Created: 10/19/99 |
A Framework for Decision Support in Concurrent Product Development
Concurrent product development involves techniques to help manage the simultaneous design process. A fundamental challenge of concurrent product development is how decisions can be supported by providing current, and complete yet succinct design evaluations. One approach involves a decision support framework consisting of a "design attribute hierarchy" (an ordered structure of all design attributes and sub-attributes for a product), and a "design object hierarchy" (a structure of all hardware components in a product). some issues of framework development have, however, remained and include: guidelines to determine the design attributes essential to product superiority; guidelines to assist design attribute hierarchy assembly; enhancements to ensure that attribute mappings alert high-level decision makers of design failures; specific rules to prevent erroneous mapping; a theory to develop the design object hierarchy structure; and methods to merge the design object and design attribute hierarchies. This thesis addresses these issues and explores solutions. to demonstrate how the solutions to these issues contribute to the framework's implementation, an example product that is part of a submarine weapon system is considered.
Braunberg, Gregory C. (M.S. Mechanical
Engineering)
Computer Aided Infrared Thermography Using Emissivity Compensated
Imaging
Infrared thermography is a full field, non-contact temperature
measurement technique. The measurement of radiant energy emitted from a
substance provides the basis for infrared thermography. In order to
utilize infrared thermography consideration must be given to the emissive
properties of the target object and to the uncertainties associated with
the measurement. This paper develops a methodology for emissivity
compensated infrared thermography to obtain true temperature measurements.
In addition, the measurement uncertainty in the temperature measurement is
developed along with guidelines for its application.
Ko, Wing Fu (M.S. Mechanical Engineering)
A Systems Engineering Approach to Design a Smart Tool Post
Structure
Precision machining has received more and more industry-wide attention
as dimensional accuracy becomes a significant measure of quality in a
product. the key in achieving today's quality requirement is, therefore,
precision of a machine tool. Since the invention of the first CNC machine
tool in the 1960s, machine tool research has entered an almost stagnant
stage. There are numerous reasons for the slow progress, and the lack of
system-wide studies of the machine tool performance is one of them.
The research presented in this thesis focused on improving machining
accuracy using a systems engineering approach. A conventional lathe during
machining is taken under consideration as a machining system. The tool
post is identified as a critical component in the machining system to
achieve the defined machining accuracy. Smart material made actuators
are used to design a new tool post structure that is capable of carrying
out an active vibration control during machining.
In this thesis research, the fabrication of the designed tool post is
completed. results obtained from the initial test strongly demonstrate its
capability to attenuate tool vibration during machining in an active and
intelligent way. Thus, the smart tool post system fulfills the design
objective of achieving microscopic level machining precision on a low cost
conventional machine tool platform. Suggestion on the actuator
specifications are made for further improvement on vibration compensation.
Transient Thermal Stress Analysis of Plated Through Holes Subjected
to Wave Soldering
The manufacture of a printed wiring board (PWB) assembly involves
attaching components onto the board often using wave soldering processes.
Because the PWB assembly is exposed to temperatures during wave soldering
which are higher than any related operating temperature, understanding the
heat transfer and potential damage mechanisms which arise during the wave
soldering process is critical to quality control and reliability. In
particular, cracks may be initiated during the wave soldering transient
depending on the manufacturing quality of the plated through holes. Theses
cracks may be arrested or masked due to the presence of solder and become
manifest after operational cycling.
This paper presents a study of the heat transfer mechanisms, stresses
and deformations which occur during wave soldering. a transient non-linear
thermal stress analysis was conducted to study the heat transfer and
potential damage mechanisms which arise during the wave soldering process.
Two different kinds of models, three dimensional orthotropic and
axisymmetric orthotropic are used and the results compared. Temperature an
stress/strain history curves are examined to determine the impact of wave
soldering operation on the fatigue life of plated through holes. the
effect of factors such as the presence of innerplanes in a PWB assembly on
the maximum stresses and strains developed in the PTH is investigated.
Osterman, Michael (Ph.D. Mechanical
Engineering)
Placement Methods for Electronic Components on Printed Wiring Boards
Based on Reliability and Routability Measures
The placement of electronic components on a printed wiring board (PWB)
is a complex problem which requires tradeoffs between several goals.
Traditionally, placement techniques have focused on improving routability
based on minimizing the total wire length between interconnected
components. However, electronic card assembly (ECA) reliability, which is
measured in terms of time to failure, cycles to failure, or the hazard
rates of the individual components, the interconnections, and the PWB, is
also affected by component placement.
The reliability of an ECA is a function of the design, manufacture,
assembly, environmental conditions, and stresses which include humidity,
vibration, shock, and temperature. The temperature dependent failure of an
ECA can generally be described by functions of component junction or case
temperatures, threshold temperatures, temperature changes, temperature
gradients, and/or temperature histories.
In this thesis, the placement problem will include reliability and
routability. The placement of components for reliability will be explored
based on physics of failure based on temperature failure dependency of
components. Placement procedures are developed so as to minimize the total
hazard rate of the components on a PWB cooled by either convection or
conduction cooling technologies. Finally, a placement procedure is
developed to combine placement for reliability and routability.
This effort is intended to expand the range of knowledge in electronic
layout by allowing for upfront consideration of both reliability and
routability. In addition, the development and verification of the
mathematical theory and tools necessary for combined placement is a
innovative and necessary step in remaining competitive in electronic
system design.
Reinhart, Hugh Steven (M.S. Mechanical
Engineering)
Automated Maintanability (M) Modeling and Analysis
This thesis examines a set of computerized techniques which can be used
during the system design process to predict and utilize maintainability
(M) system parameters associated with repair times, manpower
requirements, and false alarm rates. Included is a description of a
software module which uses statistical techniques to quantify various
aspects of maintainability. The module provides the ability to create user
specific elemental task databases which can be used to compile knowledge
of maintenance times gained through experience with similar systems
already in operation.
Hierarchical modeling techniques make it possible for M
parameters to be calculated for the entire system, or any level of a
subsystem without reentering the failure mode and repair data. This
encourages modular design and improves the finished product by speeding
the design process, allowing the analysis of more design options, and
providing estimates of the operational behavior of the finished system.
Roza, Scott A. (M.S. Mechanical
Engineering)
Vibrational Modeling of Wedge-lock Edge Guides
The accurate determination of the natural frequency of a printed wiring
board (PWB) in its working environment is an essential part of its design.
The natural frequency is critical in determining the amount of vibrational
fatigue damage the board and its components will sustain.
The greatest variable in determining the natural frequency of PWBs is
the type of support provided by the edge guide connectors. Traditionally,
the classical types of support (free, simple, or clamped) are assumed to
exist at the edges. In reality, edge guides are of a construction that
limits translation and rotation but cannot completely eliminate either.
Therefore, the actual natural frequency of a PWB will fall somewhere
between the values calculated for simple support and clamped support.
This thesis focuses on how commonly used wedge-lock edge guides perform
and affect the natural frequency of a PWB. In particular, the wedge-lock
edge guides were modeled as being rigid in translation but elastic in
rotation. Since the edge guides are assumed to be elastic in rotation,
they can be further modeled as rotational springs with an unknown spring
constant. Vibrational testing was performed on a variety of edge guides
manufactured by Calmark Corporation. The analyses of the edge guide test
data with a finite element program allowed the calculation of
corresponding rotational spring constants (Kr). Values for
Kr were then integrated into plots to produce "look-up" tables
to determine Kr in any environment. This research is critical
to the accurate modeling and design of PWB natural frequency.
Sharif, Irfan (M.S. Mechanical Engineering)
Effect of Dimensional Variabilities on Lead Compliance and Solder
Joint Fatigue Life
Lead compliance is a critical paramenter in optimal design and
interconnection reliability of surface mount leaded components. the cyclic
force transmitted to the solder joint in surface mount leaded components
is controlled in part by the lead compliance. In this paper a methodology
for computation of lead stiffness and prediction of fatigue life of the
leaded surface mount components has been developed. Three dimensional
finite element analyses are performed to obtain 12x12 stiffness matrices
for both the PQFP gullwing and PLCC J leads and solder joints. These
stiffnesses are then used in predictive fatigue life equations to estimate
the fatigue life. The stiffness matrices and diagonal lead stiffnesses for
the basis for identifying more failure resistant packages.
Variabilities in lead and package dimensions provided by different
vendors, manufacturing to JEDEC standards, are identified and their
adverse effects on solder joint fatigue life are studied with the help of
finite element parametric analyses. Eighty different finite element
analyses are performed to study the effect of change in lead length,
height, width and thickness on the lead stiffness and solder joint fatigue
life for both the PQFP and PLCC attachments. Finally recommendations are
made in order to obtain a better control on component fatigue life.
She, Jieyu (M.S. Mechanical Engineering)
A New Algorithm for Evaluating Complex System Reliability
An algorithm has been developed to evaluate the terminal pair
reliability for complex systems. the minimal path set concept is extended
to all non-target nodes, and a new concept of subpath is proposed. A
subpath is a part of a path which is disjoint to the rest of the path.
Subpath intersections are eventually used in path intersections. By means
of a disjoint branch set technique, the partial disjoint features of
subpaths are used to reduce the number of path intersections in the
calculation of the system reliability. An efficient algorithm is obtained
to calculate the terminal pair reliability for the directed and the
undirected networks.
Vodzak, John (M.S. Mechanical Engineering)
Coupled Thermal and Vibrational Fatigue Analysis of Solder Joints
for Surface Mounted Components
The combined effects of elastic and plastic strains on solder joint
reliability are investigated. The dependence of fatigue life on strain
amplitude is found to obey the classical generalized Coffin-Manson
relationship in available experimental data. The commonly adopted
approaches of relating only plastic strains to fatigue life or the total
strain to fatigue life with a single power law relationship are shown to
be inadequate when predicting solder joint reliability. Instead, both
elastic strains and plastic strains should be considered independently,
especially when the electronic assembly is subjected to a combination of
large amplitude thermal loads and relatively lower amplitude vibrational
loads.
A methodology is presented to evaluate the combined effects of
simultaneous vibrational and thermal cycling of solder joints. The damage
from the two load-types are superposed to assess the overall effective
fatigue reliability of a solder joint. As a first order approximation,
linear superposition rules such as Miner's rule are utilized. Reliability
predictions from this simple model are compared to thermal low cycle
fatigue models for three simple combined loading problems.
A simple first order mechanical model is developed to determine solder
joint strains, by considering PWB deflections that are produced by
vibration. The model determines solder joint strain as a function of the
local radii of curvature in the PWB. Solder joint strain levels predicted
by the mechanical model are shown to have amplitudes which will cause
appreciable fatigue damage when compared to thermal fatigue damage.
Watts, Jonathon David (M.S. Mechanical
Engineering)
Placement for Producibility and Assembly
Design for producibility describes a method by which products are
designed to be compatible with the most capable manufacturing processed,
equipment, and production practices available. As such, this paper
develops a methodology for placing microelectronic components on mixed
technology printed wiring boards as a function of the ratio of surface
mounted to plated through-hole components, the component package styles
used, the number of sides of the printed wiring board to be populated, and
the abilities of the production equipment to be employed. Within this
methodology, a procedure for selecting a global board-level assembly
process plan is discussed and an automated routine for determining
microelectronic component placement positions and orientations on
populated and semi-populated mixed technology printed wiring boards is
demonstrated. In addition, a procedure which compares the printed wiring
board area required to accommodate the components with globally defined or
component specific inter-component space requirements against the
available printed wiring board area is described. Finally, an automated
routine for generating the microelectronic component layout with
acceptable inter-component spacings utilizing a force directed placement
technique is proposed.