Paper ID #14227

Practical Engineering Experience in Aircraft Structural Design

Dr. Masoud Rais-Rohani, Mississippi State University

Masoud Rais-Rohani is the Associate Dean for Research and Graduate Studies and Professor of AerospaceEngineering in the Bagley College of Engineering at Mississippi State University. Masoud earned his PhDdegree in aerospace engineering from Virginia Tech in 1991. He has taught courses mainly in the areasof aerospace structures, mechanics, and design optimization. He has made extensive use of experientiallearning and computer applications in his courses, particularly the senior-level aerospace structural designcourse.

c©American Society for Engineering Education, 2015

Page 26.1234.1

Practical Engineering Experience in Aircraft Structural Design Abstract Engineering analysis and design topics in a senior-level aerospace structural design course are supplemented with simple in-class demonstrations, hands-on experience in sheet-metal fabrication, and a comprehensive engineering project involving design, analysis, optimization, manufacturing, and testing of stiffened panels. Besides providing students with opportunities to gain a deeper understanding of the concepts discussed in the course, they are introduced to other important aspects of engineering such as teamwork, communication skills, time management, personal responsibility and ethics. Details of the individual experiential learning activities and the resulting outcomes are presented and discussed. 1. Introduction Experience shows that when students are engaged in activities that integrate both cognitive and sensory systems, they are more likely to maintain focus and have better retention of the subject presented. There is anecdotal and scientific evidence to support the positive impact of experiential learning that many experienced teachers know to be true through personal interactions with students. The continuing interest in active, collaborative, cooperative, and project-based learning1-3, and pedagogies of engagement4 is fueled by the need to modify, transform, or otherwise increase the effectiveness of lecture-based courses, particularly in STEM fields. Some engineering educators have relied on the use of physical models5-7 and hands-on classroom demonstrations8-11 as a way of increasing the impact of traditional lectures in structural mechanics. Others have attempted to broaden the design education through experiential learning opportunities in manufacturing12 and construction13 engineering as a result of the pioneering work by Kolb14. Strengthening the understanding of aircraft structural analysis and design through classroom demonstrations and team projects is the main focus of this paper, which extends the prior work on the subject15. In particular, it relies on the experience of the author in teaching the senior-level Aerospace Structural Design course in the Department of Aerospace Engineering at Mississippi State University for nearly two decades. The field of aerospace structures, as a subset of the broader area of structural mechanics, deals with topics—such as deformation, stress, instability, and failure—that are influenced by the physical and engineering properties of materials as well as the geometric attributes and boundary conditions of the structural system. The thin-walled nature of flight structures is a noteworthy feature that, combined with low-density materials, is critical in reducing the weight of aircraft and spacecraft. For example, the aluminum-alloy fuselage skins of most transport airplanes16 (e.g., B737 to B747) have a thickness in the order of 0.036" to 0.071", which is roughly 4 to 8 time greater than the 0.009" thick standard printer paper. The longitudinal stiffeners or stringers used to support the fuselage skin are also of similar thickness.

Page 26.1234.2

The fact that such thin-walled structures are used in the construction of large transport aircraft has led to interesting classroom discussions. It has also provided learning opportunities ranging from classroom testing of simple cardboard models to a comprehensive engineering project involving design, analysis, optimization, manufacturing, and testing of stiffened aluminum alloy panels. Although the learning of aircraft structural design and analysis methods is essential, it is the experiential learning opportunities that help students gain practical engineering experience as future engineers. For example, without familiarity or direct experience with a particular manufacturing process, a student may suggest design features or dimensions that could not be produced. A design concept may be easily drawn on paper or captured using a CAD tool, but without adequate consideration of the manufacturing process, the result is likely to be a non-viable design. Similarly, other important factors, such as variability in the material properties or the manufacturing process and its potential impact on performance of the designed product, could be easily overlooked. Thus, a more comprehensive approach to engineering design education is one that involves manufacturing and testing. The paper aims to provide insight into the overall level of success in introducing a practical engineering experience into aerospace structural design course by describing each activity with pertinent detail. 2. Structural failure concepts of interest Instability and compressive failure of thin-walled structures are the principal topics of interest in the experiential learning activities described below. To provide proper context, a brief description of each failure concept is presented. 2.a Euler buckling When a structural member is placed under compression, it can experience global instability or buckling. For columns, buckling depends on the effective slenderness ratio !! !, and it could be either elastic or inelastic. In the elastic case, column-buckling (or critical) stress can be predicted using the Euler buckling formula given as

!!" =!!!!! ! ! (1)

where E is the Young’s modulus of the material, !! is the effective length of the column, which depends on the column length and its boundary conditions, and ! is the radius of gyration of the critical cross-sectional axis. Equation (1) is applicable to the so-called long columns when buckling occurs at a stress level below the proportional limit of the material. 2.b Local buckling This form of buckling occurs in thin-walled columns. Each flange or web member is susceptible to local buckling. The member with the lowest !! ! ! !value buckles first, where !! is the

Page 26.1234.3

buckling coefficient that depends on the aspect ratio of the member and the rotational rigidity of its supported edges, t is the wall thickness, and b is the length of the loaded edge of the member. The general equation for elastic local buckling under uniaxial compression is given as

!!" =!!!!!

12 1− !! ! ! ! (2)

where ! is the Poisson’s ratio of the material. As in the case of global buckling, it is possible for local buckling to be inelastic. In that case, the Young’s modulus is replaced by the tangent modulus, !! found using the Ramberg-Osgood formula17 !!! = 1+

37! ! !!.! !!!

!!

(3)

where n is a parameter that describes the shape of the stress-strain curve in the yield region with !!.! representing the stress corresponding to the secant line at a slope of 0.7E. The values of n and !!.! are tabulated quantities describing the properties of structural materials commonly used in the aircraft industry. 17 2.c Crippling Crippling is a form of local failure observed in thin-walled sections (i.e., short columns) under compression. Depending on the cross-sectional geometry, different semi-empirical equations have been developed by Gerard.17,18 As an example, for a multi-corner section, the crippling stress, !!" is found as

!!" = 0.56!!" !!! ! ! !!"!.! !.!"

(4) where !!" is the compressive yield strength of the material, A is the cross-sectional area of the section, and ! is a shape parameter with its value depending on the number of web and flange elements in the section. It is worth noting that the length of the loaded section does not appear in Eq. (4) because it is assumed to be small (i.e., of the same order of magnitude as the cross-sectional perimeter). For pure crippling to occur, the thin-walled short column has to have an effective slenderness ratio of less than 20. 2.d Johnson-Euler failure For thin-walled columns with effective slenderness ratio in the range of 20 ≤ !! ! ≤ !! ! ! , failure is a combination of crippling and inelastic buckling with the corresponding stress estimated using the Johnson-Euler formula17 given as !!" = !!" 1−

!!"4!!! !! ! ! (5)

Page 26.1234.4

The critical slenderness ratio !! ! ! corresponds to the transition point beyond which the column response is in the form of Euler buckling. The compressive strength at !! ! ! is equal to !!" 2. 2.e Stringers versus stiffened panels The failure modes described by Eqs. (1) through (5) apply equally to both stringers and stiffened panels under axial compression. Key differences are in the calculation of effective width in the case of stiffened panels where the skin is assumed to have a uniform stress equal to that experienced by the stringer attached to it. Additional consideration is given to panels with riveted stringers that can experience inter-rivet buckling. Further details can be found in many airframe structures textbooks, e.g., Bruhn17. 3. Cardboard column as a simple learning/teaching tool Classroom demonstrations are generally beneficial in helping students gain a deeper understanding of engineering concepts. In the context of aerospace structural analysis and design, students learn that modern airframe structures rely on the fact that thin-gage sheets of aluminum can be turned into efficient thin-walled structural members when formed into composite sections with various multi-corner shapes. As a simple classroom demonstration, the cardboard at the back of an 8.5" x 11" engineering pad serves as an excellent teaching tool. It is easy to demonstrate that when the flat cardboard is put under axial compression, it takes very little load for it to buckle in the first Euler buckling mode. However, if it is folded in the middle to form an angle section with two 4.25" x 11" flanges, all of a sudden, it shows much greater resistance to axial compression. Theoretically, the resistance of the 8.5" x 11" cardboard to buckling increases by a factor of approximately 4.4 when folded into an angle section of equal flanges. This is, of course, because of the corner introduced into the geometry. As the number of folds (corners) is increased, the rectangular cardboard could be turned into a Zee (2 folds), a Hat section (4 folds), or even more complex shapes. The cardboard column demonstration serves as excellent segue to the discussion of more advanced topics such as local buckling (instability) and crippling (failure) of thin-walled sections as highlighted earlier in Section 2. The discussion of local failure or crippling of thin-walled columns becomes more meaningful and interesting through simple demonstrations. In this case, an 11" tall Hat shaped cardboard column (with carefully measured dimensions and uniformly drawn fold lines) is placed on top of a table and students are asked to load it by placing their textbooks on top of it, one at a time as depicted in Fig. 1. The column, consisting of three 2" webs and two 1.25" flanges, is manually held in place to stabilize it while it is loaded. Careful placement of each textbook ensures

Figure 1. Cardboard column demonstration.

Page 26.1234.5

centric loading. For an added excitement, students are asked to guess, without doing any calculations, the number of textbooks (each weighing ~ 4.5 lb) that can be stacked on top of the cardboard column (~0.025" thick and weighing 1.1 oz) before it collapses. Predictions vary significantly, but the prior discussions lead to the expectation that a flat piece of cardboard folded into a Hat-shaped column is a relatively strong structural member. Of course, given the lack of precision in the support and loading mechanisms, the number of books resulting in the eventual collapse of the column has noticeable variance. However, the key element of the demonstration remains the same, i.e., the impressive load-to-weight capacity of multi-corner sections under axial compression. Repeated demonstrations of the cardboard Hat column has indicated a range of loading from 14 to 21 books for an impressive collapse load-to-weight ratio of 916 to 1,375. Although the cardboard column demonstration is far from a rigorous scientific experiment, it provides an impetus to delve deeper into the mathematical analysis and design of thin-walled columns as a preamble to more complex stiffened panels as noted below. 4. Thin-walled column assignment As a basic design-built-test experience, students analyze, build and test two specified thin-walled column (stringer) design concepts with the cross-sectional configurations A & B as shown in Fig. 2. This activity is followed by each student designing a new stringer concept for improved performance and validating it through fabrication and testing.

Figure 2. Cross-sectional design concepts for thin-walled columns.

Each concept represents a multi-corner section that can be formed by bending a 4.5" wide by 17" long strip of 0.032" thick, 2024-T3 aluminum alloy sheet. The net length is 16" as a 0.5" portion at each end is capped to facilitate the axial loading process while simulating a fixed boundary condition. Since the geometric and material properties of the two concepts are specified, students can analyze each concept to predict its axial compressive strength based on the failure modes described in Section 2. The ! parameter in Eq. (4) takes the value of 11 for concept A and 17 for

0.75"

0.75"

0.75"

0.75"

1.5"

0.5"

0.5" 0.5"

0.5"

0.5"

1.0"

16"

End Cap End Cap Thin-Walled Column

(Concept A) (Concept B)

Page 26.1234.6

concept B. Since all other parameters have the same values for both, concept B is expected to have a higher crippling strength subject to limits17 imposed on the values found from Eq. (4). Since students use a break-forming (sheet bending) machine to bend the sheet metal into the desired angular shapes, they quickly realize the challenges of manufacturing, i.e., bending a flat sheet into a multi-corner section. An important consideration is the pace at which the sheet is bent. Given the cold-forming process, it is easy to bend the sheet too fast, which would result in cracking or splitting the sheet. Another important aspect of this exercise is a better understanding of the manufacturability requirements when it comes to selecting a dimension for each part of the cross-section. Although it is easy to draw a shape with any dimensions, its fabrication may be difficult or impossible using the specified sheet-forming process. Issues such as this can easily go unnoticed unless the manufacturing process is a part of the engineering design activity. The student fabricated column concepts A and B are tested using a servo-hydraulic testing machine in a laboratory. The measured axial compressive failure loads (min, max, mean) are compared with the analytical prediction for each design concept. Also, various response characteristics such as local buckling would be easy to observe during testing for comparison of the measured load to the predicted value. The geometric variability due to fabrication inaccuracies as well as the stochastic variability in the material properties results in variability in the measured failure loads. The experience with fabrication and testing of the specified concepts provides a good learning opportunity as students work to design their own thin-walled column (stringer) concepts to beat those in concepts A and B. Prior to delving deep into creating their own design concepts, students are encouraged to work with the bending machine to help identify the limitations that could impact their choice of geometry and dimensions. Guided by this prior experience, each student designs, analyzes, fabricates, and tests his/her own concept. Figure 3 shows a few symmetric design concepts produced by students in an attempt to maximize the compressive axial failure load of the thin-walled column, using the same size (4.5" x 17") strip of aluminum alloy sheet.

Figure 3. Examples of cross-sectional design concepts suggested by students.

There is no doubt that fabrication and testing increases the level of effort and time devoted to a single course assignment, both by the students and the instructor. However, the experience is beneficial for the students, not only in better understanding and retention of the analysis and design concepts discussed in the course, but also in gaining greater appreciation for the importance of other topics such as design for manufacturing and statistical variability.

0.4"

0.75"

0.5"

0.4"

0.4"

0.5"

0.5"

0.5"

0.5" 0.5"

0.25"

0.5" 0.5"

1.0"

Page 26.1234.7

5. Stiffened panel project The thin-walled column assignment described in Section 4 was continued for several years before the design-build-test component of it was removed and integrated into a comprehensive stiffened panel project. The project is one that includes many key elements in product design and development ranging from time management, teamwork, and communication skills to manufacturability, cost, and variability in production and performance, all collected into a practical engineering experience for the senior aerospace engineering students. The project learning objectives are as follows: • Enhance the understanding of the subjects studied in the course through development of

viable panel design concepts and analysis of their performance characteristics. • Apply the fundamental principles of optimization theory to an aerospace engineering problem

with a well-defined objective function and a set of design constraints. • Gain limited experience with sheet metal forming, hand tool operation, and manual assembly. • Become better familiar with laboratory testing as means of design validation. • Improve teamwork and communication skills.

Students, divided into teams of three, are given the following design problem: An L x H x 0.032" rectangular sheet is stiffened in the short direction by an unspecified number of identical formed stringers all made of 0.032" thick 2024-T3 aluminum (bare sheet). Stringers are attached to the sheet using uniformly spaced 1/8" diameter 5052 aluminum rivets (tmax = 350 psi and sult = 325 psi). Determine the shape, size, quantity, and arrangement of stringers as well as rivet spacing for a panel (shown in Fig. 4) optimized for X to support a design ultimate axial compressive force of Fmax for at least 3 seconds before failure. The project parameters have been altered over the years. For example, both large (L = 24", H = 18") and small (L = 12", H = 9") panels have been used. Similarly, the Fmax value is varied from year to year (12,000 lb to 22,000 lb) while imposing different limits on the total number of stringers (up to 8) that could be used to support the panel. The figure of merit X has also taken a simple form such as minimum weight to a more complex form maximum composite objective function given as ! = 100(!!!∗ + !!!∗ + !!!∗) (6) where !∗,!∗,!∗ represent separate normalized design criteria for strength, weight, and manufacturability, respectively, with !!,!!,!! as the corresponding importance factors. The normalized criteria in Eq. (6) are defined as !∗ = !!"#$ !!"# (7a) !∗ = 1−!!"#$% !!"# (7b) !∗ = 1− !! !!!!"# + 1− !! !!!!"# + 1− !! !!!!"# /3 (7c)

H

L

Figure 4. Schematic of a typical stiffened panel and loading.

Page 26.1234.8

where !!"#$ is the measured panel failure load, !!"#$% is the measured weight of the panel, !!"# is the expected panel weight using the total amount of material available for the stringers, !! is the number of stringers used, !! is the number of corners in each stringers (used as a measure of design complexity), and !! is the number of rivets used to attach all the stringers to the sheet to form the stiffened panel. The parameters in Eqs. (6) and (7) are altered from year to year to vary the preference given to each design objective and the limits on the number of stringers or rivets. Without delving too deep into details, a number of additional design and manufacturing requirements are also imposed such as the unacceptability of inter-rivet buckling prior to failure, manufacturability of the stringers using the available sheet-bending machine, maximum height of each stringer, placement of the edge stringers, and a maximum tolerance for each cross-sectional dimension of the stringer. Students are warned that “You must be able to build what you design!”. Each team is required to consider a minimum of two stringer design concepts (Zee, Hat, etc.). The panel is be analyzed and optimized for each of the two stringer concepts using one or two rows of fasteners. For example, two rows of fasteners can be used to have closed Hat stringers as opposed to open stringers with a single row of fasteners. The analysis and optimization activities lead to the generation of a simple plot for the optimum value of figure of merit for each combination of optimum stringer size and quantity for a selected stringer design concept. With the lower and upper bounds imposed on each dimension of the stringer, it may be impossible to support the prescribed load !!"# below a minimum number of stringers, even at max dimensions.

Figure 5. Sequence of activities associated with the panel project Ultimately, the quantity and size of stringers for each concept that provides the optimum overall panel design are identified. With the availability of this information, each team narrows its choice to only one stringer concept, size, and quantity, and proceeds to build three identical panels for testing. The individual steps are captured with more detail in Fig. 5. This process provides a

1. Form a team of three and select a team leader 2. Propose and evaluate potential stringer concepts 3. Pick two stringer concepts 1 and 2 4. Develop the panel analysis algorithm (and equations) for the stringer concept 1 geometry 5. Identify design variables, choose associated bounds, and formulate design constraints 6. Convert the analysis algorithm into a Fortran code 7. Integrate the Fortran code into the optimization shell program 8. Run the optimization code to find optimum stringer dimensions for a fixed number of stringers 9. Vary the number of stringers and repeat step 8 10. Return to step 4 and repeat for the stringer concept 2 11. Evaluate the optimum solutions for both stringer concepts and choose the one to build 12. Draw a CAD model of the selected panel with all dimensions and stringer quantity clearly specified 13. Cut sheet aluminum and fabricate stringers 14. Measure dimensions of fabricated stringers and calculate mean and standard deviation for each dimension 15. Mark and drill holes in the skin and stringers; fasten stringers to the skin 16. Weigh each completed panel for comparison to the predicted weight 17. Cut and mount loading caps on each panel for testing 18. Test each panel using a servo-hydraulic testing machine and record data 19. Compare measured data for various events (e.g., local buckling, failure) to the predicted values 20. Write and submit a group technical report following the specified format 21. Give an oral presentation of the project to the class

Page 26.1234.9

practical engineering experience where decisions are guided by such considerations as design performance, cost, manufacturability, etc. It should be noted that prior to students getting started with the panel project and following the steps noted in Fig. 5, they are introduced to all the key concepts noted in Section 2 with prior assignments enforcing each topic; taught the basics of design optimization theory and the use of the commercial optimization code (DOT by Vanderplaats Research & Development); provided examples of two Fortran codes for analysis and optimization of thin-walled columns; introduced to the procedure for coupling an analysis subroutine with the embedded optimization code and how to interpret the optimization results; given a demonstration of the laboratory equipment and their use; and guided frequently during the course of the project. Moreover, besides the course textbook, students receive eight handouts (> 200 pages) developed by the author. One handout is specifically focused on the basics of design optimization theory as well as the formulations for various constrained optimization problems and solution techniques. 5.a Teamwork and shared responsibility Teamwork and shared responsibility are emphasized throughout the project, which lasts approximately four to five weeks from start to finish. Each team member is required to contribute to various activities including the engineering analysis, design, and optimization aspects of the project. The team leader, selected by each individual team, assumes the additional responsibility of keeping the team on task. Through submission of weekly one-page reports, each team provides an update on the level of progress being made toward the successful completion of the project. 5.b Computer programming and associated challenges Prior programming experience tends to be rather limited for many of the students in this course. Although they are taught the basics of Fortran and Matlab programming in some of the earlier courses, for most, this is the first opportunity to develop a Fortran program with this level of detail. For most students, this is also the first exposure to design optimization, in general, and the use of a commercial optimization tool, in particular. To help reduce anxiety and improve understanding of computer programming, students are given two Fortran codes that focus on the analysis and optimization of thin-walled columns with specific geometry. Many of the equations used in the panel analysis also appear in the column analysis code. In addition to discussing the code in class, students are given access to the server where the example codes reside, so they can familiarize themselves with editing, saving, compiling, and executing the codes. Since its initial inclusion in the aerospace design course in 1999, the most time-consuming aspect of the design-build-test panel project has been the computer programming. Lack of appreciable programming experience, particularly in Fortran, means students have to devote considerable time to resolve common mistakes that cause compilation difficulty, and develop an ability to debug a code to identify and eliminate the source(s) of error. Through consultation with the instructor and teamwork, students gradually overcome the hurdle and complete the analysis and optimization programming, which is essential before they can proceed with the fabrication and testing parts of the project.

Page 26.1234.10

5.c Fabrication and testing experience Each student builds a panel based on the final specifications derived from the optimization process outlined earlier and captured in the CAD drawing. Students in each team work together in the fabrication process, but they are free to divide the tasks among themselves. The requirement is that each student should experience different aspects of the fabrication process. All students receive training on the proper use of the hand tools and other equipment used. For many students, this activity represents the first opportunity they have had to use hand tools (drill, rivet plier), cut and form sheet metal, or integrate so many different aspects of engineering into a single project. Given that a panel is to be loaded under axial compression, proper alignment is critical; otherwise, panels can be put under combined compression and bending that could cause premature failure. A fixture is developed to facilitate the panel setup by the students. However, a lab technician is responsible for operating the servo-hydraulic testing machine used to load each panel to failure. Before students are allowed to test their panels, they must submit a completed testing authorization form that provides various information including the predicted value for the figure of merit, panel weight, skin buckling force, stringer local buckling force, inter-rivet buckling force, and failure force. While the panel is being tested, students look for the initial occurrence of buckling in the skin. This event is followed by the local buckling of stringers and the eventual collapse of the panel. Photos taken during the fabrication and testing process are included in the project report and oral presentation. Figure 6 shows highlights of panel fabrication and testing.

Figure 6. Highlights of panel fabrication and testing.

5.d Grading procedure The panel project grade (20% of the course grade) for each student is based on the project report (50%), panel fabrication quality (10%), panel testing (10%), design quality (15%), and oral

Page 26.1234.11

presentation (15%). To ensure fair grading, each student indicates percent contribution on the report coversheet and signs the honor code. It is possible and fairly common for the different members of the same team to receive different grades for their shared panel project. The project report gives a detailed account of the activities conducted by each team. It describes the stringer design concepts examined and selected, the design optimization problem, structural and failure analyses performed, optimization results and discussion, summary of the fabrication activities, summary of the testing procedure, comparison of the predicted and measured values, and conclusions. The computer code along with a sample input and output file developed are included in the appendix. After the panels have been tested, they are evaluated by the students and submitted to the instructor together with their project reports. The project is culminated with a PowerPoint presentation by each team. Each team’s presentation is scored on the basis of time, content, and presentation style. Examples of the panels designed by the aerospace engineering students over the years are shown in Fig. 7.

Figure 7. Examples of panels designed by senior aerospace engineering students (post failure). A detailed spreadsheet is developed and used to assign grade to each aspect of the project. The project report is scored on the basis of completeness and accuracy in the analysis and computer programming. Each fabricated and tested panel is examined by the instructor for stringer fabrication and riveting quality (i.e., uniformity in stringers and rivet spacing). The stringer dimension variability is another factor considered. This is one area where students on the same team could receive different grades. As long as each panel has been properly tested, each student receives full credit for the testing part. The most detailed grading is associated with the design quality. Here, the individual metrics specified in the project assignment are evaluated. Close attention is given to whether the panel was able to withstand the specified load, the number of stringers used, stringer complexity (no. of corners), and the number of rivets used. The design quality subscores are normalized relative to the best one in each category in a given year. Each team leader receives a small bonus credit for their extra effort and responsibility.

Page 26.1234.12

5.e Project assessment and historical perspective Since its initial inception and integration into the aerospace structural design course in 1999, the panel project has been viewed by the students as one of the highlights of their undergraduate education in aerospace engineering at Mississippi State University. The project is very demanding and students devote considerable amount of time and effort toward its successful completion. Every year, a post-project discussion session is held with the students to discuss the problems encountered and the way they could be alleviated in the following year given that the course is taught only once a year. Over the years, significant steps have been taken to improve the fabrication and testing parts of the panel. Changing the size of the panel from 24" x 18" to 12" x 9" was partly aimed at reducing the time spent on producing each panel. The fabrication and testing experience also improved over the years through the purchase and use of more accurate shear and sheet-bending machines and access to advanced test equipment and data acquisition systems at one of the major research centers in the college of engineering. The computer programming has remained the most demanding part of the project, although the integration of programming exercises earlier in the semester has made a positive difference. As noted earlier, the panel project was designed with the goal of providing students a practical engineering experience. The diversity and scope of activities embedded into the project have proven effective in meeting that objective. This project has also added a unique feature to an otherwise theoretical engineering science course with the students putting to test what they have learned in a spirited team competition. 6. Students’ assessment of the panel project As a way of soliciting the students’ views on the panel project, an anonymous survey consisting of eleven statements, as shown in Fig. 8, was developed. Each statement was scored on a five-point scale: 1 – Strongly Disagree, 2 – Disagree, 3 – Neither Agree nor Disagree, 4 – Agree, and 5 – Strongly Agree. The results shown in Fig. 9 include all the collected surveys (total = 114) from five different years combined.

Figure 8. Student survey of the design-build-test project.

1. This project enabled me to apply my knowledge of math and solid mechanics to a real-world eng. problem. 2. This project improved my understanding of related topics taught in this course. 3. This project enhanced my ability to use the techniques, skills, and modern engineering tools (such as a

computer-based analysis program) necessary for engineering practice. 4. This project improved my understanding of sheet-metal fabrication, and gave me a better appreciation of the

impact of manufacturing on product design as well as dimensional variability in fabricated parts. 5. This project improved my understanding of structural testing and the factors that can lead to differences in

the predicted and experimentally-determined values. 6. This project enabled me to apply engineering judgment in evaluating a design. 7. This project enhanced my understanding of the behavior of aircraft structures. 8. This project improved my ability to work with others in a team. 9. This project improved my communication skills through writing a report and/or making an oral presentation. 10. This project is an essential component of this course. 11. Overall, this project provided a great experiential learning opportunity for me as an aerospace eng. student.

Please write your comments about any specific aspect of this project on the back. Page 26.1234.13

Respondents predominantly agreed or strongly agreed with ten out of eleven statements. The largest scatter is associated with the responses to statements 8 and 9 indicating mixed views on the level of perceived improvements in teamwork and communication skills. The overwhelming positive views to statements 1 through 7 indicate that the project served as an effective reinforcing mechanism for the subjects covered in the course while providing students opportunities to gain familiarity with basic sheet-forming fabrication and the associated manufacturing constraints. Responses to statements 10 and 11 also indicate the positive views most students have about the overall impact of this experiential learning activity.

Figure 9. Survey response statistics from a population of 114 students over a five-year period.

Unedited written comments provided by the students in the same group are as follows: • Those rivet gun hurt your hands, but it was worth it. I will always remember this project! • Extremely, strongly, very much agree [with no. 10] • I would recommend having a sort of review of what different failure modes look like on the

panel before testing the panels. • Perhaps in the future, when time gets short, decrease your requirement, i.e., 1 panel vs. 3 or

perhaps a report w/o presentation. • Dr. Rohani is an excellent teacher and this course is one of the few courses that allow us to

not only learn engineering science, but to be an engineer. The project could be shortened a bit, but otherwise the class is great. Test could be shortened a bit.

• Wait until all portions of work we turned in before peer evals. Also, peer evals are tricky (no one wants to give their friend a bad grade), so another method may want to be considered for monitoring team participation.

• It was very hard. We didn't have enough time to do a good job. We did not have much guidance. I hate working in the lab.

• The project allowed those who haven't had the chance to build things the chance to learn the meaningfulness of designing thing to be manufacturable.

• Need more help with FORTRAN programming. • Programming was hard so I do not know still now, but it was so great experience. Thank you. • It requires a lot of time. It should be considered a lab or extra credit hour, rather than part

of the class. • Scheduling gets rough at the end of the project, otherwise great experience.

0.0 1.0 2.0 3.0 4.0 5.0

1 2 3 4 5 6 7 8 9 10 11

Mean St. Deviation

Page 26.1234.14

7. Freshman experience with a simplified panel project The success of the panel project in the senior-level Aerospace Structural Design course motivated the integration of its simplified version into the freshman-level Introduction to Aerospace Engineering course taught by another faculty member in the Aerospace Engineering Department. Freshman students are divided into teams of three. Their task is to design, analyze, fabricate, and test panels that are made of poster board with stringers glued to the skin. They are given the dimensions of a “standard” panel that is 7" x 10" and supported by two closed Hat stringers. They must calculate the moment of inertia of the stiffened panel about its centroidal axis parallel to the skin. Their calculations are performed using both MS Excel and MathCad, partly to gain familiarity with the two software tools. Each team member builds a poster board panel to the given specifications. In addition, each team can design a “custom” panel that could have any number, size, or shape stringers. The goal for the custom panel is to maximize the failure load-to-weight ratio. By maximizing the moment of inertia, students examine different geometric shapes and sizes to design their custom panels. Similar to the standard panel, they build three samples of their custom panels for testing. Using a simple apparatus shown in Fig. 10, each panel is held in position and loaded by pouring sand into the container placed on top. The loading is increased gradually until failure. By measuring the weight of the sand and the container, the failure load is determined. The project is concluded with each team submitting a written report and giving an oral presentation. Typical poster board panels weigh around 0.06 lb to 0.1 lb and carry loads in the range of 60 lb to 100 lb prior to failure for a failure load-to-weight ratio reaching as high as 1,600. As in the case of the senior-level panel project, the freshman students gain experience with the immense load carrying capacity of thin-walled structures while developing a better understanding of the engineering decision-making and analysis involved.

Figure 10. Poster board panels in freshman introduction to aerospace engineering course.

Page 26.1234.15

8. Conclusions In-class demonstrations, hands-on experience with sheet-metal fabrication, and a multi-faceted class projects involving thin-walled structures were discussed as a way of increasing student engagement and providing practical engineering experience in an aerospace structural design course. The design-build-test panel project was used as an example of a comprehensive project commonly used in support of product design and development. With a limited scope, the project was designed to help students acquire deeper understanding of the concepts taught in course and greater appreciation for other important aspects of engineering such as communication skills, teamwork, time management, personal responsibility, and ethics. Manufacturing and testing elevated the analysis and design experience that would have been very difficult to achieve otherwise. Acknowledged by students as a challenging but rewarding experience, the panel project has had a positive impact on the undergraduate aerospace engineering education at Mississippi State University. References 1. Dennis, N. D., “Experiential Learning Exercised Through Project Based Instruction,” Proceedings of the 2001

ASEE Annual Conference & Exposition, Albuquerque, NM, June 24 - 27, 2001. 2. Prince, M., “Does Active Learning Work? A Review of the Research,” Journal of Engineering Education, Vol.

93, No. 3, pp. 223-231, 2004. 3. Kolmos, A., and de Graaff, E., “Problem-Based and Project-Based Learning in Engineering Education: Merging

Models,” In A. Johri and B. M. Olds (Eds.), Cambridge Handbook of Engineering Education Research, pp. 141-161, Cambridge University Press, 2014.

4. Smith, K. A., Sheppard, S. D., Johnson, D. W., and Johnson, R. T., “Pedagogies of Engagement: Classroom-Based Practices,” Journal of Engineering Education, Vol. 94, No. 1, pp. 87-101, 2005.

5. Meyer, K.F., Ressler, S.J., Lenox, T.A., “Visualizing Structural Behavior: Using Physical Models in Structural Engineering Education,” Proceedings of the 1996 ASEE Annual Conference & Exposition, Washington, DC, June 23-26, 1996.

6. Schmucker, D.G., “Models, Models, Models: The Use of Physical Models to Enhance the Structural Engineering Experience,” Proceedings of the 1998 ASEE Annual Conference & Exposition, Seattle, WA, June 28-July 1, 1998.

7. Vander Schaaf, R. and Klosky, J.L., “Show Me the Money! Using Physical Models to Excite Student Interest in Mechanics,” Proceedings of the 2003 ASEE Annual Conference & Exposition, Nashville, TN, June 22-25, 2003.

8. Vander Schaaf, R. and Klosky, J.L., “Classroom Demonstrations in Introductory Mechanics,” Journal of Professional Issues in Engineering Education and Practice, Vol. 131, No. 2, pp. 83-89, 2005.

9. Kresta, S.M., “Hands-on Demonstrations: An Alternative to Full Scale Lab Experiments,” Journal of Engineering Education, Vol. 87, No. 1, pp. 7-9, 1998.

10. Dollar, A., and Steif, P.S., “Understanding Internal Loading Through Hands-On Experience,” Proceedings of the 25. 2002 ASEE Annual Conference & Exposition, Montreal, Canada, June 16-19, 2002.

11. Sullivan, R. and Rais-Rohani, M., “Design and Application of a Beam Testing System for Experiential Learning in Mechanics of Materials,” Advances in Engineering Education, Vol. 1, No. 4, pp. 1-19, Spring 2009.

12. Ortmeyer, T. H., Cunningham, K, and Sathyamoorthy, M., “A Manufacturing Engineering Experiential Learning Program," Proceedings of the 2000 ASEE Annual Conference & Exposition, St. Louis, MO, June 18-21, 2000.

13. Tener, R. K., Winstead, M. T., and Smaglik, E. J., "Experiential Learning from Internships in Construction Engineering," Proceedings of the 2001 ASEE Annual Conference & Exposition, Albuquerque, NM, June 24 - 27, 2001.

Page 26.1234.16

14. Kolb, D. A., Experiential Learning: Experience as the Source of Learning and Development, Prentice-Hall, Englewood-Cliffs, NJ, 1984.

15. Rais-Rohani, M., “Experiential Learning in Aircraft Structures,” Proceedings of the 2003 ASEE Annual Conference, Nashville, TN, Jun 23-25, 2003.

16. Niu, M. C. Y., Airframe Structural Design, Conmilit Press, LTD., 1988. 17. Bruhn, E. F., Analysis and Design of Flight Vehicle Structures, S. R. Jacobs & Associates, Inc., 1973. 18. Gerard, G., Handbook of Structural Stability, Part IV – Failure of Plates and Composite Elements, NACA TN

3784, 1957.  

Page 26.1234.17