Progress Report
Overview of Current Progress
An initial wave spring was printed following the direction of the initial project proposal. This was done with highly flexible material, resulting is a non-functional par. The support scheme necessary to support such a complicated geometry was extensive. So much so that, as well as parts requiring significantly more processing time, it might interfere with experimental data. Also, given that SLA printing allows for highly customizable geometry, the geometry was changed to better understand how these materials can be used to alter the stiffness of the printed components. The geometry was adjusted to be cylindrical such that orientation and internal geometry could be varied and studied in relation to part stiffness.
After altering geometry to cylindrical samples with varying thickness of internal structure, nine additional samples were printed. Printing all nine new specimens took three and a half hours in comparison with the four and a quarter hours required to print one wave spring. These parts were tested to understand the elastic stiffness for small deflections. While failure modes are not the direct subject of this project, some failure testing was conducted to understand the effect of print direction on plastic deformation and failure.
Figure 1: The three cross-sections tested
Experiments Run/Parts Tested
The following table describes the parameters used for the nine tested samples. These samples were printed with Formlabs “Tough” material and can be seen in Fig. 2.
Table 1 – Print parameters for the current test setup.
Figure 2 – cylindrical compression samples printed using SLA from Formlabs Tough material.
These nine samples were tested in compression using lab equipment provided by Professor Wendy Crone. Each sample was tested in compression three times with all forces remaining lower that the yield strength of the sample so that all tests were performed in the elastic region with no plastic deformation. On the third test, for the internal cross, the parts were tested past initial plastic deformation to understand their failure modes. The compression test setup can be see in Fig. 3 below.
Figure 3 – Compression testing setup at the ERB 210 lab. Equipment is sponsored by Professor Wendy Crone
For thick cross internal geometry, the parts printed horizontally and vertically took on distinct regions of failure seen in Fig. 4. The vertically printed part began bulging near the top of the sample with a narrow region of bulging. The horizontally printed part began to bulge nearer to the middle, with a wider region of bulging and a stronger tendency toward buckling with the specific loading condition.
Figure 4 – Large cross yielded parts
For the thin internal cross, the parts were tested beyond initial plastic deformation, and ultimately shattered. For the varying print orientations, distinct directions of layer delamination are visible. For the horizontal sample, the cracking between layers resulted in longitudinal splitting of the part. For the vertical sample, cracking is visible in the lateral direction even though loading was applied perpendicular to the cracking direction. For the 45 degree sample, cracking also occurred along the layers resulting in a shearing at the 45 degree angle. These results are visible in Fig. 5.
Figure 5 – Horizontal Print (left), Diagonal Print (Middle), Vertical Print (Right)
The data that was collected from the compression testing machine was force as a function of cross head displacement (extension). Due to the fact that these parts were relatively flexible in comparison to the machine stiffness, only cross head displacement was collected. An alternative approach would have been to implement a strain gauge to detect actual specimen displacement. Figure 6 below shows all nine samples during the elastic (linear) region of compression. Note that the data was shifted along the x-axis such that initial conditions were consistent across all specimens. There are several interesting takeaways from elastic compression tests. The first observation is that for the solid cross section, the stiffness (slope) is very close along all three print orientations. To further analyze the compression test results the slope (spring stiffness) of the elastic region was calculated as seen in Table 2. One interesting observation is that the horizontal print orientation had the highest stiffness across all three specimens. This intuitively makes sense because the force is acting along the laminate lines and should be most resistant to deformation.
Figure 6 – Compression testing results (elastic region) for the SLS Tough Material Cylinders
Table 2 – The linear stiffness (slope) was calculated across all nine specimens
To further investigate the mechanical behavior of these specimens, the small cross section was taken passed the elastic region, through plastic deformation and eventually to the fraction point (Fig. 7). The failure behavior for all three print orientations was nearly identical which is very interesting. One important point to note is that the horizontal print orientation was not initially taken to failure. After observing the plastically deformed sample, it was then compressed until failure which can be seen in the images above.
Figure 7 – Compression testing results (including yielding) for the SLS Tough Material Cylinders
Future Direction
Given the results of the compression tests and the rigidity of the current samples, additional studies will be of interest for the remainder of the semester. The first test that we would like to complete would be to rerun the tests using a material that is more flexible. We have some concerns about the strength of the “tough” material because we would ideally like to model this as a spring. Another test that we would like to complete is to actually design a coil spring with the tough material and do a compression test for different print orientations. This would be interesting because we could further investigate the stiffness dependence on print orientations. Finally, we would like to complete a test using an FFF manufacturing process with a relatively flexible material while varying percent infill. We believe that this will provide valuable insight as to how infill contributes to the stiffness of a product. If more time is available, we would ideally like to complete a statistical analysis on the above tests to show robustness in our test approach.
Please let us know if you have any comments or suggestions as we continue to investigate the SLA effects on polymer springs! Additionally, please zoom in for clarity on pictures, graphs and tables.