I :2

iii.

 

I...
:31»

5...”... ,;

 

 

 

LIBRARY
Michagan State
Unlversliy

 

 

 

PLACE ll RETURN BOX to remove this Mommy": you: ncord.
TO AVOID FINES Mum on at More data duo.

DATE DUE DATE DUE DATE DUE

 

 

 

 

1‘ q _ q r’!
n. . a, - -
9" *H 1 .1
.fi!_ E .7
v 0
0‘4 .
If

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU In An mm. MONEquul Oppoctunlty Inflation
Wm1

 

CORRELATION BETWEEN TRANSPORTATION TIME
AND VIBRATION TABLE TIME

By

Deanna Joyce Fell

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

School of Packaging

1995

ABSTRACT

CORRELATION BETWEEN TRANSPORTATION TIME
AND VIBRATION TABLE TIME

By

Deanna Joyce Fell

Vibration due to shipment is a major cause of product damage. Vibration tables
are used in laboratories to recreate transportation environments. From a packaging
perspective, this device can aid in determining the packaging required for proper
protection in some known distribution channel. An assumption of many investigators is
that the vibration table is a time-saving device. If this assumption is correct, what then
is the correlation between transportation time and vibration table time? Determining
such a correlation was the focus of this research. In this experiment, cereal was used
because damage, defined here to be settling of the contents, could easily be gauged
over time. An Environmental Data Recorder (EDR) was used to record the vibration
generated by an automotive mode of transportation. A Power Spectral Density (PSD)
plot generated by the EDR was used to recreate the ride on the vibration table. It was
found that for the product and road used, about one-third of the actual transportation

time was required to cause the same amount of settling on the vibration table.

To God, I give all the glory for giving me the ability to do this thesis.

To my husband, Heath, I dedicate this work. Thank you for all of the support and
patience these last two years.

To my best friend, and twin, Delana I don't think I could have survived without you.
You're next!

Also, to my parents Jim and Brenda. Without you I could not have made it. Thank
you, thank you.

iii

ACKNOWLEDGMENTS

To a truly remarkable teacher and my major professor, Dr. Gary Burgess, I want to
extend my sincere thanks for giving me the opportunity to do this research and for all
of his time and encouragement.

Also, I want to give my thanks and appreciation to Dr. Paul Singh for serving on my
committee and for all of his time and support.
I would also like to thank Dr. George Mase for serving on my committee.

I would like to thank the Consortium of Distribution Packaging for funding this
project.

iv

TABLE OF CONTENTS

LIST OF TABLES
LIST OF FIGURES
CHAPTER 1 - INTRODUCTION AND LITERATURE REVIEW

CHAPTER 2 - EXPERIMENTAL DESIGN / MATERIALS
2.1 Product/Package System Under Investigation
2.2 Equipment

CHAPTER 3 - EXPERIMENTAL PROCEDURES
3.1 Experimental Design
3.2 Programming the Environmental Data Recorder
3.3 The Actual Transportation Environment
3.4 The Simulated Transportation Environment
3.5 Damage Criteria

CHAPTER 4 - DATA AND RESULTS
4.1 Data From the Actual Transportation Environment
4.2 Data From the Simulated Transportation Environment
4.3 Correlating Road Time To Table Time

CHAPTER 5 - SUMMARY AND CONCLUSIONS

APPENDIX - EDR DATA

LIST OF REFERENCES

vi

viii

OO\I\I

10
10
12
17
21
23

25
25

32
42

49

52

LIST OF TABLES
Table 1. Pattern of Removing Cereal from Shipping Container in the
Actual Transportation Environment TRIPS 1 and 3

Table 2. Pattern of Removing Cereal from Shipping Container in the
Actual Transportation Environment TRIPS 2 and 4

Table 3. Recording Control Parameters TRIP 1

Table 4. Memory Requirements TRIP 1

Table 5. Recording Control Parameters TRIP 2
Table 6. Memory Requirements TRIP 2

Table 7. PSD Data Used to Program the Vibration Table
Table 8. Settling Data For TRIP 1

Table 9. Settling Data For TRIP2

Table 10. Settling Data For TRIP 3

Table 11. Settling Data For TRIP 4

Table 12. Average Settling Per Trip For Bran Cereal
Table 13. Average Settling Per Trip For Flake Cereal

Table 14. Average Settling Per Trip For Cartons In Top
Position

Table 15. Average Settling Per Trip For Cartons In Bottom
Position

Table 16. Settling Data From Simulated Environment

vi

11

11

15

15

16

16

22

26

26

27

27

29

3O

3O

3O

34

LIST OF TABLES (cont'd)

Table 17. Average Settling Per Trial For Bran Cereal
Table 18. Average Settling Per Trial For Flake Cereal

Table 19. Average Settling Per Trial For Cartons In Top
Position

Table 20. Average Settling Per Trial For Cartons In Bottom
Position

vii

39

39

40

40

LIST OF FIGURES

Figure 1. PSD Plot In Bar Graph Form

Figure 2. PSD Plot of TRIP 1

Figure 3. PSD Plot of TRIP 2

Figure 4. Measuring of Settling of the Cereal

Figure 5. Actual Transportation Environment Settling Versus Time

Figure 6. Settling Versus Time For Bran Cereal, Flake Cereal,
Cartons In the Top and Bottom Position

Figure 7. Settling Versus Time For the Simulated Environment

Figure 8. Settling Versus Time For Bran Cereal, Flake Cereal,
Cartons In the Top and Bottom Position

Figure 9. Correlation of Actual and Simulated Data

viii

19

20

24

28

31

38

41

43

CHAPTER 1
INTRODUCTION AND LITERATURE REVIEW

In the past few years, measuring the dynamics of the distribution environment
has become a common practice for almost every major manufacturer in the United
States [3]. Packaging engineers are continually studying the levels of shock and
vibration in their distribution cycles. Although shock levels play an important role in
determining the proper packaging for some products, this research was geared only at
studying the vibration levels. It is the transportation environment that is characterized
by vibration and the handling environment by shock. More specifically, random
vibration was studied here. The intent of such field studies is to use the data from the
actual transportation environment to develop laboratory testing environments which
simulate the distribution environment.

Vibration tables can be used in the laboratory to simulate the transportation
environment which a package/product system may encounter. Vibration tables, or
shakers as they are sometimes called, can simulate different modes of transportation
such as automotive or rail. For this study, only the automotive mode of transportation

was used.

2

Although expensive, a vibration table cOuld be used to simulate a particular
roadtrip in a much shorter time than conducting the vibration analysis in an actual
transportation environment such as a truck. It is easy to see how the vibration table
could be beneficial in terms of being a time-saving device. It is of no surprise to any
investigator that the two most limiting factors of an experiment are time and money.
For many engineering professionals, the cost of a vibration table could be justified if it
were shown to be a method of saving time.

The purpose of using the vibration table, from a packaging perspective is to
determine how much packaging is necessary to protect the product for some specific
trip. The determination of the correct amount, or kind, of packaging will vary
depending on the standards of the packager. In most cases, the packaging professional
will have their company's standards of quality by which they need to adhere. In other
words, a product's damage criteria will be set by the manufacturer of the product. For
this study, a product was chosen so that damage to the product could easily be assessed
over time. Cereal was selected because it is damaged in distribution by breaking up
and settling. Vibration can be attributed as the primary reason for such settling. The
displacement of the cereal over time was decided to be the criteria by which the
damage would be gauged for this experiment. The cereal was purchased from a local
grocer. Because the product was purchased, it was assumed that it had already been
through a distribution channel and had already settled. Therefore, it may be reasonable
to assume that further settling may be minimal. It is also important to note that the

damage criteria was determined by the investigator, and in no way reflects the damage

criteria set by the manufacturer.

Furthermore, it can be noted that there are several sources of dynamic
excitation which must be taken into consideration. The first would be caused by the
mode of transportation, or vehicle. The second source of excitation of interest here
was the dynamic input caused by the pavement roughness. Therefore, the findings of
this research are only applicable for the test conditions given. The method used,
however, can be applied to different types of transportation and roads. There is one
other significant factor to consider and that is the product. There are numerous types
product and each has a distinct mode of failure or criteria by which it is rendered
damaged.

One general problem with simulation which has been noted for many years and
still remains a problem is that simulating the vibration which occurs in the actual
distribution environment cannot be done exactly [2]. Basically, it is virtually
impossible to re-create the vibrations in the exact same manner that they were produced
in the actual transportation cycle. Developing a laboratory testing environment which
identically mimics the actual transportation environment is extremely complex.

In this investigation, an Environmental Data Recorder (EDR) was used to
record the vibration in the truck as a result of traveling primarily over expressways
[10]. The road chosen for travel was one which was under construction. Rough road
surfaces generate higher acceleration levels which was the reason such a road surface

was used [1]. Also, the rear of the truck has been shown to have the highest

acceleration levels [1]. For this reason the EDR and the package/product system were

4
acceleration levels [1]. For this reason the EDR and the package/product system were
positioned in the rear of the truck. The EDR was basically a sensor/recording device
for shock and vibration. Recalling that the transportation environment is characterized
by vibration and the handling environment by shock, this device was used as a means
of recording the vibration. The EDR was a battery-powered device which has a built in
tri-axial accelerometer [8]. The Dyna Max software was used to take the data
recorded by the EDR and generate a Power Spectral Density (PSD) plot [10]. The use
of the PSD plot, in this case, is to send information to the random vibration controller
to vibrate the vibration table. This controller is an electronic device which produces
many pure sine waves in combination. The PSD signals the controller how the sine
waves can be mixed together in the proper frequencies and amplitudes to recreate the
random vibration from the actual transportation environment. The PSD is this re-
creation that was used to drive the vibration table. It is actually a statistical
representation of the random vibration occurring in the truck. The PSD is a plot of
Power Density (PD) versus frequency.

In most instances the PSD plot is given in a line graph form which is indicated
in Figure l by the dotted line. The line graph was the form which was plotted by the
DynaMax software. Figure 1 also indicates the PSD plot in bar graph form by the
solid line. It is important to understand that the graph actually represents a statistical
picture of the random vibration. At any frequency, the PD is the square of the
standard deviation of a random sampling of acceleration values taken from a 1 Hz

bandwidth tuner set to that frequency [1].

5

Because different analyzers can have different bandwidths, RMS G2 must be
divided by the bandwidth so it does not have any influence on the PD results. All of
this was done internally by the EDR software. For example, in Figure 1 the bandwidth
is 1 Hz so every RMS G2 would be divided by 1 Hz to get PD. Despite the
SOphistication of this method, there are still hurdles to overcome when using PSD plots

to control a vibration table. One of the problems serves as the focus of this research.

Figure 1. PSD Plot in Bar Graph Form

I Power Density

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(CZ/Hz)
1 Hz bandwidth tuner
?\
I \
/ ‘ .:
A’ \ I \\
" \
, /
/ I &\ 4’ \\
V I
, , ~ ‘_ , \
, , .. , “Th

‘ , a? _..\.._I frequency

The objective of this research was to determine if there could be some
correlation made between the time it takes to conduct the experiment on the road and
the time it takes to conduct it on the vibration table. One hypothesis is that there is a

one to one ratio which would indicate that the vibration table does not necessarily save

6

time. Another hypothesis is that this correlation is dependent on the product, the mode
of transportation, and the route of transportation. Therefore, a product was chosen
where the damage could be determined easily. The mode of travel was a ten foot rental
truck. The route of travel was kept constant throughout the experiment and is
discussed later.

The American Standard for Testing Methods (ASTM) designation D—4728-87
was used as a guide in using the vibration equipment. The problem with using this as a
guideline is stated in section 10.2 of the procedure. It states to ”conduct the random
vibration test for the length of time stated in the applicable specification, if any, of for
a predetermined period, or until a predetermined amount of damage may be detected.
The test may be stopped momentarily to inspect for damage”[5]. This suggests that
there is no specific correlation between table time and the road time. In ASTM D-4169
it is recommended to use a 180 minute duration for performing the random vibration
test to simulate transportation [4]. Since many engineers use these standards for their
testing purposes, one must question how they determine the length of time. The

uncertainty explained here served as a motivation for conducting this research.

CHAPTERZ

EXPERIIWENTAL DESIGN / MATERIALS

2.1 Product/Package System Under Investigation

For this study two types of cereal were used as the test specimens. Twenty-
four, twenty ounce cartons of flake cereal were used. Also, twenty-four, twenty ounce
cartons of a bran cereal were used. In all, there were forty-eight cartons. All of the
cereal was purchased from Meijers grocery in Lansing, Michigan.

A corrugated container which would normally be used for shipping purposes
was also obtained from the Meijers store. The dimensions of the box were 24" x 15
3/4" x 15 1/4”. The shipping container held twenty four of the cartons of cereal at one
time, twelve cartons on the bottom layer, and twelve on the top layer.

Prior to conducting the experiment, all of the cartons of cereal were opened.
The inner liner containing the product was also opened. The level to which the cereal
had previously settled was traced on the inside of the inner liner using a black marker.
This was done for all forty eight cartons.

In the first phase of the experiment, which will be referred to as the actual
transportation environment, the 24 cartons of cereal were labeled using a two letter

code indicative of the sequence in which the cartons were removed from the corrugated

8
container. This code included the layer, top or bottom, in which the carton was
positioned. The code also had a number which indicated which trip the carton was
removed. There were four trips total.
The second phase of the experiment was simulation of the trip on the vibration
table. In this phase, four cartons of cereal were removed after each predetermined
interval of time. After the settling was measured, the cartons were placed back into the

container. The cartons of cereal were labelled in the same manner as in the first phase.

2.2 Equipment

In order to create an actual transportation environment, a ten foot truck was
rented through Michigan State University and Kildee Kar Kare, 2649 Grand River,
Okemos, MI 48864.

The device used to record the vibration data from the transportation
environment was an Environmental Data Recorder. Specifically, the EDR was an EDR-
3-10, S/N 0401, +/- 10G FS Range, with an RS-232 Interface cable. This equipment
was leased from Instrumented Sensor Technology, Inc., 4704 Moore Street, Okemos,
MI 48864. Other equipment leased from IST included magnetic brackets used to
mount the EDR to the steel floor of the truck and the Dyna Max Module DM-3
software used to interpret the recorded vibration and generate a PSD plot. Wood
beams were used to brace the shipping container in the moving truck. Metal brackets
were used for support on the vibration table.

Lansmont Corporation's Vibration Table and System Station were used to

9
simulate the transportation environment encountered by the EDR in the truck. The
System Station was a computer which was simply programmed by touching the monitor

[11]. The System Station was used to drive the vibration table.

CHAPTERS

EXPERIMENTAL PROCEDURES

3.1 Experimental Design

It was the opinion of this investigator prior to the test that neither the type of
cereal nor the orientation of the package in the shipping container would have any
significant effect on the amount of settling. Nevertheless, when designing this
experiment, it was very important to make certain that the test would not be biased in
any way. In order to test this hypothesis, two types of cereal were used. There were
also two positions in the shipping container that the cartons could set. When taking the
measurements, care was taken to do so in a manner that neither of these factors could
be said to have any influence on the conclusion of the experiment. Statistically
speaking, the experiment was designed so that the factors were not confounded. An
example of confounding a factor would be if the six measurements of settling from the
first run were all taken from cartons on the top layer and the second set of
measurements were taken from the cartons positioned on the bottom of the shipping
container. At the conclusion of the experiment one must then ask if the orientation of
the package had any influence on the settling. As this example illustrates, this is the

reason for careful planning and designing of the experiment. For the actual

10

11
transportation phase of this study, six cartons of cereal were removed after each trip.

The patterns in which the cereal was removed is illustrated in Table 1 and Table 2.

Table 1. Pattern of Removing Cereal from Shipping Container
in the Actual Transportation TRIPS 1 and 3.

 

 

 

 

 

 

 

Package Cereal Type Package Orientation
1 bran top
2 bran top
3 bran bottom
4 flake A top
5 flake top
6 , fle _ _f bottom

 

   

 

 

 

 

Table 2. Pattern of Removing Cereal from Shipping Container
in the Actual Transportation TRIPS 2 and 4.

 

 

 

 

 

 

 

 

 

 

 

 

Package Cereal Type Package Orientation
1 bran top
2 bran bottom
3 bran bottom
4 flake top
5 flake bottom
6 flake bottom jl

 

12

The data collection for the simulated transportation environment was conducted
in much the same way as previously explained for the actual transportation data. The
difference was that only four cartons of cereal were removed and inspected for damage
rather than six. Of these four cartons of cereal, two were flake cereal and two were
bran cereal. Also, two of the four cartons were from the top and two from the bottom
of the shipping container.

In addition, all of the measurements for the actual transportation were done on
the same day and the weather conditions were fair all day. The measurements for the
simulated transportation were also taken on the same day. Therefore, time, weather or

driving conditions were not considered factors in this experiment.

3.2 Programming the Environmental Data Recorder

An EDR was used to record the vibration in the rear of the truck for the actual
transportation environment. Prior to mounting the EDR in the rear of the truck, it had
to be programmed. The Dyna Max software, which was also leased by IST, Inc., was
used to do the programming. Several parameters needed to be adjusted before use.
Such programming is referred to as setting the Recording Control Parameters (RCPs) in
the manual provided by the manufacturer. The RCPs had to be carefully determined to
assure that a representation of the entire trip was recorded. It is the user's
responsibility to be sure that the memory capabilities of the recording device was not
exceeded. This particular EDR had a memory of 3550 KBytes.

The route of transportation chosen was sixty-six miles in length. The maximum

13
length of time to complete one trip was estimated to be eighty-five minutes. Taking
into consideration the time it would take to complete the trip, an engineer of IST was
consulted to aid in determining how the RCPs were programmed [7]. Not all of the
parameters had to be changed. The most important were those which determined the
length of recording time (the time to complete the trip). Essentially, a simple
calculation was used to determine three of the parameters. These parameters included
the event length, dead time, and the number of events. Equation 1 was used to

determine these parameters.

Recording Time = Events(Dead Time + Event Length) (1)
Because the trip would take an estimated 85 minutes, the parameters were selected by
trial and error in order to maximize the length of recording time. The Event Length
was the time (seconds) that the EDR would record an event and the Dead Time was the
time (seconds) that the EDR would shut off between recordings. The Events was the
number of events that the EDR was capable of recording. The maximum number of
events was given as 850 by an engineer of IST,Inc [7]. As indicated in the EDR
Manual provided by IST, the EDR had to be set in the overwrite recording mode in
order to construct a PSD [9]. This means that the device will automatically begin
recording over the start of its memory when the length of recording time has been
exceeded. Consequently, it was imperative to use a length of time believed to be the
maximum to complete one trip while at the same time taking into consideration that
excess time would be a waste of recording space. The trigger level is the lowest G-

level that will be recorded. The RCPs were programmed as indicated in Table 3. The

14
memory required for recording the information is shown in Table 4. As indicated in
Table 5, two of the RCPs were changed from their initial setting as shown in Table 3.
The Dead Time and Event Length were switched. The reason for doing this was to
determine if these parameters had any significant influence on the PSD plots generated
in each of the trips. The parameters which were changed are highlighted in each table.

Table 6 shows the memory requirements of the EDR for the RCPs shown in Table 5.

Table 3. Recording Control Parameters TRIPl

I PARAMETER

15

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

PROGRAMMING
Accelerometer Sample Rate 1024 samples/second
Eiger Level 0.1 G
_'_I‘£ig_ger Duration Thresh 0.0391 seconds
Pre-Trigger Length 512 (0.5) seconds
Post-Trigger Length 512 (0.5) seconds
Accelerometer Input Selection internal
Recording Mode overwrite
Number of Events 850 events
Dead Time After Event 4 seconds
Event Length 2 seconds
Table 4. Memory Requirements TRIP 1
Acceleration Data 3410.8 KBytes
Temperature Data 0. 3 KBytes
Total Required Memory 3;:— 331g fil kBytes

 

 

 

 

 

16

Table 5. Recording Control Parameters TRIP2

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Accelerometer Sample Rate 1024 samples/second
‘_Ir_ig_ger Level 0.1 G
l_'l:1g_ger Duration Thresh 0.0391 seconds
, Pre-Trigger Length 512 (0.5) seconds
. Post-Trigger Length 512 (0.5) seconds

Accelerometer Input Selection internal

Recording Mode overwrite

Number of Events 850 events

Dead Time After Event 2 seconds

Event Length 4 seconds

 

Table 6. Memory Requirements TRIP 2

 

 

 

Acceleration Data 3441.2 KBytes

 

Temperature Data 0.3 KBytes

 

 

 

 

 

17
3.3 The Actual Transportation Environment

The intention of this portion of the study was to re-create the actual
transportation environment as near to the real one as possible. Cereal is generally
distributed via a semi-trailer mode of transportation and so a ten foot truck was used to
re-create the actual transportation environment. Primarily because of financial
restraints, a larger truck was not used. The corrugated container with the twenty-four,
twenty ounce cartons of cereal was placed in the rear of the rental truck. The
corrugated container was secured in place using two wooden beams. The two beams
were placed horizontally across the width of the truck on either side of the corrugated
container.

The EDR was mounted in the truck with the aid of magnets. The EDR was
mounted at the rear of the truck as near to the shipping container as possible. The
reason for placing the EDR near the container was to record the same ride that the
product experienced. The path which was driven for each of the four trips was

identical and is explained in detail below:

-Started at the Packaging building's loading dock.
-Turned left, or west, onto Wilson Road.
-Turned left, or south, onto Harrison Road.

-Turned right, or west, onto Trowbridge Road.

18
-Entered the US 127 South expressway.

-Took Interstate 96 West for approximately 20 miles to the Portland Exit. At
the exit turned right, then left into the McDonald's parking lot.

-From the parking lot, turned right onto to the road and then entered the I-96
East expressway by turning left.

-Took I-96 East toU.S.127 North.

-Exited US 127 North at the Trowbridge Road exit.

-Turned left, or north, onto Harrison Road.

-Turned right, or east, onto Wilson Road.

Returned to the Packaging building.

This entire trip was sixty-six miles in length. This path was chosen because the
expressway, for the most part, was under construction. All of the driving occurred on 1
Friday, October 7, 1994.

After each trip, six cartons of cereal were removed from the corrugated
container to be inspected for damage and then returned to the container before the next
trip. The damage criteria will be discussed in section five of this chapter.

As previously discussed, the PSD plots for the first two trips were generated by
the EDR. The PSD plots for TRIP 1 and TRIP 2 are shown in Figure 2 and Figure 3,
respectively. Immediately following the first run, the EDR was switched to standby
mode. In this standby mode, all of the data from the trip was preserved and could not
be over written. The EDR was connected directly to a computer and the data

information uploaded as directed in the Dyna Max User Operating Manual.

219

.aficoasommdgam

Anzv 30:03605u

5.
H3

 

 

 

m m u u u m m m u u m m m m m m in ..
O O O O O O O O I O O O O 0 O I 0
O O O O O O I O O O I O O I O 0 0
O I O O O O O O O O O O O O O O O
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 o
O O O O O O O O O I O O O O O O O
a0’00‘0000'0000'00000W000000“00000000’000000000000.000000000000000000000’900000.0000000000000000000000'coco-0000000.00.0000000000000-coco-0.000.000fi
O O O O O O I O O O O O O 0
O O O O O O O O O O O O O O O 0 0
O O O O I O I O O O O O O O 0
00¢00M0000w0000W00000W000000-00000000’00000000000000.000000000000000000000’H000m000W000M0000~00000.0000000m000000000m000000000000mean-00000000000000.000 v
00900h0000‘0000“000000\000000”00000000‘000000000000’000000000000000000000900.000”000“0000“00000“00000000‘00000000u000000000000000000000000000.00.00.00 v
0 O O O O O O O O 0 O I O O O O r
00-00.0000.000000000."000000.00000000'000000000000.000000000000000000000-00000000000000000000000000000'000000000000000000000.0000000000000000000000
I O O O . O I O O O O I O I O O O O
00-000000000000.0000‘000000000000000-0000000000000000000000000000000000D0000000.000.00000000000000000UI0000000000000000000000000000000000000000000v
00¢00h0000VO000V00000”000000~00000000#000000000000v000000000000000000000o00u000M000“0000“00000“00000000‘0.000000”000000000000000000000000000000000000v
00’00’00000000'00000.000000000000000’0000000000000000000000000000000000‘0000000.000.0000000000000000000000000000000000000000\0000000000000000000000 v
nunaaunuanaaana:auEnwnnunuawnnnnnnunnuwuruunannéunannnwnnnuwfinunnnwuunnnnunnnnu
" u u u u u u u u u u u u u u u n
O O O O O O O O O O I O I I O O O
I O I O O O O O I C O O O O O O 0
O O I O u 0 O u C O 0 O O O 0 0 O
I O O O O I O O O O O O O O 0
00-00:00000000000000.0000»0000)|0000"000000000000m000000000000000000000’000000.000.0000fl00000000000000D00000000u0000000-0000000000..-00000000000000v
O O O O O O O I 0 O O 0
O O O C O O n O O O O m H M H 0
O O O O O O O O O. O O O 0 O
00‘0000M0000W0000W000e0W000000 000 M0 000000000W000000000000000000000h0000m000”000m0000W00000m0000000m¢00000000m000000-00n0eM0n0000000000-000000000v
f00h0000h00000W0000W00000w000000 00000000h0000 000 0%000000000000000000000’0000-000.000“0000“00000*0000000Vooea0000me000000000-000000000000000000000000 v
00'00h0000000000000000" 00 000000eoh0000 00 00 0N000000000000000000000900".00"000”0000“.0000“00000000.000.00000000000o00.00000000000000000000000.000 v
00.00000000000D00000‘000000‘ 0 ”00000000'000000 0 000 "000000000000000000000h000uo00".00“0000”00000"00000000‘00000000u00000000000000000...00.0000000000000 v
0’0000‘000090000900000V 00 *00000000’000000 0000 0000000000000000000000u000.00a.000‘0000h00000&0000000900000000-000000000000h0000000000000000000000v
00-0000000000000000000 00 0000000000‘00000 0000 000000000000000000000‘00000000000000000000000000000.000000000000000000.0000.00000000000000000000v
0’00.000000000'00000 0000000000000.’00000000000 000000000000000000000‘0000000000000000000000..00000000000000.0000.000.000000000000000000000000000r
0¢00§000000O0000w000 00000u000000004000000000000 000000000000000000000‘0000M000U000N0000R00000”.00000o¢’e0000000“00¢00000.000‘000000000000000a0000000v
u u u u u u n n u n n u u u u u
0 0 0 0 0 0 0 0 0 0 0 e 0 e n u
.p'.m....m.. nI...‘."....."U.......pl....C...’..“....... ..O ...‘pCD.C.C”...m....m.....m....l.0”‘I...OI.."OI.......C..“OO.CO..I...OCOOODOCCIIf
" u u u u n u u . u u . u u u "
0q00fl0000" 0"00000"000000"00000000q000000000000"000000000000000000000400"000"000“00000 0.000000In,00000000M000000000000§0000000-000000000000000v
‘00008H0000m 000m00000m000000mooe0aooe‘n0000000000000m000000000econ-00000.0‘”-olm000W000M0000M00000N000 eco’coooooeomeeneecocoa-oweooeeeo-0.00000000000-w
00 00‘ 0000‘00000”000000u00000000’00000.000a000ofloe00000000000000.0000hoouu000"...”0oeohaoa0oha00000 0.0a0ooueaooocaoecoon-0000000000000000000 v
0 rec.“ 0000‘00000”000000M00000000’00000000000000u00000000000000000000060no”.ecu-00fl0000h00ooohoeeoo.00‘00 essence0000..a0000000000000000000000 0007
0 000' 000W000000'00000ou00000000‘00000000000000M000000000000000000000‘0000“000“000&0000000000N00000000,-000 00n000000000000A00000000000000 00000.7
0 00‘ 000.00000Q00000000000-000{000.0000-0000000000000000000000000‘000000‘00000000000000.0000000‘00000 0000000000.econoooeooouuooo enuuuuuunn
3M. .3. uuumuuuuumuuuuunmnn"unnunflunununnunuuum”u””3“”nuuuunuuuuuuuMnumu”umnunMuuuuMuunnumflnuuumunnununfl..uuu”ununuuuunuuunu. 11..
..9..h.. ...CvUOCODVD..C.I”CI...C..#.ICI...IDCIDVIICOOCI.IDIOOUOODOCIIh‘IDWII.”...“. ...OmIIDODWIOOICIDC‘. IIIIIII WDIIUOOICIUCI’HIDOCICDI.OIIOOOCIOIIOII
CDPCOh......IDHDCCCO"IICOOOmDCOICOOD+ICOCUOUIICCDHIIOCDOIOI... IIIII IIIIPUO”.IC"COCHOCIDMCCCDDWOCCOOOIH.IIIOIOOD"OIOO IIIIII OI...COCO-...CDOOIOOOOIIICV
eemoo M00000.00000.0000000000000000000000’000000000000'000000000000000000000‘0000"u00”000“0000.00000O0000000m cccccccc m 000000000000 Q” 000000000000000000000 Iv
..hICI'....v....v.....vIO...O“.C.OI...."C.I...0......¥CDIO...DCOOIUIOICIOIOC"CIO“CIIvlII“.|IC~...C.*I..OCCD?IOI IIIII u IIIIIIIIIIII hCIQOOICODOICIOCCCUOICCV
..fi...&....v....v...l.’0......00...OCOSCIDOIUIIOCCDleIIIOOICDDOOCOOIODOOICHDII—lIC'IOIADDCDa.0...“..0000OHIO-OOOODI~ODO ......... s. IIIII .IIDIOIIODOOI....fi
00‘000.0000'0000.000000000000000000000‘0000000000000'0.0000060000000000000‘0000000000000-000-0000.0000000’00000-0000 0000.00.00...-...0000-0000000000-
§~H~ 00V.00oVaoooovooooooaoaoeoaoohoooooooonoeeouoonee.Innocence-ueeeoouooo-one“ooohoooofiaeeoofieooo-oon.use...oofiaeaoouooocoohooeoo000000ooooaoouooon
~DC‘C...‘DICII‘QICOCC....00..."l...................‘OCOOCDIIOUQOC€C...‘..II....U...‘I...I..I...‘OCUOOOC...DOCOICIIODC. ..... 0......IC.DI.O...

 

 

 

uzza

moie—

230.0

and

ad

(Zn/.6) 2 use

2C)

5.

Anzv DUB-.30. Ln.—

9.

«ax—.333...” es»...—

 

 

,mzsa

    

 

 

  

 

P P 1” P P D P D P P F D P P F P F
. . . . . u u u u d u u u u u u n ..
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 I I I . I I I . 0 I 0 I I I 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 I I I I . O I I I I I I 0 0 I 0
I 0 I I I I I 0 I I 0 I 0 I I I I
00,000000000000.000000000000000000000,0000000000000000000000000000000000,00000000000000000000000000000’000000000000000000000‘0000000000000000000000 v
I I I 0 I I I . I 0 I 0 I I . 0 I 0
I I .. . I . I O . . I I . I I I 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
00h000fi0000¥0000¥00000v000000.00000000’00000000000000.000000000000000000000¢00.000.000‘0000fl00000“0000000”000000000.000000000000h0000000000000000000000f
0 I I I 0 0 I I I I 0 . I I I I I
I 0 I I 0 I I I I I I I 0
00‘0000000.0000‘00000\000000“00000000‘000000000000”000000000000000000000‘000000.00000000000000h0000000‘00000000M00000000000000000000000000000000000 v
I I I I 0 I . . I I . I 0 I 0
00’0000000.0000n00000.000000000000000-000000000000H000000000000000000000E000000000000000H0000000000000.00000000000000000000000000000000000000000000 v
I 0 I I . . u I I I I 0 I . 0
00,0000000000000000000000000000000000-000000000000H000000000000000000000-000000000000000000000000000000'00000000000000000000000000000000000000000000 0
.09II“....”...Iv.....”00....”I.I.....9.........I.I”.........000......00.09..”0II”...“IIII”...II“IIIII.*000000.0“IIIIIIII0000,00IIIIIIIIIIIIIIIIIIIIIT
00.00000000000000000000000000000000000’0000000000000000000000000000000000‘000000000000000000000000000000.00000000000000000000000000000000000000000000 v
I. .0 I... ..I. ...0. 00.0.. I....... .........I.0 I....I.000.I.....0.IIA... .00 .I. I... 0.... II.IIIII’.000.III .00000000000 000.....OIIIIIIIIIIIII I
00 000 0000 0000 00000 000000 00000000 000000000000 000000000000000000000’000 000 000 0000 00000 00000000000000000 000000000000 00000000000.0000000000 0
0 I I I 0 . I I . I I I I I I I I I
I I I I I . . I I I I 0 I I I I I
I 0 I I I . O 0 . I I I I I I I 0
I I I I I . . I I 0 I I I 0 I I I
I I I I I I I 0 I I 0 I I I I I 0
. I I I I . O 0 I I I I I I I I I
00'00.0000.0000000000.000000000000000-0000000000000000000000000000000000.000000000000000000000000000000.00000000000000000000000000000000000.00000000 V
O 0 I . I . . . I I I O I I I I 0
I I I I I I I I I I I . I I I I 0
I 0 0 I 0 I O I I I 0 I I 0 0 0 0
00fi000A0000V0000V00000V000000-00000000fi0000000000000.00000000000000000000050000.000”000*0000h00000Q0000000n000000000-0000000000005.000000000000000000000 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0h000h000090000¥000009 0000—00000000fi 000 0000.000000000000000000000’0000-000¥000fi0000~00000fl0000000V00000000.000000000000’00000000000000000000000 v
0.00000000000000000l0000000 000" 00 0 00 00 00H000000000000000000000P00"000H000H0000M00000"00000000.00000000H000000000000000000000000000000000000 v
0.0000000000V0000V000000 0 00 0 00 00 0 000, 0"000000000000000000000.00000000fl000h0000“00000h00000000‘00000000u000000000000000000000000000000000000 v
0’0000fi0000’0000V00000 0 .00000 0 00000 00000 .000000000000000000000’0000-000U000fi0000‘00000*0000000u000000000.000000000000h0000000000000000000000 0
'0000000000000000000 00 0000000 ‘000 000000 00000000000000000000000000000000000000000000000000000.00000000000000000000000000000000000000000000 0
’00.000000000000000 00 0000000 ,00000000000 000000000000000000000’00000000000000000000000000000'00000000000000000000000000000000000000000000 0
\0000‘000000'00000'0000 00 0w00000000‘0000000000000 000000000000000000000‘0000M000W000N0000”00000“00000000'00000000u000000000000“0000000000000000000000 v
I I I I 0 0 I I I 0 0 I 0 I I 0
I I I I I . I I I I I I I 0 I 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 I I I 0 0 I 0 I 0 I 0 I I Q 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 o
. ‘.......... .. ‘...............¢.................. ...............‘..........‘...........I......‘0..I.I...IIII000....0‘0000..00I..00....I....r
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
I I 0 I I I 0 I 0 I I 0 I 0 0
I 0 I 0 I . I I I I I 0 I 0
0 ’0000000000 0 000000000000000000,0000000000000000000000 0 . 00000 0000000000000'000000000000000000000\0000000000000000000000 r
0 0 I I I I 0 I 0 I I I I 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 000.0000'00 0 0 0000000000000000000‘000000000000.0'000000000000000000000‘00000000000.00.000000 000000}0.0000000000000000000‘0000000000000000000000V
0 0 0 I I I 0 I 0 0 O 0 I 0 0
. I ..h....v.0 0. ...O‘......“I.0.....9...O........“................I.IIIh...“...“...“...........“.I ...O‘...I.00.“.00.......II“0........I.I.I.I.I.I. I
I. ...h.....‘. I. .....‘.0....“D.......9...I........u.......I..l.....I...I’....u...”...“....“l....“....0 I .I...I..“I.....0....I’II..........I...I.. I'
00 00.00000. 00000000000000.00000000‘000000000000000000000000009000000000‘0000u000u000Q0000“00000N0000000- 000-00u000000000000‘00000000000000000 0000 0
00 0.0 0‘0000000 0N000000000000000000000‘0000000 00000000000000000‘00 000000 I
00 0000000. 0 00000I00000000-0000000000000000000000000000000000.00000000 0000000000000000000000 000000000
00 0‘0000' 0'00000'000000000000000’0000000000000000000000000000000000‘0000000.000.000000000000000000.00000 000000000000000000000000 0000000000 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 o 0 0
00’0 .0000 000‘00000‘000000000000000’000II00000000.000000000000000000000’000.000.000.0000.00000'00000000000000000.00000000000 000000000000000000000 T
. I . . I I . 0 . I I . . I 0 I I
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
00.00 00000 00000000000000000.00000000'0000000000000000000000000000000000-000000.000.000000000000000000'00000000000000000000000000000000000000000000I
0 0 0 n 0 0 0 0 0 0 0 0 0 u 0 0
0 o 0 o 0 0 0 0 0 0 0 0 0 0 0 0
00‘0- ‘0000 00000000000000000.00000000‘00000000000000000000000000000000000’000000.000.0000O0000000000000'0000000000000000000000000000000-000000000000 I
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
...".. “0... ..I.v.....v.lII..“........h.............“.l...l....0..........0”.......v.0.~....“.....*00..I..”..I...0....0.0.III.I..fl...........D........00v
000n00 ‘00000 000.0”00000w000000*000000000"0000000000000~0000000000000000000000Hl00-000v000*0000*00000‘0000000H000000000m000000000000‘00000000000000000000000V
00.00 \ 0 00000000000000000.00000000‘00000000000000000000000000000000000‘00000000000I000000000000000000'00000000000000000000000000000000000000000000v
DD. ....v.....v.0.0...I.......h....0........”.....I.....I......I.I.”...‘...m...“.0..“.....~......."0.I.0I00.”..000000...Ifil..........III..I.....v
0 g ...I‘I’...‘.I.............£...I..........IIU..I0..I..l.0..l..’.......‘......I.........I.I...‘IIII000......IIIO.II.’0.I...0..I..II..00.... V

 

 

no-0—

88.0

80.0

8.0

(ZH’.6> 2 use

2 1
3.4 The Simulated Transportation Environment

The vibration table was used in the laboratory to simulate the ride recorded by
the EDR on the ten foot truck. Power density and frequency data had to be retrieved
from the PSD plot in order to program the vibration table. This was accomplished by
using the Dyna Max software and the PSD plot from TRIP 2 to determine the Power
Density (Oz/Hz) values for the first fifty frequencies. Although there was not a
significant difference between the two PSD plots generated in the Actual Transportation
Environment, the PSD from TRIP 2 did have slightly higher PD values than in TRIP 1.
For this reason, the PSD values from TRIP 2 were used to program the vibration table
for the Simulated Environment. The frequencies used were from 1 Hz to 49 Hz in
increments of 1H2. Table 7 represents the data that was used to program the vibration
table. The data was entered into the Touch Test Vibration System Station manually.
The raw data from the EDR can be found in Appendix A.

Again, the test specimens under investigation were packaged as intended for
shipment. Twenty-four cartons of cereal, not the same as previously tested, were
packaged into the same corrugated container as used in the actual transportation
environment. Restraining devices were attached to the vibration table to prevent the
corrugated container from moving freely. This method of restraint was used to mimic

the wooden beams used in the truck to secure the container.

22

Table 7. PSD Data Used to Program the Vibration Table

FREQUENCY POWER DENSITY

(HZ) (9‘21 H1)
1 0.w0489
2 O.m1880
3 0.000948
4 0.000484
5 0.000338
8 0.000288
7 0.000255
8 0.000228
9 0.000218

10 0.000241
1 1 0.000242
12 0.000198
13 0.000174
14 0.000207
15 0.000193
18 0.000187
17 0.000171
18 0.000153
19 0.000140
20 um
21 0.000047
22 0.000035
23 0.000048
24 0.000078
25 0.0MO87
28 0.000039
27 0.000033
28 0.000029
29 0.000028
30 0.000023
31 0.000023
32 0.000032
33 0.000038
34 0.000027
35 0.000023
38 0.000022
37 0.000023
38 0.000024
39 0.000032
40 0.000039
41 0.000051
42 0.000088
43 0.000080
44 0.000072
45 0.000055
48 0.000053
47 0.000051
48 0.!!!)088
49 0.0001 10

2 3

According to ASTM D-4728-87, the random vibration test can be stopped at
predetermined times to inspect for damage [5]. For this investigation the
predetermined interval of time was ten minutes. Because the entire length of the actual
trip was 335 minutes, ten minutes was thought to be appropriate.

After each trial, which will be defined as each ten minute interval, the vibration
table was stopped and four cartons of cereal were removed and inspected for damage.
Of these cartons, two were from the top layer and two were from the bottom layer.
Also, two of the cartons were flake cereal and two were bran cereal. These cartons of
cereal were inspected for damage and placed back into the container to keep the
organization of the package intact. This test was continued until the amount of settling
was near that of the worst condition measured in the actual transportation portion of

this study.

3.5 Damage Criteria

For this experiment, a product was chosen where damage criteria could be
easily defined. Cereal settles during distribution. It is often noted by the manufacturer
of the product, on a flap of the carton, that the product is sold by weight and not by
volume to reflect this fact. Furthermore, such a statement usually indicates that some
settling of the contents may occur during shipment and handling. For purposes of this
investigation, the damage criteria was taken to be the amount of settling and was
simply gauged by measuring the displacement of the cereal over time in the carton.

Figure 4 helps to show how the settling of the cereal was measured. Prior to testing,

24

the outline of the cereal was traced using a marker. This line was not a horizontal one,
but irregular as shown in Figure 4. The height of the cereal was measured prior to
testing. This height was actually an average of the ten heights taken at the locations
marked with an X in Figure 4. After testing, the height of the cereal was determined
again. This height was also an average of ten heights. Settling was defined as the
difference between the average height prior to testing and the average height after
testing. The ruler used to measure was heights graduated in 1/16ths of an inch.

Therefore, the error would be about 1/32nd of an inch.

Figure 4. Measuring of Settling of the Cereal

 

 

 

 

 

CHAPTER4

DATA AND RESULTS

4.1 Data From the Actual Transportation Environment

Four of the same sixty-six mile trips were driven to re-create an actual
transportation environment for the cereal. After each trip, six cartons of cereal were
removed and the average amount of settling for each was determined. Ten
measurements of settling were taken for each carton of cereal to determine the
average. Again, settling was calculated by determining the average height of the cereal
prior to testing and subtracting from it the average height of the cereal after testing.
The average settling for each trip was calculated by computing the mean of the settling
for the six cartons of cereal.

Trip 1 took 82 minutes to complete. The speed of travel for Trip 1 was
determined to be an average of 48.29mph. The settling data for Trip 1 is shown in
Table 8. Table 9 shows the settling data for Trip 2. The second trip took 88 minutes
to complete for an average speed of 45mph. Table 10 contains the settling data for
Trip 3. The third trip took 81 minutes to complete for an average speed of 48.9mph.
Finally, the settling data for the fourth trip, Trip 4, can be found in Table 12. Trip 4

took 84 minutes to complete and the average speed was 47.14mph.

25

Table 8. Settling Data For Trip 1

26

 

 

 

 

 

 

 

 

 

 

 

 

Package Cereal Package Average
Type Position Settling (inch)
1 bran top 0.24
2 bran top 0.29
3 bran bottom 0.26
4 flake top 0.27
5 flake top 0.34
6 flake bottom 0.29
Average Settling For Trip 1 0.28 inches

 

Table 9. Settling Data For Trip 2

 

 

 

 

 

 

 

 

 

 

Package Cereal Package Average
Type Position Settling (inch)
1 bran top 0.5 1
2 bran bottom 0.59
3 bran bottom 0.42
4 flake top 0.58
5 flake bottom 0.44
6 flake bottom 0.44

 

 

Average Settling For Trip 2 0.49 inches

 

 

 

Table 10. Settling Data For Trip 3

27

 

 

 

 

 

 

 

 

 

 

 

 

Package Cereal Package Average
Type Position Settling
(inch)

1 bran top 0.62

2 bran top 0.78

3 bran bottom 0.69

4 flake top 0.73

5 flake top 0.7 8

6 flake bottom 0.74

Average Settling For Trip 3 0.72 inches

 

Table 11. Settling Data For Trip 4

 

 

 

 

 

 

 

 

 

 

Package Cereal Package Average
Type Position Settling
(inch)

1 bran top 0.76

2 bran bottom 0.79

3 bran bottom 0.74

4 flake top 0.83

5 flake bottom 0.73

6 flake bottom 0.91

 

 

Average Settling For Trip 4 0.79 inches

 

 

 

28

Figure 5 is a plot of the amount of settling of the cereal over time in the Actual
Transportation Environment. Each point is the average of six replications. Recall that
six cartons of cereal were removed and the amount of settling measured after each trip.
The four points on the plot represent the averages of settling from each trip.

Regression analysis was performed to find the slope of the line. The slope was found

to be 0.0024 inches/minute.

Figure 5. Actual Transportation Environment Settling Versus Time

 

0.3 p

015 L

Average Settling (inches)

04 _.

 

 

02

 

82 I70 25l 335
Time (mimtes)

29

Recall that one expectation in this experiment was that neither the cereal type nor the
orientation of the package in the corrugated container would have any significant
influence on the amount of settling. In order to show how this factor affected the
settling, the data for each trip has also been averaged for each cereal type (bran and
flake) separately and for each position within the corrugated container (top or bottom).
These averages for TRIPS 1, 2, 3 and 4 were tabulated and are shown in Tables 12,
13, 14 and 15, respectively. Furthermore, to compare the amount of settling over time
for each of these factors, this data was also plotted. Figure 6 depicts the amount of
settling versus time for bran cereal, for flake cereal, and for cartons in the top and
bottom layers of the corrugated container. Using linear regression, the slopes of each
of these graphs were determined and compared to the slope for Figure 5. All slopes
were approximately 0.0024 inches/minute suggesting that neither the type of cereal nor
the orientation of the carton had any influence on the settling of the product.

Table 12. Average Settling Per Trip For Bran Cereal

 

 

 

 

 

Trip Average Settling (in)
1 0.26
2 0.5 l
3 0.70
4 0.77

 

 

 

 

30

Table 13. Average Settling Per Trip For Flake Cereal

 

 

 

 

 

Trip Average Settling (in)
1 0.30
2 0.49
3 0.73
4 0.82

 

 

 

 

Table 14. Average Settling Per Trip For Cartons In Top Position

 

 

 

 

 

 

 

 

Trip Average Settling (in)
1 0.29
2 0.54
3 0.73
4 0.79

 

Table 15. Average Settling Per Trip For Cartons In Bottom Position

 

 

 

 

 

Trip Average Settling (in)
1 0.28
2 0.47
3 0.72
4 0.79

 

 

 

 

31

Figure 6. Settling Versus Time For Bran Cereal, Flake Cereal, Cartons In the Top
and Bottom Position

 

Average Settling (inches)

 

 

 

 

0.2 .
82 no 25' 335

Time (mimrtes)
+Settling For BranCereal +SettlingForFlalreCereel
-A- Settling For Top Position -a— Settling For Bottom Position

32

4.2 Data From the Simulated Transportation Environment

Table 16 represents the settling data forthe cereal under the simulated
conditions. As indicated in Table 16, the measurements were taken every ten minutes.
After each ten minute interval, which will be referred to as a trial, four cartons of
cereal were removed and the measurements of settling taken. Again the average
amount of settling for each carton of cereal was determined by computing the‘mean of
ten measurements. For each interval of time, or trial as it is referred to in Table 16,
the average settling was computed for the four cartons of cereal.

After the first six trials, the settling for all twenty-four cartons had been
measured once. For the remainder of the trials, sixteen cartons of cereal were re-
examined for settling. It is important to note here that some error in the settling for
trials 7, 8, 9 and 10 may be attributed to the fact that the cartons of cereal had been
measured previously. Furthermore, the data indicates that a correlation can be made
for the data from the first six trials of the test. However, the data from trials 7, 8, 9
and 10 seem to agree with the correlation of the data for the first six trials which
suggests that the re-measuring of the data did not have a significant effect on the
measurements.

Figure 7 is a plot of the displacement of cereal over time in the simulated
transportation environment. Regression analysis was also performed to determine the
slope of the line generated by the data in Table 16. The slope of the line was 0.007
inches/minute.

As in the Actual Transportation phase of this experiment, the data here was also

33

averaged according to the type of cereal (bran or flake) and the orientation which the
carton was placed in the corrugated container (top or bottom). The data was separated
in this manner to show if either of these factors had any influence on the settling of the
cereal. For each trial, or each ten minute interval, the average amount of settling for
the bran cereal is shown in Table 17. This data was plotted versus time in Figure 8.
The average settling per trial for the flake cereal was computed and is shown in Table
12 and plotted in Figure 8. The average settling per trial for the cartons in the top and
bottom positions are shown in Table 19 and Table 20, respectively. Figure 8 also
charts the average settling over time for the cartons in the top and bottom positions,
respectively. The slopes of the lines in Figure 8 where computed by linear regression
and compared to the slope of the line in Figure 7. All slopes were approximately
0.007 inches/minute which suggests that none of the factors had any significant effect

on the settling of the cereal.

34

Table 16. Settling Data From Simulated Environment

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Trial Time Package Cereal Package Average Settling
(minutes) Type Orientation (inches)

1 10 1 bran top 0.14
2 flake top 0.15

3 bran bottom 0.13

4 flake bottom 0.14

Average Settling For Trial 1 '5 0.14 in

2 20 l bran top 0.21
2 flake top 0.22

3 bran bottom 0.23

4 flake bottom 0.21

Average Settling For Trial 2 is 0.22 in

3 30 l bran top 0.33
2 flake top 0.29

3 bran bottom 0.32

4 flake bottom 0.28

 

 

Average Settling For Trial 3 is 0.30 in

 

 

Table 16. (cont'd)

35

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Trial Time Package Cereal Package Average Settling
(minutes) Type Orientation (inches)
4 40 1 bran top 0.35
2 flake top 0.38
3 bran bottom 0.39
4 flake bottom 0.33
Average Settling For Trial 4 's 0.36 in
5 50 l bran top 0.39
2 flake top 0.43
3 bran bottom 0.39
4 flake bottom 0.43
Average Settling For Trial 5 is 0.41 in
6 60 1 bran top 0.53
2 flake top 0.51
3 bran bottom 0.51
4 flake bottom 0.47

 

 

Average Settling For Trial 6 is 0.51 in

 

 

36

Table 16. (cont'd)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Trial Time Package Cereal Package Average Settling
(minutes) Type Orientation (inches)
7 70 1 bran top 0.53
2 bran top 0.56
3 flake bottom 0.57
4 flake bottom 0.57
Average Settling For Trial 7 is 0.56 in
8 80 1 bran top 0.71
2 bran top 0.63
3 flake bottom 0.59
4 flake bottom 0.60
Average Settling For Trial 8 is 0.63 in
9 90 1 bran top 0.74
2 bran top 0.77
3 flake bottom 0.80
4 flake bottom 0.73

 

 

 

 

 

 

 

Average Settling For Trial 9 is 0.76 in

 

 

Table 16. (cont'd)

37

 

 

 

 

 

 

 

Trial Time Package Cereal Package Average Settling
(minutes) Type Orientation (inches)
10 100 l bran top 0.80
2 bran top 0.77
3 flake bottom 0.76
4 flake bottom 0.84

 

 

Average Settling For Trial 10 is 0.79 in

 

 

 

Figure 7. Settling Versus Time

38

For the Simulated Environment

 

0.8 _

0.6 ._

Oe‘ #—

Average Settling (inches)

0.2 ._

 

 

40 50 60 70 80

Time (1mm)

11!)

 

39

Table 17. Average Settling Per Trial For Bran Cereal

 

Trial Average Settling (in)

 

1 0.13

 

0.22

 

0.32

 

0.37

 

0.39

 

0.52

 

0.55

 

0.65

 

\OWQONM‘FUN

0.77

 

0.78

y—e
O

 

 

 

 

Table 18. Average Settling Per Trial For Flake Cereal

 

Trial Average Settling (in)

 

0.14

5—5

 

0.21

 

0.30

 

0.35

 

0.43

 

0.49

 

0.57

 

0.61

 

\OOOQGMAUJN

0.75

 

0.81

p—e
O

 

 

 

 

40

Table 19. Average Settling Per Trial For Cartons In Top Position

 

Trial Average Settling (in)
1 0.14
0.22
0.31
0.36
0.41
0.52
0.55
0.67
0.75
0.79

 

 

 

 

 

 

 

 

 

\OOOQGUIJib-DN

 

 

 

p—e
O

 

 

Table 20. Average Settling Per Trial For Cartons In Bottom Position

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Trial Average Settling (in)
1 0.14
2 0.22
3 0.29
4 0.36
5 0.41
6 0.49
7 0.57
8 0.59
9 0.76
10 0.80

 

41

Figure 8. Settling Versus Time For Bran Cereal, Flake Cereal, Cartons In
the Top and Bottom Positions

 

0.3 +-
3 0.6 _
g
o

0.4 _
3
<

0.2 _.

 

 

 

it! If so so {It
Time (minutes)
+ Settling For Bran Cereal + Settling For Flake Cereal
-A- Settling For Top Position + Settling For Bottom Position

42

4.3 Correlating Road Time ToTable Time

As shown in the data for the Actual Transportation Environment and the
Simulated Environment, neither the type of cereal nor the orientation of the package in
the corrugated container proved to be a significant factor in the settling of the cereal.
Therefore, the data used for correlation purposes was that which was not separated for
these factors.

Figure 9 represents the correlation of the data between the Actual and Simulated
environments. For this particular study, having a criteria such as settling makes
determining the "damage” easy. The reason the line for the Actual Transportation
Environment begins so late is because the first trip took 82 minutes to complete. Also,
the data for this line is so infrequent because the points represent the data for each of
the four trips.

For illustration purposes, if the correlation between table time and
transportation time was known, one could use this information to gauge the damage.
This could be an extremely valuable tool in determining the appropriate amount of time
a package/product system should be on the vibration table for a known distribution
time. For example, if the actual distribution time is 240 minutes then the amount of
settling is approximately 0.6” and so the time necessary to re-create this amount of
settling on the vibration table is approximately 80 minutes from Figure 9. It takes
approximately 50 minutes on the vibration table and 135 minutes in the truck for the
cereal to settle 0.4 inches. This graph shows that there is not a one to one relationship

between transportation time and table time. The correlation is more like 2:1 to 3: 1.

43

Based on the slopes in Figure 5 and Figure 7 obtained from regression analysis the

correlation suggested here for cereal is 0.007/0.0024 which is about 3 to 1.

Figure 9. Correlation of Actual and Simulated Data

 

 

 

 

0.3 _ I
O
’5
a:
g 0.0 L.
V
0
an
0.4 _. I
E O
0
>
<
O
0.2 _ 3
I
0
no 20 30 40 so 00 70 00 02 90 100 no 251 335

Time (minutes)

-I— Settling in Simulated Environment 0 Settling in Actual Environment

CHAPTERS

SUMMARY AND CONCLUSIONS

The focus of this research was to determine if a correlation between vibration
table time and actual transportation time could be found. As previously described, an
EDR was used to record the vibration environment in a ten foot truck. Because the
rear of the truck has been shown to have the highest acceleration levels, the EDR and
the package/product system were positioned in he rear of the truck [1]. The Dyna Max
software was used to plot the PSD from the actual transportation environment. The
vibration table was able to simulate the transportation environment by entering the PSD
data via the Touch Test Vibration System Station control panel. The method described
here coincides which Method A, closed loop- automatic equalization, of the ASTM
standard D—4728 for Random Vibration Testing of Shipping Containers [5].

The EDR was a dimple device to use. Programming the RCPs was , perhaps,
one of the-more difficult aspects of the experiment. Fortunately, changing the RCPs
had little effect on the PSD plot. Recall that the times for the Event Length and the
Dead Time were switched in TRIP 1 and TRIP 2. In TRIP 1 the Event Length was 2
seconds and the Dead Time was 4 seconds. In TRIP 2 the Event Length was changed

to 4 seconds and the Dead Time to 2 seconds. Both of the plots were made on the

44

45
same scale for comparison purposes. When overlaying the two plots they were very
similar between frequencies 1112 and 50 Hz. The largest difference between the two
graphs was approximately 0.0004 G’le which occurred at 2 Hz. There are other
areas which appear to have larger discrepancies, but because the PSD plot is on a log-
log scale, some of the differences appear large but are actually orders of magnitude less
than that at 2 Hz.

It appeared that altering the Event Length and Dead Time parameters had no
significant effect on the PSD plots. One explanation may be that the 2 second window
of time in which the event was recorded was long enough.

The Touch Test Vibration System Station control panel was capable of receiving
a total of fifty entries for the PSD plot. In this study, the PD values for only the first
fifty frequencies, 1 Hz to 49 Hz, were determined using the Dyna Max software. For
frequencies greater than 49 Hz the PSD plots from the Actual Transportation
Environment show a graph with several peaks. These peaks may indicate that there
was some other source of vibration than that produced from the road. This other
influence of vibration could be from the flexing of the metal to which the EDR was
magnetically fixed. If the PD values for every 2 Hz had been entered into the Touch
Test Vibration System Station control panel, a PSD representative of the entire PSD
from the Actual Transportation Environment would have been used. This means that
the PD values for the frequencies 2 Hz to 100 Hz, in increments of 2 Hz, would have
been determined. The reason for using the PD values for frequencies 1 Hz to 49 Hz

was to avoid using the PD values at the higher frequencies which seemed more severe

46

than what actually happened in the truck. Again, these high PD values may be
attributed to the flexing of the metal floor to which the magnets on the EDR were
attached. Evidently, the Simulated Environment can in no way re-produce the exact
vibration which occurred in the Transportation Environment. Also, it becomes evident
that it is a very difficult task to determine what data to use from the Actual
Transportation Environment in attempting to re-create it on the vibration table.

Product damage is the criteria that should be used to correlate Transportation
Time to the Vibration Table Time. For this investigation a product was chosen where
damage could be easily assessed. Measuring the settling, or displacement of the cereal,
over time was a seemingly simple way to evaluate the damage for cereal. It is
important to realize, however, that it is not always easy to define damage criteria for
products.

Specifically, the findings of this research are as follows:

1. The average settling for cereal is 0.0024 inches / minute for the Actual
Transportation Environment.

2. The average settling for cereal is 0.007 inches / minute for the Simulated
Transportation Environment.

3. Neither the type of cereal (bran or flake) nor the position of the carton in the
shipping container (top or bottom) had any effect on the average settling of the
cereal. This was true for both the Actual and Simulated Environments.

4. Based on findings in 1 and 2 above, it took approximately one-third of the
transportation time to simulate the same amount of settling on the vibration table
for the product, mode of transportation, and route of transportation used in this
study.

The findings of this research support the notion that the vibration table is a time-

47
saving device. In fact, the evidence in this study shows that the vibration table was
capable of creating the same amount of damage as the actual transportation
environment, in one-third of the time. Although justification for the vibration table is
not necessary, it is interesting to note that if the findings of this research were
applicable for all product/package systems and routes of automotive transportation, the
vibration table would certainly be worth its cost.

Although there may be flaws with the techniques used in this experiment, the
fact remains that the relationship between transportation time and vibration table time is
not an equal one. It has become quite apparent that determining a correlation is an
extremely complex task. There are literally hundreds, maybe thousands, of
experiments which could be done to check the major finding of this research, that
transportation time and vibration table time are not equal. Different products and
routes of transportation could be used in further attempts to find some correlation.
Also, different modes of transportation could be investigated. It would certainly be
worthwhile to study other products. One idea would be to use a granular product such
as salt, sugar, or flour and measure the amount of product breakup due to vibration.
The amount of breakup could be measured using different size sieves. Another
example of future work may include overlaying real time (G versus time) plots of
actual versus vibration table and determine if a 3:1 relationship can also be found.

There are some questions which were raised in the process of doing this thesis
which are not answered here but, perhaps, may be in the future. Since correlation

between transportation time and vibration table time is not 1:1, is simulation on the

4 8
vibration table merely a matter of using less test time, or should PD levels be
decreased? What are the effects of the random. controller giving the table the same
frequencies and same amplitudes but in a different order? Also, what effect does the
PSD data have since the process of generating this data is one of averaging out the
small vertical accelerations?

Needless to say, there are numerous options for packaging professionals and
students to investigate in the world of vibration. Whether the product be a stereo
component, computer, apples or cereal, the effects of vibration should always play a
role in determining how the product should be protected.

As described in this thesis, the need for simulating the distribution environment
is a tremendous one. The problem with simulation is determining the length of time
the product needs to be on the vibration table to get comparable damage in reproducing
the actual transportation environment. Simulating the actual environment incorrectly
could have some harsh consequences. If the simulated environment is too severe, one
may find that more packaging material than necessary is needed to protect the product
and so packaging cost rises. The consequence of under-packaging can also be a costly
one in terms of product damage.

The findings of this research in no way suggest that the vibration table time is,
at all times, one-third of the actual transportation time but that these times are not
equal. For the particular product, mode of transportation, and route of transportation
in this study, this was certainly the case. The method outlined in this research is

definitely applicable to other sets of conditions.

APPENDIX

EVENT

Q

ififiéfifififi

6§§$B§§88§

83

42

174

810

DATE TIME
MM-DD-YY HH—MMas
923-94 193-30
9394 14-3-69
93-94 1442-4
9394 15—6-25
9394 192931
93-94 14-12-32
9394 14-50-41
93-94 14-56-4
93-94 14-16-56
9-3-94 14-27-57
93-94 143919
93-94 14-13-12
9394 14-47-51
93-94 14-16-47
93-94 14—49-55
93-94 14-35-7
9394 14-14-39
93-94 14-54-4
92994 143436
93-94 14-52-3
93-94 14-26-3
93-94 15356
93-94 14-32-39
9394 14-47-14
93-94 14-17-19
93-94 14-12-3
93-94 1446-33
93-94 15910
93-94 15-6-38
9394 14-31-52
923-94 14-13-3
93-94 146446
93-94 1466-36
93-94 14—52-40
9394 14-915
93-94 1920-35
93-94 14-55-5
9394 14-57-19
93-94 14-5-69
93-94 14-31-46
9394 15-16-53
93-94 14-1049
923-94 “-5536

93-94 1991

93-94 146-34
9394 14-32-20
93-94 1946

9394 14-35-20
9394 144956
93-94 14-2927
93-94 14.591

9394 191942
9394 14-46-26
9394 146231
93-94 15-12-36

APPENDIX - EDR DATA

MEAN: MEANy MEAN: PEAKx My

6'9

6'9

6'2

6'2

6‘:

0.039
0.141
0.01 4
0.1 18

PEAK z
6':

RMSx
6':

0.030
0.188
0.180
0.171
0.192
0.139
0.132
0.129
0.141
0.13
0.1%
0.138
0.132
0.173
0.123
0.132
0.162
0.139
0.151
0.148
0.162
0.146
0.137
0.148
0.13
0.158
0.149
0.158
0.134
0.156
0.151
0.137
0.140
0.128
0.153
0.163
0.148
0.157
0.133
0.129

0.129
0.172
0.150
0.116
0.173
0.158
0.148
0.130
0.134
0.137
0.148
0.127
0.154
0.145

RMS 2
6'2

50

OUR x OUR y 00R 2
exam“ unnmu NKUMB
008 0&7 003
0.012 0.010 003
0.011 0.014 0.011
0.012 0.014 0.010
0.017 0&2 00%
0.012 0&5 0.03
0.012 0.03 003
0.014 0019 0.03
0.012 0&3 0GB
00'8 003 003
0.012 0.041 0&7
0.012 0.010 0.03
0.009 0019 003
0.011 0&7 003
0.010 0th 0.011
0.010 0.010 003
0.013 000 0GB
008 0&4 003
0012 0&1 0%
0.011 0&2 003
0.010 0&1 0.003
0.010 0.011 0.03
0.01 1 0.01 0 003
0.09 0&0 003
003 0.010 0.03
0.012 0&4 003
0.010 0&1 0&7
0.013 0&1 0.03
003 0.019 0.09
0.010 0040 0.”
0.020 0.014 0.0.3
003 0.010 003
003 0.019 0.03
0.010 0.011 003
0&0 0.020 0.0!
003 0.010 003
0.09 00!) 0.03
0&6 0&4 01m
0&1 0GB 00$
0&5 0.014 0.03
0.010 0.011 0.03
0&5 0&0 0.011
0&6 0&1 0.010
003 0.010 0.010
0.014 0.022 003
0.010 0‘ 003
0.012 003 0.03
0019 0042 0GB
0040 0.030 0&7
0.015 003 0.03
0013 0.01 1 0&7
0.011 0.019 0.03
003 009 0.03
0.011 0&3 003
003 0.011 0.011

VEL 2
ftleee

0.044
0091
-0.188
0.143
0.103
0103
-0.104
0103
0.102
«0.13
-O.114
0100
-0.124
-0.192

0131
0116

LIST OF REFERENCES

9.

LIST OF REFERENCES

. Burgess, Gary, Lecture Material from Advanced Dymmics Course, Spring 1994.

Michigan State University.

. Robson, J. D. An Introduction to Random Vibration, Edinburgh: Elsevier

Publishing Company, 1964.

. Marcondes, Jorge A., Mark B. Snyder, S. Paul Singh. "Predicting Vertical

Acceleration in Vehicles Through Road Roughness”, Journal of Transportation
Engineering, 33-49 (January/ February 1992).

. ASTM D4169-93, "Standard Practice for Performance Testing of Shipping

Containers and Systems”, Selected ASTM Standards on Packaging, 237-247, 1994.

. ASTM D4728-87, ”Standard Test Method for Random Vibration Testing of

Shipping Containers”, Selected ASTM Standards on Packaging, 740-745, 1994.

. Ming, Xu. Effect of Vibration and Package Type on Bruising in Apples. Masters

Thesis. 1992.

. Don Hatfield, Engineer at Instrumented Sensor Technology, Inc., (517)349-8487

Personal conversation, October 5, 1994.

. VanValkenburgh, Paul. "GM Motorsports Safety Technology research Hitting

the Wall...’, Racecar, V013 No 4.

Instrumented Sensor Technology, Inc., User's Manual for EDR-3, 1994.

10. Instrumented Sensor Technology, Inc., 4704 Moore Street, Okemos, Michigan.

11. Root, Dale. Touch Test Vibration System Station Version 1.1 Beta, Released

March 2, 1993, Copyright 1990-1993.

52

RIES

"010111101