Summary of our progress so far
Since the project proposal, 5 samples of each 15 degree grain orientation between 0 and 90 have been printed. These samples have been tested for their tensile failure, and data has been collected for the entire process, but most notably the maximum failure strength of each sample.
Execution of the project proposal
The execution of the project proposal occurred in three stages.
During the first stage the 7 different degree orientations from 0 to 90 degrees were programmed into gcode using a custom slicing software organic to the polymers lab. This allowed for the control of bead orientation amongst other print parameters outlined in the project proposal.
The second stage consisted of printing the test coupons with the Lulzbot TAZ5 3D printer in lab ME 1001. A single print consisted of printing five coupons all with the same degree orientation. A single coupon would be printed in its entirety before the next coupon was printed. This pattern repeated until the printing was complete. Each print took approximately 4.5 hours, which 5 test specimens being produced for print. The printing took place on a dedicated printer in the Polymer Engineering Center at UW-Madison, so was not at risk of imposing on other students usage of printers at the UW MakerSpace. All prints were completed between 26 and 28 March 2019. Printing conditions are captured on the table below.
Table 1: Printing conditions
| Print Bead Orientation | 0 degrees | 15 degrees | 30 degrees | 45 degrees | 60 degrees | 75 degrees | 90 degrees |
| Date/Time Print Started | 26 March, 9:00 am | 26 March, 2:00 pm | 26 March, 7:00 pm | 27 March, 9:30 am | 27 March, 3:30 pm | 28 March, 8:00 am | 28 March, 12:45 pm |
| Print temp [C] | 220 | 220 | 220 | 220 | 220 | 220 | 220 |
| Bed Temp [C] | 100 | 100 | 100 | 100 | 100 | 100 | 100 |
| Indicated Ambient [F] | 72 | 73 | 73 | 70 | 75 | 70 | 73 |
| High Ambient Temp [F] | 72 | 73 | 73 | 73 | 75 | 77 | 77 |
| Low Ambient Temp [F] | 68 | 68 | 70 | 70 | 70 | 70 | 70 |
| Indicated humidity [%] | 20 | 16 | 16 | 20 | 16 | 27 | 23 |
| High Humidity [%] | 29 | 29 | 20 | 20 | 21 | 27 | 27 |
| Low Humidity [%] | 16 | 16 | 16 | 16 | 16 | 16 | 16 |
The third stage consisted of tensile testing the printed coupons. All coupons were tensile tested between 3:00 pm and 6:00 pm on 28 March 2019. Prior to testing each coupon was measured to determine its cross sectional area in the region where breaking would occur. The 3D printed coupons didn’t have perfectly square dimensions so nominal dimensions were taken. Figure 1 below represents an extreme cross sectional area with deformations on the first layer or two of printing creating a “skirt” type feature on the bottom of the print. Figure 1 also represents a “mound” on the top of the print which occurred during the 0 degree sample due to print beads stacking up while the 3D printer printed from the outermost bead to the inner bead. Referencing Figure 1, the faces of the coupon used to determine its cross sectional area are those represented by the 3.20mm and 13mm dimensions. The height of the “mound” and width of the “skirt” aren’t to scale, rather they represent the approximation of the shape witnessed on the 3D printed coupons. After cross sectional areas were obtained, the coupons were loaded into the test cell and tensile tested to their failure points. The coupons were held in the test cell jaw clamps with sandpaper which prevented slipping of the coupons during the test. After tests were completed the data was compiled and is represented in following sections of the report.
Pictures and data
The results from the tensile test were exported as Excel files and are ready for analysis. The most relevant parameter for our project is the maximum stress of each sample. However, the test also provided data that may be useful in the future such as strain and Young’s modulus. Figure 2 below is the sample of the plots generated by the testing machine. It illustrates the tensile stress versus strain result of the 45-degrees specimens.
From the all the testing results, the plot of maximum stress at each printing angle was created (Figure 3). The trend agrees with the expectation to some extent. Figure 3 shows that overall as printing angle increases, maximum tensile strength decreases. This is, however, just an initial observation. Further analysis is needed in order to fully understand the trend of the testing results.
Besides the quantitative results, the failure characteristics of each specimen were also recorded. They are shown in Figure 4 – 10. When the printing angle is below 45 degrees, the specimens tend to fail along the cross-sectional plane. However, as the printing angle exceeds 45 degrees, the failure region tends to be along the axial direction of the filaments.
Next steps
For the next update on this progress report, we aim to analyse the data we have already acquired. Primarily, we want to quantitatively apply failure criteria analysis to our specimens. Focus will be on Tsai-Hill criteria as this is commonly used for anisotropic materials, but multiple criteria will be applied and evaluated. Another will be the Osswald-Osswald Criterion and the criterion upon which it is based, the Gol’denblat-Kopnov Criterion. The goal will be to analyze the accuracy of these models with our data. Moving forward, this could allow for more accurate failure prediction for parts that use FFF printing. This is crucial for the use of manufacture and use of FFF printed parts in situations where safety is a large concern.
Potential future steps would also include visual aids for viewing the failure. In creation of future prints, a colored grid on the test specimen itself would help show distortion that occurred before the specimen broke.
Improvement on specimen printing
Although the current printing process gives a satisfying specimen quality, there is still room for improvement for future printing. Since the specimens are designed based on the ASTM D-638 Type 1 coupon, the “mound” and “skirt” (Figure 1) cause the cross-sectional profile to deviate from the standard. One potential solution to reduce the “mound” is to reduce the material feed rate at the center of the part for the last 2-3 layers. This aims to get rid of the extra material on the center region. For the “skirt”, adjusting the printing parameters for the first layer may be a solution. One possible adjustment is to reduce the feed rate when printing the outermost edges of the first layer. All in all, we will discuss these potential solutions with Gerardo and may conduct a print trial to see their effects.