The project needs 3D modeling with FEA and hand calculations to compare the results.
There will be two separate reports: one for Lab 1 and another for Labs 2-5.
LAB 1:
- Use instructions!
- Build a realistic loading scenario using Inventor CAD. Just make sure it's a scenario I can replicate at home with a yardstick.
- Compare Hand Calculations with FEA.
- In the report use only photos, sketches, tables, charts, graphs and no more than 250 words, documenting the experiment, level of success, any conclusions.
LAB 2-5:
- Use the instructions!
- Use the template as a guide for the report.
- Compare Hand Calculations with FEA.
- In the report use pictures, sketches, tables, charts, graphs, documenting, level of success, any conclusions.
- I provide examples on how some labs should look like in FEA.ENGR 305: Cantilever Beam Lab – Picture Version Goal: To create an awareness of the differences between the real world and calculations. In this lab you will compare measured and predicted deflections of a simple wooden yardstick. Process. 1. Form a lab team of 2 to 4 students. 2. Get a yardsticks from Professor Tubergen, Professor Haan or from other lab teams. 3. Working with your lab partner(s), determine a simple loading scenario that you can build using the yardstick. This loading scenario should meet the following criteria: a. It should align well with analytical results which are shown in the textbook in Appendix B19. I recommend scenarios 1, 5 or 6. b. The applied load will not break the stick. 4. Build the loading scenario with 1 yardstick being mindful not to modify it significantly. That means no cutting and any drilled holes should be located on the centerline and be 3/32” or smaller. 5. Test the setup to determine whether the experiment has some (within 50%) level of agreement with the calculations. 6. If, based on 1 through 5, you think you have a reasonable experiment, obtain 3 yardsticks and repeat the experiment with each one. 7. Validate your experimental results from hand calculations as well as FEA in Inventor or Solidworks. Questions to resolve: 1. What are the best material properties to use? What are the dimensions of the beam? 2. What is the best method of measuring results? 3. How do I build it? 4. How do I make this repeatable for multiple beams? 5. How well do my hand calculation match the computer FEA? 6. How well do my hand calculations match the experimental data? Outcomes: 1. An understanding of the modeling process and the relationship between models and reality. 2. An appreciation for the variability in experimental data using similar test specimens. Deliverable: Using only photos, sketches, tables, charts, graphs and no more than 250 words, document your experiment, your level of success, any conclusions. 1 report per team. Load and Support Slope at End Maximum Deflection Equation of Elastic Curve ase (Length L) + A) (+ upward) (+ upward) 2 3 op ne 2 } 251 3EI yon atx=1L atx=1L VT Tek * Lab Exercise #2 Model this flat bar is 10 mm thick. Model the bar in inventor and load it with P=10kN. What do you estimate the stress to be when you calculate Stress=Force/Area. What area did you use? Does your answer make a little sense? Use the snipping tool or other method to capture images of your FEA in either Inventor or Solidworks and submit them with a writeup which contains answers to the above questions. Lab Exercise #3 FEA Problem: Build a wedding cake problem (with 3 sections, not the cheap kind) and add an axial load (either tension or compression). Duplicate the solution for the displacement by hand to validate the results for the total displacement. Provide a table (either by hand or MSOffice product which tabulates the geometry (length, dia, material) and materials for each section along with the displacements from FEA and hand calcs. You should turn in a document with your hand calculations and a few pictures showing your FEA results. It will be easiest to compare total displacements. If they disagree a lot, you may have to look at the displacement of the individual sections. Lab Exercise #4 EA Problem: I want you to do a scuba tank problem. I don't want to worry about axial loads, however, so this is just a cylinder with internal pressure. Use an ID of 6.375" and an OD of 7.25". The material is 6061-T6 aluminum. Use an internal load of 3000 PSI. Compare your results to hand calculations. The problem you will have is that the model that you build is in cartesian-coordinates and the hand calculations that you do based on the book are in cylindrical coordinates. This will not be a huge challenge, but it will set the groundwork for coordinate transformations which will need to be done. Lab Exercise #5 Build a wedding cake problem (with 3 sections, not the cheap kind) and add a TORSIONAL load (either clockwise or counter-clockwise). Duplicate the solution for the displacement by hand to validate the results for the total displacement. Provide a table (either by hand or MSOffice product which tabulates the geometry (length, dia, material) and materials for each section along with the displacements from FEA and hand calcs. You may use one material for the entire assembly (cheesecake). You should turn in a document with your hand calculations and a few pictures showing your FEA results. It will be easiest to compare total displacements. If they disagree a lot, you may have to look at the displacement of the individual sections in order to explain why this occurred. Experiments in FEA A COMPARISON TO HAND CALCULATIONS Ren Tubergen, Ph.D., P.E. | ENGR 305 | 11/15/XX PAGE 1 Introduction: In order to demonstrate the skills that have been developed in this class, work was done by hand and utilizing commercial FEA software. The results from each method were compared to each other, demonstrating the simple problems could be effectively solved using either method. Because the results matched, it is almost certainly the case that both analysis were done correctly. PROCEDURE: Four simple analysis types of loading scenarios where created which allowed us to test the major different types of loads that can be encountered. Those scenarios were axial loads, torsional loads, pressure vessel loads, and beam loads (flexural loads). Each scenario had been discussed during class and had equations which would easily check stresses, deflections, or both. By recreating the simple situation for which the equations applied with the FEA software package, we were able to validate the solutions. This will be demonstration in the next sections as each type of loading is evaluated individually. Axial Loads: The first type of loading scenario that was created was an axial load. This was done by creating a wedding type problem and putting it under a tensile (or compressive) load. It was anticipated that each section would deform according to the equation: Delta = Sum for all section (PL/AE) The plot below shows that the answer from the FEA showed that the total deflection over the three sections was XXXX. This matches pretty well with what was calculated by hand. Hand calculations showed that we would have XXX. Figure 1 showing AXIAL FEA TABLE 1 Summarizing results FEA HAND CALCS % Diff PAGE 2 The differences are a result of blah blah, this approximation in the hand calculations This assumption is not in the FEA where “another approximation” is taking place. This can be shown in the details of the plat. You can see in the region shown in Figure 2 that the exaggerated deflection shows blah blah blah, which is simplified out of the hand calculations. Overall, our error was reasonable and can be explained easily. The approximations make the hand calculations vary “higher/lower” than the FEA and that is borne out by the results. Torsional Loads: This section should replicate the axial load section. Pressure Vessel: This should look like Torsional Loads which should look like axial. There will be differences because, especially pressure vessels, are harder to load that the other scenarios. Errors might be higher and you may have had to make bigger approximations in the physical model. Stress Concentration: Get the theme at this point. Any other loading scenarios (like beam) Bonus points for other scenarios!!!!!!! Discussion Each of the above sections should point out that you are capable of using the tool. That means that you have matched hand calculations to a reasonable degree or can explain the differences. Overall differences should be 5% or less. If not, you should be able to explain them. This is the key section for me in that the lessons of the 4 sections can be combined into an overall theme. I am not sure what that looks like for you. I would expect some people to comment about how the biggest struggle with using FEA is the creation of the model which is a reflection of personal skill. That won’t be true for everyone. It may be that you have learned one of the big lessons about FEA and that is that boundary conditions are very important or that while the code can do stress risers, the quality of the solution depends on the mesh. It may be that you have found you can’t trust the results locally around the places where loads were applied. You may have something else that you would like to bring to the table(VM Stress or ????). Here is the place to discuss it. It doesn’t need to be positive. Criticism of the software or process is valid (or personal skill). PAGE 3 Conclusion: Put a brief conclusion which is a recast of the introduction and discussion with a hint of the quality of the data correlation. Notes: I think by the time you get done, this will take 3-15 pages of text and tables. Depending on how many pictures you include, that could double the size of the report. Remember that each image should make one or two points, but no more. If it doesn’t make a point, don’t include it. Don’t include it to show me that you did the work. Include it only if it make the discussion easier. For example, you can relatively quickly describe a cantilever beam with words (cantilever beam fixed on the left side). It is not as easy to tell me how you added a torsional load to a shaft. The cantilever beam may not require a photo, but the shaft probably does for clarity.