The Incredible Egg: Investigating the Design Education ... · generation of a requirements list,...

1Corresponding Author 1 Copyright © 2010 by ASME

Proceedings of the ASME 2010 International Design Engineering Technical Conferences &

Computers and Information in Engineering Conference

IDETC/CIE 2010

August 15-18, 2010, Montreal, Quebec, Canada

DETC2010-28817

THE INCREDIBLE EGG: INVESTIGATING THE DESIGN EDUCATION CHALLENGES

AND COMPLEXITY OF THE EGG DROP PROJECT

Scott Ferguson1

Assistant Professor North Carolina State University, Raleigh, NC

[email protected] (919) 515-5231

Heidi Klumpe Undergraduate Research Assistant

North Carolina State University [email protected]

John Turner Undergraduate Research Assistant

North Carolina State University [email protected]

ABSTRACT Designing a system that will protect an egg dropped from a

predefined height is a common experience for many K-12

students and undergraduates in engineering. Often presented in

the context of the scientific method, results from these

experiments are used to teach concepts of impulse, acceleration,

and impact modeling. When done in the classroom, students

are usually given a box of pre-defined supplies and a small time

frame with which to complete the design. But what educational

challenges and outcomes can be gathered from this experience

when the problem is tackled using the systematic design

process? In this paper, outcomes from the various steps of the

design process conducted over a six-week research project by

two high school students are presented. Results include the

generation of a requirements list, the creation of a functional

model, results of brainstorming sessions, concept analysis,

model validation through experimentation, optimization, and

final design testing. Challenges faced during each step of the

design process, and the surprising complexity of the problem,

are also discussed. Additionally, the challenges associated with

teaching design principles to high school students for a

multidisciplinary and multiobjective problem are addressed.

1. INTRODUCTION The egg drop project is an experience shared by almost all

K-12 and engineering students. Results from this experiment

are often used to teach concepts of gravity, impulse,

acceleration, and rudimentary impact modeling. Motivation for

using this project in K-12 education is the desire to help

promote better differentiation between science and technology.

By using concepts already familiar to students – everyone has

dropped and broken something, at least once – and

incorporating concepts associated with problem solving and

design, a better understanding for the role of technology in

society is developed. For students in grades 5-8, it is suggested

that activities be constructed such that “only a few well-defined

ways to solve the problem” are involved. To accomplish this,

multiple constraints are often imbedded into the problem.

However, these constraints can be relaxed for students in grades

9-12. The following outcomes and assessment criteria are

suggested by the National Science Education Standards these

students in the area of technical design [1, p. 192]:

Identify a product or design an opportunity: Students

should be able to identify new problems or needs and

to change and improve current technological design.

Propose designs and choose between alternative

solutions: Students should demonstrate thoughtful

planning for a piece of technology or technique.

Students should be introduced to the roles of models

and simulations in these processes.

Implement a proposed solution: A variety of skills can

be needed in proposing a solution depending on the

type of technology that is involved. The construction

of artifacts can require the skills of cutting, shaping,

treating, and joining common materials - such as

wood, metal, plastics, and textiles. Solutions can also

be implemented using computer software.

Evaluate the solution and its consequences: Students

should test any solution against the needs and criteria it

was designed to meet. At this stage, new criteria not

originally considered may be reviewed.

Communicate the problem, process, and solution:

Students should present their results to students,

teachers, and others in a variety of ways, such as

2 Copyright © 2010 by ASME

orally, in writing, and in other forms – including

models, diagrams, and demonstrations.

These outcomes are in addition to those originally defined for

students in grades 5-8. One item of note from this category is

specifically discussed in the context of the egg drop: [1, p.164]

Assessment activity: Following the egg drop activity,

students each prepare a report on one thing they

propose in order to improve their team's container and

how they would test the effectiveness of their

improvement.

To maintain a limited number of solutions, and minimize

budgetary impacts, the egg drop problem is typically

constrained by limiting students to a set of pre-specified

materials with which to complete their design. When conducted

in a classroom setting, time constraints also play a significant

role in shaping the outcomes associated with this experience.

But what happens when material and time constraints are

removed, or at least relaxed?

In this paper, the outcomes from two high school seniors

(now freshman undergraduates) following a systematic design

process conducted over a six-week research project are

presented. Results from this approach include the generation of

a requirements list, the creation of a functional model, results of

brainstorming sessions, concept analysis, model creation

through experimentation, optimization, and final design testing.

For each step of the design process, the challenges faced – from

both an education and application perspective – are discussed.

Additionally, insights into the additional problem complexity

when material and time constraints are relaxed are presented.

The outline of this paper is as follows. Section 2 describes

the motivation for introducing the students to a systematic

approach to engineering design, and discusses the major steps

of the design process. The results generated from the planning

stage of the design process are presented in Section 3, along

with student reflections and an assessment of the result with

respect to the National Science Education Standards. Similar

outcomes are then discussed with respect to the conceptual,

embodiment, and detailed design phases in Sections 4-6.

Finally, conclusions are presented in Section 7.

2. ADOPTING A SYSTEMATIC APPROACH TO

ENGINEERING DESIGN Conducting a Google search for references to the „egg drop

project‟ reveals 6.58 million results, with multiple high school

and university websites that discuss concept ideas, materials

lists, rules, and assessment criteria [2-4]. It is interesting to

note, however, that design assessment is almost always binary:

did the egg break – yes or no? While students are typically

evaluated on multiple criteria like weight, cost and success,

rarely does it seem that a true understanding of how close a

system was to success or failure is ever attained. Instead, it

often seems that students are making design decisions with

regards to performance criteria without comprehensive or

accurate predictions of system performance. Such decisions, or

conclusions on product architecture, appear to be more inquiry-

based in nature. That is, the students do not always know how

the system will perform until after the final result is observed.

One approach to gathering information that students are

introduced to early in their education is the scientific method.

The scientific method, as an approach, focuses on rigorous data

collection, with information being the resultant outcome.

However, when the desired output is more than information, a

design process is required. As defined by the Accreditation

Board for Engineering and Technology, engineering design is:

“. . . a decision-making process (often iterative), in

which the basic sciences, mathematics, and the

engineering sciences are applied to convert resources

optimally to meet these stated needs. [5]”

Thus, a design process focuses on rigorous information

synthesis as a means of developing a final product.

Regardless of the design process model followed [6-10] it

is well established that successful process navigation

necessitates the understanding of the interconnected (although

sometimes unexpected) relationships between subsystems and

components. According to the systematic design approach

proposed by Pahl et al. [6], the fundamental elements of design

include: planning and task clarification, conceptual design,

embodiment design, and detailed design. In conceptual design,

a designer searches for working principles and develops

concept variants. After these variant designs have been

evaluated against technical and economic criteria, a principle

solution is established. Embodiment design then builds upon

this principle solution through the selection of material and

form, validating the performance of the product through a series

of calculations and simulations. The form of this procedure is

similar to approaches proposed by Dewey [11], who suggested

that the problem-solving process consists of defining the

problem, identifying the alternatives, and selecting the best

alternative. Osborn [12] has further enforced this notion,

stating that problem solving involves fact finding, idea finding,

and solution finding.

For this project, two students were instructed to follow the

systematic approach to engineering design as described by Pahl

et al. [6]. This design process was chosen in particular because

of its wide acceptance within the engineering design community

and because the inputs and outputs associated with each step are

clearly defined. This was viewed as particularly important for

students who were getting their first exposure to regimented

instruction on engineering design principles. Additionally, this

design approach embraces the idea of establishing function

structure before establishing the physical form and layout of the

design. Given the motivation of exploring the consequence of

removing material and time constraints from the egg drop

problem, creating a functional model was considered a useful

endeavor to prevent user-implied constraints associated with

form and function.


The next section introduces the results obtained from the

first step of the design process: planning and clarifying the task.

For each step of the design process, this paper will discuss the

results generated for each task, student reflections and

assessment of results with respect to the National Science

Education Standards introduced in Section 1.

3. PLANNING AND CLARIFYING THE TASK The first task of the design process is to identify the design

objectives, and specify the properties that the design must have

or not have. The objective, inputs and outputs of this task are:

Task objective: Collect information about requirements

and identify constraints

Task input(s): Customer needs

Marketing information

Task output(s): Requirements list (demands and wishes)

Product proposal

For this project, the supervisor served as the customer. The

background supplied to the students was that they were required

to design a system that could protect one-dozen eggs being

dropped from a height somewhere between 25 and 40 feet.

They were told that the eggs were symbolic of medical supplies

being dropped from a hovering helicopter, and as such would

have a negligible initial horizontal velocity. Modifications to

the eggs were not permitted, nor could the system be a danger

to the user or anyone on the ground (unless they were standing

directly under the system at the point of impact). Remaining

demands and wants that needed to be defined were left to the

students to extrapolate from conversations with the customer

over the six-week period, or set to ensure “market

competitiveness”. The finalized product demands established

by the students are shown in Table 1.

Product demands

No egg modification allowed

System must carry 12 eggs

System cannot be volatile or hazardous

Survive falls between 25 and 40 feet

90% survival rate of cargo

Reusable

Straight drop (i.e. from hovering helicopter)

Table 1. List of product demands

Many of the product demands established by the students

came directly from the foundational story they were given as

motivation for the project. While mostly straightforward, one

interesting result was that they specifically defined a 90%

survival rate of the cargo. When asked about this, they

explained that a decision had been made to quantify

acceptability as 11 out of 12 eggs surviving on average. They

noted that improper packaging, manufacturing defects, and

other uncertain variables could prevent a 100% success rate

from being achieved across multiple drops.

„Wishes‟, or requirements that are taken into account

whenever possible, were also developed. As shown in Table 2,

these requirements were arranged from greatest to least

importance. The students felt that given the vagueness of the

problem statement supplied, there was not enough evidence to

make them absolute demands. Instead, they developed this list

as targets and criteria that they would like to achieve so long as

it did not affect mission success.

Product wants (most to least)

Cheap (minimum cost)

Able to withstand non-flat landing

Biodegradable / green materials

Waterproof / ability to float

Common components

Redundancy

Able to withstand initial horizontal velocity

Table 2. List of product wants

The desire to minimize cost and withstand a non-flat

landing both appeared as obvious „wants‟ for the designed

system. An unexpected outcome was the appearance of

biodegradability or the use of green materials. When asked for

insight into their thought process, it was revealed that Ms.

Klumpe was a proponent of, and interested in, green and

sustainable systems. Furthermore, the „want‟ related to

„common components‟ was an artifact of a discussion with the

students regarding research into product platforming [13-14].

Throughout the course of the project, the original

requirements list was slightly refined. A significant change to

product demands came in the form of more detailed definitions

of product failure. Under this new definition, failure of an egg

was now considered across three criteria:

1. More than one crack (includes "spider web" cracks)

2. Lengthwise crack longer than half egg length, or

width-wise crack one quarter the circumference

3. Leaked fluid/damaged yolk sac

In assessing the requirements list, and discussing the

process with the students, it was clear that a major hurdle they

faced was assigning the different requirements to an individual

designer. They felt that they did not know enough about the

structure of the system to make that determination at this phase

of the design process. Pahl et al. [6, p.157] also note that it is

common for factors like low-cost production and reusability to

be accidently left off a requirements list because it is an

assumed requirement. While these factors were included in

Table 1 and 2, it is significant to note that ease of assembly and

quick turn-around time did not make these lists. In discussing

the end product with the students, both strongly felt that these

factors were important elements of their design, but were not

explicitly incorporated into the requirements list.


Overall, the students successfully completed the assessment

criteria of „identifying a product or designing an opportunity‟.

No major product demands were ignored, and sufficient insight

was developed to progress to the conceptual design phase.

Student feedback about this phase of the design process

commented on how when first making the requirements list, it is

easy to define what you think needs to happen, but in reality the

process is much more complex. Additionally, it was felt that the

least important „wishes‟ were too hard to incorporate into the

system for the first generation of the design. They believed that

with additional redesigns, it would be possible to account for

these additional criteria.

Having identified the requirements of the product, the next

step in the design process is arriving at a conceptual design.

This is the focus of the next section.

4. CONCEPTUAL DESIGN Understanding the requirements that the system must meet

allows a designer to begin the search for a working concept.

This is completed by translating consumer demands into the

engineer‟s language of potential problem and solutions.

Research and brainstorming are used to establish system

functionality and define principles that meet this function. The

objective, inputs and outputs of the conceptual design phase

are:

Task objective: Arrive at a principle solution

Task input(s): Requirements list

Task output(s): Function structures

Working principles

Concept variants

The first challenge the students faced in the conceptual

design process was extracting generalized functional

requirements from the requirements list. To help the students

arrive at a generalized functional mode, two afternoons were

spent discussing the notion of functional modeling [9, 15].

Functional modeling is a technique that uses a common design

language to model the functionality of a product or process. A

black box model of the system is created using verb-noun pairs

to yield functional chains. These functional chains present a

designer with information regarding where system interactions

exist and what inputs and outputs are required of each

functional block.

Three principle functions of the system were defined:

1. Slow down fall (decrease Vf)

2. Cushion/support eggs (energy-absorbing insulation)

3. Extend impact time/absorb impact (crush zones,

airbags, etc.)

4.1 ESTABLISHING FUNCTION STRUCTURES Attempts at constructing functional models by both

students are shown in Appendix A in Figures A.1 and A.2.

Examining Figure A.1, the upper level functionality that was

defined includes:

Dissipate kinetic energy (KE)

Minimize impact force

Hit target accurately

Load / remove cargo easily

Maintain reusability

Stabilize system (in flight)

Withstand dynamic conditions

The number of higher-order functions associated with this

system speaks toward the complexity inherent in egg drop

projects when constraints are relaxed. Impact problems like

this require concepts associated with energy, linear momentum

and constitutive models. Students must dissipate the kinetic

energy of the egg at impact by using principles of work to exert

a force on the “insulating materials” over some distance. From

a momentum perspective, the velocity at impact must be

dissipated to remove all momentum, requiring a larger impact

time to reduce the magnitude of the force.

At the K-12 level, many teachers will establish constraints

that prevent parachute-like items from being used to reduce the

size of the solution space. However, from a functionality

standpoint, this would not remove the higher-order functions of

„Withstand dynamic conditions‟ and „Dissipate KE‟, as other

feasible solutions could be generated to achieve the same

functionality. The complexity of the problem is also seen when

investigating „Minimize impact force‟. Here, the student

defined two lower level functions, „Maximize t‟ and

„Minimize F‟ from the impact equation of:

t

mvF

(1)

In Figure A.1, basic solution principles are defined for each

lower-level function. In trying to maximize the impact time, the

student suggests attaching a malleable surface or collapsing a

section of the system. To minimize the force at impact, the

student suggests adding insulation, a large base, modifying the

orientation of the egg, and adding a suspension system.

The functional model created in Figure A.2 doesn‟t follow

standard convention of verb-noun pairs, instead attempting to

define independent variables, considered attributes, and

potential attribute groupings in as generalized a manner as

possible. This model acts as more of a combination of

functional model and requirements list, in that it accounts for

the cost of the eggs and environmental friendliness. However,

there are some aspects of this model that do establish

generalized functionality, providing some value.

Discussing the functional modeling process with the

students after the project was completed revealed that they were

generally happy with the outcome of the process. They

believed that creating a generalized functional model helped

decompose the problem into smaller, more manageable


components. Feedback was provided that functional modeling

helped with the brainstorming process, as it was easier to tie

product components together to achieve desired results.

Creating a functional model for a product can be a

significant challenge, even for a designer with prior experience.

A main challenge that often arises is when the size and

complexity of the model becomes overwhelming even for the

most rudimentary product. However, the feedback from this

project tended to look favorably upon the experience, even if

the model was not generated correctly. The information and

understanding gained from this process proved to be something

that the students would refer to throughout their six-week

project as they considered possible configuration changes.

4.2 SEARCHING FOR WORKING PRINCIPLES Having developed a generalized function structure, the next

step of the process was to search for working principles that

provided desired functionality. To facilitate idea generation, the

6-3-5 method of brainstorming [6, p.85] was used in a weekly

lab meeting. The two students worked in conjunction with 4

graduate students who are part of the System Design

Optimization Laboratory at NC State. To complete the 6-3-5

method, each student created 3 unique concepts, using

sketching, labels, and phrases to illustrate their idea. After a

pre-determined amount of time, each student passed the sheet of

paper in front of them clockwise. Each student then added

additional details to the three new concepts in front of them,

with the restriction that nothing could be removed from the

design. This process resulted in 18 designs, with a subset of

those results shown in Figure 1.

Many of the results from the brainstorming session

provided concepts that were realistically infeasible. All

students involved in the brainstorming process, however, had

positive comments about the experience. Ms. Klumpe also

recently mentioned that she taught her group members this

brainstorming process for a project in E101 at NC State. It was

also considered beneficial to set aside a specific block of time

for the sole purpose of brainstorming. Comments were made

that while students often have the intention to go through a

brainstorming procedure, it is commonly overlooked or pushed

aside in the interest of “getting to building more quickly”.

While the results of the brainstorming session did provide

mostly infeasible or impractical concepts, the overall trend

provided insight into the final principle solution. This is

discussed in the next section.

Figure 1. Results of 6-3-5 method of brainstorming

4.3 PRINCIPLE SOLUTION AND ASSESSMENT After analyzing the trends in the result obtained from the

brainstorming session, the students defined a principle solution

as shown in Figure 2. Arriving at this decision in the conceptual

design process was aided by analyzing the results of the

brainstorming session using rudimentary technical and

economic criteria. The requirement of using these criteria at

each phase of the design process is another reason why the Pahl

et al. [6] approach to engineering design was selected.

After considering the technological need and economic

impact of the different variant solutions, the students worked

together to establish a principle solution that took the strongest

elements from the possible solution set. It is interesting to note

that while the principle solution is by no means a unique or

Insulation

Crush Zone

Egg

Parachute

or

Accelerometer

Figure 2. Resultant principle concept


novel solution, the procedure by which they arrived here was

well documented and allowed them to validate their decision-

making process.

By this stage of the design process, the students had begun

to consolidate their ideas and choose between alternative

solutions. While these decisions were made at the very highest

level of product form, they were facilitated by evaluating

possible functionalities against technical and economic criteria.

This approach is significantly different than a “tinker-toy”

approach to product design. Now that a principle solution has

been developed, attention turned to giving it physical form.

This is the focus of the next section.

5. EMBODIMENT DESIGN The goal in embodiment design is to develop the solution

principle to a point where it can be prototyped, tested, and

manufactured. This is also the most difficult part of the

engineering design process for students because it requires a

culmination of their engineering education. It is at this step that

students must combine abilities in architecture determination /

component selection, model building, engineering analysis,

prototyping, and experimentation to arrive at a definitive layout.

This layout develops the system to a point where “a clear

check of function, durability, production, assembly, operation,

and costs can be carried out [6, p. 227]”. The objective, inputs

and outputs of the conceptual design phase are:

Task objective: Arrive at a definitive layout

Task input(s): Principle concept

Task output(s): Preliminary layouts

Analysis of proposed concepts

Optimization of design

Final layout

To successfully arrive at a definitive layout, the next step of

the project was to define the arrangement of the system from a

spatial perspective and determine component shape and

material. However, to validate the final configuration it was

necessary to iterate between periods of analysis and synthesis.

The first phase of this approach involved specifying the design

variables, constraints, and objective functions associated with

the problem.

The design variables associated with the proposed solution

principle are shown in Table 3. For each major component

(egg, insulation, etc.) there are multiple design variables that

must be defined. For example, while the students were told that

large eggs would be supplied for experimentation and testing,

no comment was made about the color of the supplied egg.

Therefore, one decision the students had to make was whether

they should choose a brown egg or a white egg – a binary

choice. In addition to egg color, the students also had to make

an integer-based decision regarding the orientation of the egg

(pointed side down, pointed side up, egg sideways), and the

location of the egg in the container (x, y coordinates as

continuous variables).

Similar decisions had to be made for other components as

well. For instance, when determining insulation the design

variables include: material, material thickness, volume of the

material in total, and whether insulation should be included in

the design. When considering 12 eggs, there are technically 72

different design variables that must be accounted for. If only

one egg was to be placed in the system, 28 design variables

would still be needed to define the problem. Additionally, for

the manner in which the problem was described there are 2

specific constraints, as shown in Table 4.

Component Design variable Type

Egg (x12) Color (brown or white) Binary

Orientation Integer

Location in container (x, y) Continuous

Insulation Material Integer

Material thickness Discrete

Volume (x, y, z) Continuous

On / Off Binary

Parachute Shape Integer

Volume (x, y, z) Continuous

Material Integer


On / Off Binary

Crush zone Number of components Integer

Material Integer


Area (x, y) Continuous

On / Off Binary

Container Length Discrete

Width Discrete

Height Discrete

Material Integer


Table 3. Component design variables and variable type

Constraint

11 of 12 eggs must survive each drop

Eggs and insulation must fit within container

Table 4. Problem constraints

From an optimization standpoint, there are six design

objectives that can represent the problem. These objectives are

shown in Table 5. Here, three objectives are to be minimized

(cost, weight, size) while three objectives are to be maximized

(survivability of eggs, repeatability, reusability). Furthermore,

this is a mixed-variable optimization problem whose solution

requires an understanding of multiobjective optimization.


Objective function Goal

Cost Minimize

Weight Minimize

Size Minimize

Survivability of eggs Maximize

Repeatability Maximize

Reusability of system Maximize

Table 5. Objective functions

5.1 INITIAL MODEL DEVELOPMENT FOR INSULATION Understanding the design variables associated with the

problem, however, is only one element of the overall puzzle.

Without models to evaluate the performance of different

configurations, ensuring product success is nearly impossible.

To begin the process of model development, the students first

turned their attention to the primary component of this project –

the egg. The first question to answer about the egg was how

much impact force was required to produce egg failure as

defined in Section 3.

An experimental study was performed to determine the

significance of two design variables associated with the egg:

color and orientation. Eggs were dropped from various heights

onto a paved parking lot (similar in texture to the ground used

for final product testing) using a full factorial experiment. As

shown in the results in Table 6, the pointed end of an egg is

most able to withstand the greatest force before failing.

Additionally, brown eggs were found to be slightly stronger

than white eggs. Not shown in this table is that an egg turned

on its side was found to crack at significantly smaller heights,

which was an expected result.

Egg color Orientation Average height of

failure (cm)

Brown Pointed end 3.75

Brown Rounded end 2

White Pointed end 3.6

White Rounded end 2

Table 6. Average failure height of unprotected egg

In addition to determining the average height of failure for

an unprotected egg, this experiment also allowed for the

estimation of impact time. References were found that the

pointed end of an egg was able to withstand 50 Newtons of

force, on average [16-17]. Using the mass of an unprotected

egg with energy and momentum equations resulted in an impact

time of 0.002 seconds to be calculated. This result was then

used as an estimated starting point for prototype system testing.

Developing an analytical or experimental model to predict

the performance of different insulations proved to be a greater

challenge. Despite numerous research attempts it was

concluded that material and performance properties for

household items – such as bubble wrap, Cheerios, and

Styrofoam – were not readily available. Insulation materials

studied during the course of this experiment included:

Jell-o

Cornstarch and water

Styrofoam

Cheerios

Bubble-wrap

Shredded paper

The objective of these tests was to quantify the energy / work

absorbed by different thicknesses of each material.

The first experiment designed to capture these properties

used a three-egg container dropped at 1-foot intervals. The

maximum drop height of the apparatus was 17 feet. In setting

up this experiment, insulation was placed only on the bottom of

the container, an assumption that worked well for drops from

low heights. As the drop distance increased, the container had a

tendency to tumble on impact, making it difficult to determine

which forces caused egg breakage.

To reduce the error associated with testing, the students

developed a guide-wire system to control the motion of the

container during free-fall in an effort to reduce the uncertainty

associated with force direction on impact. As shown in Figure

3, the system was constructed out of two wooden boards,

fishing line (low friction), and eyelet screws. The line length

was designed to be adaptable to reduce the overhead associated

with testing at various heights, and was secured to provide

consistent tension. The bottom board was later secured to the

ground as initial tests saw this board bouncing on impact.

Reel of

Fishing Line

Egg

Insulation

Rubber mat to increase ? t (improve

accelerometer readings)

Figure 3. Initial test rig for insulation property

determination

Student feedback regarding their testing system noted the

fixed height of the system sometimes prevented egg breakage.

To solve this issue, very thin metal plates, weighing 216 and

244 grams respectively, were added to the container with the

hope of increasing impact force. Also, when resetting the

system, thin plastic straws that were added to the container to

guide the system would get damaged by the taut fishing wire.

This made the straws more susceptible to catching, resulting in

increased friction. Finally, a major complaint was the time-


consuming nature of set-up, especially after an egg broke within

the container.

5.2 INSULATION MODEL USING AN IMPACT

ACCELEROMETER The lengthy set-up time associated with cleaning up and

replacing eggs that broke during testing led to a desire for a

more robust solution. The USB Impact X250-1 data logger was

purchased from Gulf Coast Data Concepts [18] to measure the

acceleration of the system for various insulation materials and

thicknesses. This system was capable of measuring +/- 250g in

all 3-axes, had an internal battery, and interfaced with a

standard PC using a USB port.

Given the size of the accelerometer (4” x 1” x 1”), the egg

was removed from the test rig and the accelerometer was

secured in the container‟s egg cavity. The height of the

accelerometer left room for insulation only on the container‟s

bottom and sides. Data collection was now more efficient, but

some technical issues needed to be addressed before the data

could be validated. Some of these issues included that:

the amplitude of the noise could be one or two g‟s

the accelerometer had to be zeroed correctly before

each run

the device needed to be secured to ensure readings

along a single axis to prevent confounding

During accelerometer testing, the sensor occasionally

reached its upper limit for certain design configurations. A

rubber mat was added to extend the impact time. Plans were

then made to establish the quantitative effect of the mat on the

observed acceleration. However, the mat failed to significantly

alter the results and was removed to ensure consistency.

Experimental results using the impact accelerometer when

0.25”-thick Bubble wrap was used as the insulating material is

shown in Figure 4.

Figure 4. Accelerometer data for

0.25”-thick Bubble wrap as insulation

Results from the accelerometer provided insight into the

number of g‟s experienced by the sensor for different material

types and thicknesses. Future model development will include

modeling the sensor as a point mass that is connected to

insulating materials modeled as springs and dashpots. The

acceleration data collected will then be used to determine the

effective spring coefficient and energy dissipating element for

each material at a given thickness.

5.3 INSIGHTS FROM INSULATION MATERIAL

TESTING Many insights about the practicality of different insulation

materials were drawn from the experimental testing. Jell-o,

although relatively successful at protecting an egg from large

falls, was considered too messy and difficult to replace.

Cornstarch and water were removed from consideration due to

the increased weight when compared to other insulation

options. Styrofoam was found to be effective and reusable, with

the biggest complaint being the mess made during construction.

To ensure continuity between test runs, it was necessary to

rotate and replace the Cheerios, equally distributing the broken

and unbroken bits. This life-cycle failure contributed to

Cheerios not being used in the final design. The students also

struggled with reliably calculating the thickness of the shredded

paper, due to variations in packing. Additionally, knowing

when to replace this material was a concern.

As time constraints prevented determining the effective

spring coefficient and energy dissipation elements of the various

materials, experimental data was used along with energy and

impulse equations. Here, the students made the simplifying

assumption of an impulse time of 0.002 seconds to calculate the

amount of force absorbed by a specified thickness of insulating

material, as shown in Figure 5. Overall, the following results

were collected for a given material at a given thickness:

average break height in meters

force absorbed per gram of material

force absorbed per dollar of material

Figure 5. Results from experimental testing

5.4 PARACHUTE AND CRUSH-ZONE TESTING Despite severe time limitations at this point in the six-week

project, the students attempted parachute and crush-zone

testing. The conclusion drawn about parachute testing was that

no substantial, quantifiable results could be obtained in the

allotted time. The best result from limited experimentation was

an upside-down, plastic grocery bag attached to the container.

For crush-zone testing, four concepts were tested, two of

which are shown in Figure 6: airbag, angled accordion, spiral

accordion, and cardboard tubes. Between tests, the airbag was

replaced if damaged, and the different accordions were


“fluffed” between trials. From testing, it was revealed that the

greatest failure of the angled accordion was the system‟s

inability to fall straight, causing alignment issues at impact.

5.5 OPTIMIZATION OF PROPOSED CONCEPTS Despite weeks of experimental testing, there was not nearly

enough data generated to create a full-fledged optimization

model for this problem. However, there was enough material

test data to facilitate informed and improved decision-making

with regards to the insulation material. To aid this decision-

making process, background material on optimal alternative

selection when considering multiple attributes was introduced

[19]. This experience allowed the students to learn about the

potential pitfalls associated with ranking, scaling, and

normalizing attributes. During these discussions they were also

introduced to the concept of optimization, Pareto optimality

[20], and use of utility theory in engineering design [21].

The idea that attributes can be weighted was leveraged to

create an interactive optimization algorithm for insulating

materials that required replacement (bubble wrap, Cheerios,

etc.). For a given insulation thickness, the amount of material

replaced per drop can be varied. The amount of material

replaced directly influences the overall cost of the insulating

material per average drop. However, replacing larger amounts

of material also increases the probability of system success. By

assigning weights to the attributes of cost and system reliability

(not breaking an egg), the utility of the proposed scheme was

calculated. An example of this approach is shown in Figure 7.

Figure 7. Optimizing the amount of

material replaced per drop

5.6 FINAL LAYOUT Arriving at a definitive layout for the system occurred

during the last week of the six-week project. While multiple

experiments had been run in an attempt to model the effect of

different design variables on system performance, the amount

and fidelity of the information gathered was not comprehensive

enough to develop equations that described the entire system.

Instead, decisions on the final design were based on two types

of student experience: 1) experience gained from working with

different materials during the experimentation phase, and 2)

past experience with egg drop projects, discussions with other

students, and a mental model of what the system should look

like gathered from outside research.

In addition to performance criteria, ease of use was

specifically targeted. When discussing this decision with the

students, it was determined that they were interested in running

multiple tests as quickly and easily as possible. This decision

was motivated by the need for final test results to be used in a

poster symposium that occurred at the end of the six-week

research experience. The final design is shown in Figure 8.

Figure 8. Final design layout

This design primarily used Styrofoam as the insulating

material. Shredded paper was used to cover the top of the eggs,

as it was concluded that minimal forces would be seen at this

location. Instead, the paper was used to securely fasten the eggs

in their defined slots during free-fall and to allow for ease of

access during testing set-up. Airbags were placed at the corners

of the container, and a plastic grocery bag served as the

parachute. Additionally, holes were cut in the container in an

effort to minimize weight

Final testing showed that the final design successfully

completed the defined requirements listed in Table 1. During

multiple tests from 40 feet, an average of 95% of the eggs

survived free fall. To test the “optimality” of the system, the

free fall height was increased in increments of 10 feet. At a

Figure 6. Experimental designs for crush-zone testing


height of 50 feet, 2-3 eggs on average would fail, indicating that

the thickness of the material selected was correctly chosen for

the mission. Complete system failure occurred when dropped

from 80 feet. Here, a gust of wind added horizontal velocity to

the system, causing it to collide with the fire escape used to test

the system at the 60 foot mark. This collision caused the system

to tip, and half of the eggs were ejected to experience

unprotected free-fall. The system landed upside-down,

breaking the remaining eggs at impact.

5.7 ASSESSMENT OF EMBODIMENT DESIGN TASKS The embodiment design phase was by far the most difficult

task for the students to complete. The assessment criteria

defined in Section 1 state that when choosing between

alternative solutions, students should be introduced to the roles

of models and simulations. As part of this task, students had to

generate models for multiple components: the egg, insulation,

parachute, and crush zone. Data for these models came from

experiments created by the students to extract this data. While

the information gathered provided useful insight into the

problem, the six-week period was not enough time to plan,

develop, and run the necessary experiments needed to create a

comprehensive model.

In reality, the students ended up with two design projects.

In addition to designing the system to protect the eggs, the

students were also required to design and validate their testing

mechanisms. Interviews with the students revealed that they

enjoyed the process of gathering data for the problem, as they

felt it gave them a better understanding of how the different

components of the system worked. Much like the design

process, they felt that successfully developing a testing scheme

was iterative in nature. While the functional model gave them

suggestions for component layout, failures in testing provided

insight into what did or did not work. They were also surprised

by how much information they gained from failures in the

testing approach itself. Often, they were able to gather data on

system phenomena that they did not intend to capture because

of a failure in the testing mechanism or a poorly designed

experiment.

In achieving the final design, the students believed that they

gravitated toward focusing on two objectives: reusability and

survivability. A non-specified system requirement that received

much attention was reset time. Also, despite the limitations of

gathering data for model development, listing and exploring the

design variables associated with this problem showed the

students how much control they had over system performance.

They felt that in many high school projects they never

completely understood of all the design variables associated

with a problem, which often led to a disconnect in

understanding performance changes.

Finally, the students were surprised at how well the

“simple” final design performed. When compared to their

brainstorming session, which resulted in many infeasible or

impractical designs, they felt their final design lacked a degree

of novelty. However, by creating system models and analyzing

the data, they were able to predict the performance of their

design against the criteria that it needed to meet. By comparing

this design against other concepts, they felt that they had made

the best-informed decision possible. Implementation of the

proposed solution was also achieved by constructing a final

version of the system and conducting drop tests from multiple

heights. It was also interesting to note that Styrofoam was

chosen by the student who originally was a proponent of using

green materials – a listed „want‟ in Table 2.

6. “DETAILED” DESIGN The end result of the detailed design phase is product

documentation. In this light, the assessment criteria of

communicating the problem, process, and solution will be

addressed. As part of this six-week research experience, the

two students were required to present their work at a poster

symposium on NC State‟s campus. Their individual posters

discussed the culmination of their research efforts and allowed

them to discuss their experience with other student researchers

and university faculty. Procedure logs, documentation, and

reports were also collected by the advisor to document the

experimental efforts and lessons learned. They also contributed

to the writing of this paper.

Additionally, the students performed a self-assessment

activity on the work that they completed. As stated in Section

1, students must also prepare a report on an aspect of the system

they would improve. This activity was completed by develop

four alternative solutions to the testing apparatus originally

presented in Section 5.1. Two of these proposed alternatives

are shown in Figure 9 and outlined below.

Swinging pendulum

Variability: change in mass, or change in elevation, changes

the absorbed kinetic energy

Advantages: Small size; easier to reset; can apply a point force

Egg

Pendulum Stop

Pendulum Stop

Insula

tio

n Mass

Swinging pendulum

Egg

Sto

p

Insulation

Sto

p

Mass

Inverse drop

Figure 9. Alternative testing apparatuses


Disadvantages: (1) May require large masses to get data that

approximates high drops; (2) have to minimize egg movement

(so instead absorbs kinetic energy, rather than transfers it)

without “sandwiching” and crushing the egg, (3) does not

provide linear motion after initial impact, (4) may measure

compressibility of material more than energy absorption.

Inverse Drop

Variability: change the mass or change the initial height to

change the falling mass‟s final velocity.

Advantages: Small size; easier to reset; can apply a point

force; linear movement throughout impact; (the mass could

easily be replaced with the egg, although a larger unit may be

required to get adequate height)

Disadvantages: (1) May require large masses to get data that

approximates high drops; (2) have to minimize egg movement

(so instead absorbs kinetic energy, rather than transfers it)

without “sandwiching” and crushing the egg; (3) may measure

compressibility of material more than energy absorption; (4)

gravity‟s direction parallel to the force being applied may

increase the complexity of the math.

7. CONCLUSIONS AND FUTURE WORK The traditional egg drop project is a common experience

for many K-12 students and undergraduates in engineering. To

reduce the feasible design space, students are commonly given a

box of pre-defined supplies (material constraints) or constraints

placed on allowable components. In this paper, the true

complexity of the problem is investigated when these

constraints are removed, or at least relaxed. When a system

designed to protect 12 eggs is considered, the possibility exists

for 72 different design variables define the problem. If a

container is designed for only one egg, 28 design variables

would still be needed to define the problem. In addition to this

complex number of design variables, six unique performance

objectives must also be considered – 3 of which are to be

minimized, 3 of which are to be maximized. Therefore,

students are faced with a constrained mixed-variable

multiobjective optimization problem.

Assessment of student outcomes from this project show

that following a systematic design process satisfies the

suggested criteria defined by the National Science Education

Standards, and helps students make informed configuration

decisions by assessing against technical and economic

conditions. Discussions with the two students revealed that they

thought it was very important to go through the steps of the

design process, as they had a better understanding of how

design variables in a problem affected system performance. In

developing a requirements list, the students believed that they

had tangible goals - or things to solve for - where following a

scientific method is typically more of a binary response to

answer a hypothesis. In developing models to assess the

performance of their proposed concepts, they developed an

increased understanding of the multidisciplinary aspects of

system and how component interactions can drastically impact

system performance. Students also expressed that the

“exciting” parts of engineering design took about 20% of their

six-week window, while the remainder of the time was spent

developing, running, and analyzing experimental data.

Furthermore, they found that important information could be

learned not only when the system failed an experiment, but also

when the experiment itself failed and revealed unintended

information about system performance.

Testing of the final design showed that even without

comprehensive models of the system, the container was not

drastically over-designed. The two students associated with this

project quickly learned complex engineering design principles,

and were able to adopt them into their project. They were

excited by the multiobjective and multidisciplinary nature of the

problem, and quickly identified future challenges associated

with this project. The categories for these aspects of future

work include:

Data Analysis

What is the relationship between the accelerometer‟s

reading and the egg cracking?

How much uncertainty is associated with the readings

over a large number of drops?

How does the surface area of the container bottom

change accelerometer readings?

Is there a way to test the accelerometer from larger

heights and then scale the system? Can a metamodel

be fit to this data?

How does the performance of the system degrade over

time?

Configuration challenges

What would be the ideal characteristics for a (fictional)

material for the system?

How does the multi-egg layout alter performance

outside of increasing system weight?

How do we best measure the interactions between the

components? How about between the materials?

How do you quantify requirements like

biodegradability rather than assigning an arbitrary

value?

ACKNOWLEDGMENTS The authors would like to acknowledge the support

provided from the NC Space Grant and the NC State Reaching

Incoming Student Enrichment (RISE) Program

REFERENCES [1] National Committee on Science Education Standards and

Assessment, National Research Council, 1996, National


Science Education Standards, National Academy Press,

Washington, DC.

[2] Davis Creek Elementary, 1998, “5th

Grade Eggdrop

Project”, <daviscreek.cabe.k12.wv.us/eggdrop.html>

[3] “Egg Drop Project”,

<http://www.ciesc.k12.in.us/ProfessionalDevelopment/Sea

m/documents/physics/EggDropProject.pdf>

[4] Belgrad, S., “The Amazing Egg Drop”,

<http://www.csun.edu/~sb4310/The%20Amazing%20Egg

%20Drop.htm>

[5] ABET Engineering Accreditation Commission, 2008,

“Criteria for Accrediting Engineering Programs,” ABET,

Inc.

[6] Pahl, G., Beitz, W., Feldhusen, J., and Grote, K. H., 2007,

Engineering Design: A Systematic Approach, 3rd

Edition,

Translated by Ken Wallace, Springer, London.

[7] Suh, N., 2001, Axiomatic Design: Advances and

Applications, Oxford University Press, New York.

[8] Pugh, S., 1991, Total Design – Integrated methods for

Successful Product Engineering, Addison-Wesley,

Massachusetts.

[9] Otto, K., and Wood, K., 2001, Product Design: Techniques

in Reverse Engineering and New Product Development,

Prentice Hall, New Jersey.

[10] Cross, N., 1989, Engineering Design Methods, John Wiley,

New York.

[11] Dewey, J., 1933, How We Think, Heath, Lexington,

Massachusetts.

[12] Osborn, A. F., 1963, Applied Imagination: Principles and

Procedures of Creative Problem Solving, 3rd

edition,

Scribners, New York.

[13] Simpson, T.W., Siddique, Z., and Jiao, J., eds, 2006,

Product Platform and Product Family Design: Methods

and Applications, Springer, New York, NY.

[14] Jiao, J., Simpson, T. W., Siddique, Z., 2007, “Product

Family Design and Platform-based Product Development: a

State-of-the-art Review,” Journal of International

Manufacturing, 18(1): 5-29.

[15] Stone, R. B. and K. L. Wood (2000). "Development of a

Functional Basis for Design." Journal of Mechanical

Design, 122(4): 359-370.

[16] Tung, M. A., Staley, L. M. & Richards, J. F., 1968, “Studies

on egg shell strength, shell stiffness, shell quantity, egg size

and shape,” British Poultry Science, 9:221-229.

[17] Voisey, P. W. & Hunt, J. R., 1967, “Relationship between

applied force, deformation of egg shells and fracture

force,” Journal of Agricultural Engineering Research, 12:

1-4.

[18] Gulf Coast Data Concepts, LCC, 2009, “USB 3-axis Self

Recording Impact Accelerometer,

<http://www.gcdataconcepts.com/x250-1.html>

[19] See, T. K., Gurnani, A., Lewis, K., 2004, “Multi-Attribute

Decision Making Using Hypothetical Equivalents and

Inequivalents,” Journal of Mechanical Design, 126(11):

950-958.

[20] Pareto, V., Manuale di Econòmica Polìttica, Società

Editrice Libràia, Milan, Italy, 1906; translated into English

by A. S. Schwier, as Manual of Political Economy,

Macmillan, New York, 1971.

[21] Lewis, K., Chen, W., and Schmidt, L., (eds.), 2006,

Decision Making in Engineering Design, American Society

of Mechanical Engineers, New York.


Figure A.1. Functional diagram of Ms. Klumpe

Figure A.2 Functional diagram of Mr. Turner

The Incredible Egg: Investigating the Design Education ... · generation of a requirements list,...

Documents

Transcript of The Incredible Egg: Investigating the Design Education ... · generation of a requirements list,...