is assigned as fixed boundary condition in analysis.
The longer end is attached to moving head hence displacement boundary condition is specified
to this end. Figure 5 shows the loads and boundary conditions applied to this model. A new
material with the name of Plastic, Aluminum and brass are created in Solidworks to feed the
tested material properties. Analysis is solved for each material.
Figure 5. Loads and boundary conditions as applied in Solidworks analysis
c. Results from simulation
Results from linear static analysis for Plastic material is shown in figure 6 which has von-
Mises’ stress plot, strain plot and displacement plot. Since a linear analysis is performed, the
stresses increase linearly and show a very high value although the strain just crosses the
material limit. If non-linear material properties are used, the stresses also would match the
experimental results.
Figure 7 shows the analysis results for Aluminium material and Figure 8 shows analysis results
for Brass material.

5
Figure 6. von-Mises’ stress, equivalent strain and displacement plot from analysis with Plastic
material

6
Figure 7. von-Mises’ stress, equivalent strain and displacement plot from analysis with
Aluminium material

7
Figure 8. von-Mises’ stress, equivalent strain and displacement plot from analysis with Brass
material
3. Experimental results
a. Failure Load
The failure loads for each material are different. Since Plastic is the softest, it has lowest failure
load. It is listed in the table below:
Table 2. Failure loads for all 3 material tested currently
Material Name Plastic Aluminum Brass
Failure Load (N) 377 4307 3297

8
b. Failure Location
The failure location of the samples were in the gauge region where a small notch was already
present before the testing was started. This notch helps localize the stresses once the material
reaches its elongation limit.
4. Finite element model results
a. Maximum stress corresponding to experimental failure load
In Solidworks analysis, the maximum failure load corresponding to each material was applied
to pull the sample from longer gripping region. The stresses obtained are linear hence values
are high but if they are corrected for non-linearity, it would match experimental values.
Table 2. Maximum stress corresponding to failure load in Solidworks
Material Name Plastic Aluminum Brass
Max. stress (MPa) 412 9355 16280
b. Location of maximum stress
The location of maximum stress for all the samples was near the fillet region where gauge
region is connected with gripping region. This is because of the load transfer mechanism.
Second region is the gauge region itself.
c. Mesh convergence
Analysis was run with 4 different mesh sizes and mesh was refined in the gauge region. A size
of 0.25 mm was obtained after which further refinement did not improve the stress results
further.
Figure 9. Converged mesh used for current analysis in Solidworks