Get documents about "Industrial PC Based Data Acquisition and Processing System
Shared by: xvi11400
Categories
Tags
data acquisition, data acquisition system, signal conditioning, data acquisition systems, data loggers, process control, acquisition boards, control systems, control system, industrial computers, process automation, test and measurement, automation controllers, personal computer, pc-based data acquisition
-
Stats
- views:
- 11
- posted:
- 7/6/2010
- language:
- English
- pages:
- 34
Document Sample
Industrial PC Based Data
Acquisition and Processing System
For:
Team Members:
Megan Bessick
Brett English
Pete Jankovsky
Ramesh Narayan
James Roll
Matt Sigg
Faculty Advisor:
Dr. Srinivasa Vemuru
Industry Advisor:
Lance Murdock
Industry Mentor:
Ron Hulsey
Date: November 5, 2004
Executive Summary
The goal of this project is to design a proprietary force monitoring and
analytical measuring system. The system will be used in conjunction with Grob’s
machining and assembly equipment. The final system will consist of an industrial
PC with LCD display, data acquisition card, and a custom user interface which
will allow the user to control the various applications of the system. This
statistical software will not be fixed for specific applications, but rather allow for
further customization as needed. Upon completion of this project the prototype
will be capable of monitoring industrial processes by creating pressure curves
and depth measurements via the data obtained from the inputs. The software for
this system will graph each part’s data points on a two dimensional graph. This
information, when compared with user entered standards, will then indicate
whether the part passed or failed. All data collected will be stored on a self-
maintaining archival hard drive, whose functionality will be implemented during
development. Ultimately, this will lead to the determination of the integrity of the
parts being tested.
Another objective of this project is to design the system in a manner that
promotes further system development and expansion. This will ensure that Grob
employees are able to benefit from the system in future years of operation.
- 1 -
Table of Contents
Executive Summary ................................................................................................ 1
List of Tables........................................................................................................... 3
List of Figures ......................................................................................................... 3
Company Background ............................................................................................ 4
System Specifications............................................................................................. 5
Operational Description .......................................................................................... 6
Constraint Analysis ................................................................................................. 7
Design Deliverables ................................................................................................ 10
Implementation Considerations............................................................................... 10
System Design........................................................................................................ 11
Hardware Research ......................................................................................... 11
Hardware Decisions......................................................................................... 12
Software Research .......................................................................................... 13
Software Decisions .......................................................................................... 15
Software Development Flow ............................................................................ 17
User Interface Specifications ........................................................................... 18
Project Scheduling .................................................................................................. 19
Budget ................................................................................................................... 20
Conclusion .............................................................................................................. 21
Appendix A: References ......................................................................................... 22
Appendix B: Decision Matrices ............................................................................... 24
Appendix C: Complete Gantt Chart......................................................................... 30
Appendix D: Team Member Resumes .................................................................... 33
- 2 -
List of Tables
Appendix B:
Table 1: Decision Matrix for Operating System....................................................... 25
Table 2: Decision Matrix for Programming Language ............................................. 26
Table 3: Decision Matrix for the DAQ Component .................................................. 27
Table 4: Decision Matrix for Industrial PC Selection ............................................... 28
Table 5: Decision Matrix for User Peripherals......................................................... 29
List of Figures
Figure 1: Operational Flow Chart ............................................................................ 6
Figure 2: Software Flow Chart ................................................................................ 18
- 3 -
Company Background
Grob International was founded in 1926 by Ernst Grob. Although Grob
has locations throughout the world, its headquarters is in Mindelheim, Germany.
The United States office is in Bluffton, Ohio. Grob manufactures a variety of
products geared towards the automotive industry. Grob’s product line includes
horizontal machining centers, flexible machining systems, automated assembly
systems, and transfer machines. Some of Grob’s major customers include Ford,
General Motors, DaimlerChrysler, and Harley-Davidson.
Many of Grob’s products are custom designs in which the product is
tailored to the customer’s needs. The Bluffton, Ohio location employs roughly
one-hundred and eighty people, whose jobs consist of engineering, designing,
building and testing the machining and assembly systems. Grob’s machining
and assembly systems popularity has increased overseas due to compact
designs that efficiently use floor space. These products are designed in such a
manner that a cast part can be transformed into a fully functional part that is
ready for installation.
Grob is continually striving to uphold superior product and service quality.
Environmental obligations are weighted heavily throughout company
functionality.
Grob products include the following: special machine tools customized to
customer requirements, automatic assembly equipment, highly-dynamic
horizontal spindle machining centers, flexible manufacturing equipment, Gantry
robots, and automation systems.
The company supplies machining and assembly equipment as an
integrated system and is responsible for this complex technology. The Bluffton,
Ohio plant serves automotive customers in North America, Europe, and Asia.
Systems are designed, constructed, assembled and tested in this facility. The
design, construction and testing requires the use of Industrial PC to display
pertinent information and archive data. The data is acquired from sensors placed
on the machines operating at the Grob Systems (transducers), which record
information, such as load data and force data. It then displays this information in
form of graphs. This system needs to be developed further to add features for
storing and retrieving data as needed by Grob employees. Currently, Grob
purchases completed measuring systems from other vendors to use on its
equipment. The vendor provided software used with these systems contains
several functionalities such as acquiring data, processing data, timer controls to
control the acquisition of data, calibration and diagnostics etc. This project is
Grob’s first attempt at creating its own analysis system.
- 4 -
System Specifications
Grob Systems, Inc., the customer, requires an industrial computer system
that will have a function of acquiring raw data, processing the data, presenting a
useful user output, and verifying a part's design conformity. These actions will be
performed on each part handled by Grob's manufacturing machine, and
respective data will be stored and associated with that specific part using a
timestamp. The data will be read in from sensors for which the customer will
provide specifications. The computer will then process the raw data to convert it
to values representing the sensor's specific measurement. The output will be
specified by Grob, including such items as pressure curve, current measurement,
etc. The system will then compare its processed data to standard or expected
values provided by Grob to determine an individual item's pass or failure.
This computer system will be designed by the team: multiple alternatives
will be considered and a prototype will be produced. Documentation will be an
important focus during the entire process. Upon completion of the project, the
team will present Grob with sufficient commented code and other documentation
to allow Grob programmers to maintain the system. Considerations will be taken
in deciding system appearance and user interface. The hardware cost to Grob
should be in the range of $4,000-$10,000 per unit and shall be designed as
components.
- 5 -
Operational Description
The system first reads in a sample from an analog signal. It then converts
this value to a digital value. It then tests if the samples received indicate that a
part has been completed, and continues reading in samples until a part is
completed. The samples are next compared with previously defined tolerance
levels. As samples are processed, they are sent to the display to be graphed.
After all points are processed, the system determines if the part passes or fails
and displays this result. While this is happening, the space remaining in the
storage device is checked, and deletes the oldest file if more space is required.
The values for the samples and the result of the pass fail operation are then
written to the storage device. See Figure 1 below.
Analog In Check for full Delete oldest
data storage lot data when Output part
User configurations more space data to file
required
Convert to Compare
Digital measured values to Pass / Fail
set tolerances and Output
determine pass or
fail
Test for part Display
operation
completed
Figure 1: Operational Flow Chart
- 6 -
Constraint Analysis
Engineers make complicated decisions when determining their designs.
More complicated designs should require more consideration than a
straightforward problem, so it follows that the senior design project will
encompass many decision considerations. This analysis focuses on the
constraint issues likely to be encountered in Grob Group’s design. Some issues
that are important aspects of engineering include: economic, environmental,
sustainability, manufacturability, ethical, health and safety, social, and political. A
complete system life cycle design can be further developed by considering all of
the aforementioned factors.
This system life cycle approach will assist in creating a product that
conserves time and money in the later development stages. As the world is
developing into a global environment it is important for engineers to think about
the design process from the systems engineers’ point of view to ensure products
are meeting the needs of their customers without harming the global public.
Economic
The major economic factors are the costs associated with producing a
prototype and the cost associated with mass-producing the product.
The cost of producing a prototype would include the average cost of $25
per design employee per hour, the costs of parts and equipment such as licenses
for the operating system software, parts, testing equipment, and overhead costs
such as renting building space, electricity, and administrative costs. A budget of
$4-$10k was designated by Grob Systems, Inc for use in developing the
prototype.
The costs associated with mass producing the product are very similar to
the prototype costs but also include the costs associated with the machinery
needed during the assembly period, and the reduced cost of skilled laborers. An
additional economic factor associated with this product will be the cost of
employing someone who can provide support for Grob customers after they have
implemented the system in customer solutions. Since this is an industry
sponsored project and none of the team members will be employed by Grob after
project completion, the cost of bringing Grob employees up to speed with the
project so they can maintain it and update it as needed also must be considered.
This cost and time can be reduced by thorough documentation.
Environmental
The disposal of the product at the end of its life cycle is our major
environmental concern. Our self-regulating archival storage system reduces
media waste by allowing for data to be stored and deleted in a first, in first out
manner. In an attempt to reduce the amount of power consumption, our system
will be implemented with electrical efficiency in mind. Efficiency can be achieved
by choosing a PC with a lower operating frequency. A large display is
unnecessary, so the utilization of LCD technology rather than CRT technology
will be a benefit to the environment during disposal. An LCD display will also
result in lower power consumption and lower interference generation.
- 7 -
It is also important to consider the constraints of the operating
environment. Staying up to the industrial standards in the selection of our
industrial PC is the most important concern in this area. The customer specified
that the system must be up IP65 standards, meaning that it is robust enough to
work in industrial environments. We therefore only considered PCs that met that
requirement.
Sustainability
Sustainability of the hardware components can be estimated through
ratings of mean time between failure and expected lifetime. The product will also
be sustained through easy location and replacement of defective parts. The
safety of the data must be ensured in this product—a power outage should not
equate to loss of data. This can be helped by immediately writing data to the
hard drive upon acquisition rather than storing in memory. Redundancy in the
hard drive can also be investigated using a RAID hard drive array.
To enable our product to be sustained by our company sponsor after
development, it is important that we have documented our code thoroughly.
Thorough documentation of the code will help the company in maintaining and
upgrading the system. It is also important to provide documentation to the
customer through a basic user manual. This manual will cover navigation
through the software we designed and will include some troubleshooting points.
Ethical
One major ethical factor within this project is the issue of reengineering
our product to have the same functionality of a currently patented product. We
have researched the U.S. Patent Office filings and found some existing patents
for similar systems. The group is independently determining the implementation
of this functionality and shouldn’t presently be concerned with licensing.
Ultimately, it is up to Grob to abide by these patents should they choose to
implement our design.
An additional ethical factor that must be incorporated into our project is the
appropriate citation of researched sources in our written documents throughout
the project. All sources of research have been logged which will eventually be
recorded in a bibliography to give credit to original authors and researchers.
Manufacturability
Although our participation in this project will not include the manufacturing
stage, this factor is still important to consider in designing the system. Our major
concern is that Grob will manufacture this system for multiple applications, so it is
important to design a flexible system.
The first manufacturability factor to consider is the use of standardized
and common parts and materials. By using these types of parts, the
manufacturer can minimize the amount of inventory needed and standardize the
handling and assembly operations.
Software deployment is also an important manufacturability component to
factor in. In an attempt to make the installation process as efficient as possible it
- 8 -
would be advantageous to create an image to use when installing the software
onto the PC.
Health and Safety
Ergonomics is the project’s only noteworthy health concern. Ergonomics
should be considered in the placement of input devices to reduce soft-tissue
strain on the user. Componentizing the system would be beneficial in that
transportation and set up in the new environment can be accomplished easily.
Social
One major social factor is the user interface. The use of large buttons and
text sizes will allow for easy user input. The user interface must also be
designed keeping in mind that the user may be left or right handed. Another
important factor is the use of a complementary color scheme that is easily read
by all users, including those who may be colorblind. The main interface should
be usable by anyone with a basic knowledge of how to operate a computer.
Common color standards can be implemented in the design: green for good part,
red for bad part. Finally a user manual will need to be produced for training
purposes.
A second social factor is the ability for Grob to improve their customer
service. By providing this part of their equipment in-house instead of
recommending another company, Grob will be able to give more complete
customer service and build better relationships with their clients.
Finally, another social factor is that our product will assist in the detection
of errors in the manufacturing process. By detecting errors and removing bad
pieces before they go any further in the manufacturing process, our system is
helping to increase the number of reliable products that go to market. This will
result in safer products, such as vehicle components being manufactured by
Grob’s machining equipment. Since the system will require little attention from
the user while still detecting defective parts, more of his attention can be focused
on more pertinent aspects of his job, which should also result in more reliable
end products.
Political
Political factors lend to maintaining good business relationships. Since
our product is being produced for an industry corporation, it is important that the
group maintains a good faith relationship with the industry contact so that they
are encouraged to participate in future Ohio Northern University senior design
projects. Political considerations generally imply a governmental relationship, but
this should not be an issue with this project—no government agency is involved.
Government and industrial standards will be followed as required. Further, Grob
has stated that it does not require the project to be kept confidential, but asks
that the team not actively advertise to Grob’s competitors.
- 9 -
Design Deliverables
Upon completion of this project our group will deliver to Ohio Northern
University College of Engineering and Grob a working and tested prototype
complete with documented source code. The prototype will be able to monitor
multiple industrial processes, but is currently specific for the application of
pressure curves and depth measurements. This prototype will be able to analyze
captured data and determine whether the manufactured part falls within specified
tolerances. When a part falls outside these tolerances, the system will mark the
part as a bad part. This system will also store all data captured from each part
onto on a hard disk drive. The system will constantly monitor the amount of
stored data so that once the amount of data reaches a specified size the system
will perform a self maintenance cycle and delete the oldest of the data to allow
for new data storage.
Implementation Considerations
Beyond the previously diagnosed design constraints there are three major
implementation considerations that must be reviewed while designing this
product. The issue of developing both the hardware and software via individual
components must be worked into the design to allow for easy maintenance and
upgrade in the future. The system hardware should be developed so that
individual components can easily be replaced or upgraded in the future. The
code associated with the software for this project should also be developed in
several independent components which can then later be tied together to
implement the whole software system. Building the software package so that it
consists of individual components will allow for easier maintenance and upgrades
in the future. Another important implementation consideration associated with
maintenance and future upgrades is the importance of documenting the source
code adequately. Since Grob employees will be responsible for upgrading and
maintaining this product when it goes into production, it is important that they
understand how the code works by reading the well documented comments. A
third implementation consideration is the calibration of hardware cards. If a card
is chosen that requires maintenance calibration, additional software components
will need to be developed to handle this calibration from the user interface.
- 10 -
System Design
Hardware Research
In determining the hardware solution, it is first necessary to specify
essential components of the system and the method of implementation.
We looked at two methods of obtaining an industrial PC that would meet
our specifications. As we see them, our options are to either assemble a PC
from individual components or purchase an industrial PC that was already
assembled. By purchasing an industrial PC that was already assembled, we
have the benefit of a system that has already been tested in the industry. The
company will have received feedback from their previous customers and has had
an opportunity to fix problems with their systems. Another benefit in choosing a
proven system is that the compatibility support of the system and components
are assured. Purchasing a system from a known and reputable source provides
access to adequate documentation and technical support. If we were to build a
pc from scratch we would be limiting the amount of support available and
warranty considerations between components could become an issue.
After deciding that we wanted to use a PC that has been proven to be a
viable system with compatible components, we need to look at whether we want
to use a one piece system that has the monitor built into the same case as the
PC or a system that has a monitor separate from the CPU which could be
mounted any where with in the machinery that is convenient. It is our conclusion
that we should use a one-piece system. In either case data and power lines
would have to be routed through the machinery negating this as a factor
influencing our decision. Using this system we will be able to cut down on time
required to install the hardware as well as the amount of materials needed, in
doing so we will also reduce installation costs.
When selecting an industrial PC it is important to consider the type of
environment it will be operating in. Safety in the work environment is a necessity
and as a result, great consideration is taken when electrical equipment is chosen.
Companies such as 3A and Factory Mutual ensure that safety is maintained in
the work place. The main functions of these companies are to examine, test, and
certify products to guarantee that they comply with standards for sanitary
applications, various electrical situations, and hazardous locations. The Ingress
Protection (IP) rating system is characterized by the letters IP followed by two
numbers (i.e. IP65). The first digit (i.e. IP65) designates the protection against
solid objects. The second digit (i.e. IP65) designates the protection against
liquids. Our customer has specified that the system should conform to the IP65
standards.
The customer requires a data acquisition card (DAQ) capable of capturing
data from different types of sensors depending on the application. Grob
specified a number of sensors that may be used in this project. One important
type of sensor is the linear variable displacement transducers (LVDT) to measure
position at high precision. The sensor outputs a voltage level within a specific
range that proportionally changes with position. The team investigated one of
- 11 -
these sensors that would require a high resolution: an LVDT with ±0.5 inch stroke
and infinite resolution. Grob specifies that the sensor used will have a 10 Volt
range and measure increments in the magnitude of microns. The team has
determined that a 16-bit DAQ card will provide measurement resolution of
167.8µV, which relates to position resolution of 0.381µm for the sensor.
The amount of hard drive logging capacity needs to be determined to
make an intelligent decision among hard drives. The amount of data that will be
logged per part has been worst case estimated by the customer as 775 bytes.
The customer has also specified that the worst case processing is 4800 parts per
day, which generates approximately 3.55MB of data per day. A 260 work day
calendar year would require 923MB of storage space. The minimum hard drive
capacity of the PC models investigated is 20GB, so hard drive capacity will not
be a major driving issue.
Hardware Decisions
One major component of the system is the user peripheral. The system
can feasibly be implemented with a traditional keyboard and mouse contained in
a pull-out tray, a touchpad mounted on the panel, or a touch screen. Ease of use
for maintenance describes the ability for a maintainer to interface the system for
setup. Programmability includes the ease of implementation for the input in
software drivers. Ergonomics refers to the user’s ease of use in interfacing the
system during normal use. Other factors involved in the consideration include
reliability and cost. Based on our rankings of these factors as in the decision
matrix of Table 5 in Appendix B, the group recommends using the traditional
keyboard and mouse user input.
A number of various industrial PC vendors were contacted regarding the
design of this system. Two outstanding vendors are Beckhoff and Daisy Data,
from which possible solutions were investigated. The customer expressed strong
importance on choosing a system meeting the industrial rating and appearance
of an industrial system. Other factors involved in the consideration include
processing power, storage capacity, PCI expandability, price, memory, display,
and company reputation. Based on our rankings of these factors as in the
decision matrix of Table 4 in Appendix B, the group recommends purchasing a
Beckhoff system, model C3320. This solution fits into a 19 inch rack mount, has
a 15 inch LCD display with 1024x768 pixel resolution, has a programmable
keypad on the front, conforms to the IP65 industrial standard, and contains an
Intel Celeron 2 GHz processor, 512MB memory, and 40GB hard drive.
The data acquisition card is critical to the success of this project. Once
the card’s specifications were determined, many similar models were discovered.
The determining factors between cards were price, company reputation, and
documentation and support. Company reputation is a very important factor in
choosing a product that will be reliable and replaceable. Documentation and
support is essential to the development of compatible drivers and troubleshooting
errors. Based on our rankings of these factors as in the decision matrix of Table
3 in Appendix B, the group recommends purchasing a National Instruments data
acquisition card, model PCI-6014. This card has 16 input channels with 16-bit
- 12 -
resolution, a ±10V input range, a sampling rate of 200,000 samples per second
and two analog output channels.
Software Research
While research was being done involving the hardware implementations
for the project, research was occurring for the implementation of the software
components. The two key software decisions that were researched were:
researching the operating system that should be used and the programming
language that should be used in writing the software.
Although there are many different types of operating systems to consider,
the project research was narrowed down to an in-depth comparison of the
Microsoft Windows XP professional, Microsoft Windows XP embedded, Microsoft
Windows C.E., and Linux embedded operating systems. This narrowing of
possible choices was mainly due to the reputation, reliability and available
documentation of these major operating systems.
The initial operating system investigated was the familiar Microsoft
Windows XP professional operating system. All research gathered on this
operating system indicated that it was best used for a PC that was used by
multiple users and ran general applications, such as notepad, Internet Explorer®,
etc. It was rated highly on its flexibility and reliability, yet requires a 1.6 GB
footprint. After this initial research was conducted, our project specifications
steered us toward the embedded operating systems category.
The next embedded operating system researched was Microsoft Windows
XP embedded. It was found to be the high-end embedded operating system for
advanced devices such as point of sale terminals, ATMs, kiosks, and high-end
electronics. It was designed to deliver the power of XP in component form for
special purpose dedicated devices. With over 10,000 individual components, the
size of footprint can range from a minimal size of 8 MB. Due to its smaller
footprint, it has a faster boot and log on time than the traditional XP professional
version. It also allows for flexible booting from a variety of locations. An
additional feature includes over XP professional’s ability to “lock down” the
interface, by allowing the user to install only the necessary components needed
for your application and replacing the Windows shell with a custom interface.
The licensing cost associated with XP embedded is almost half as much as the
cost for XP professional. XP embedded’s configuration is so similar to XP
professional that anything that can run on XP professional is suppose to be able
to run on XP embedded without requiring any modification. XP embedded also
comes with a large assortment of drivers for common input devices. One of the
limiting factors associated with XP embedded is that it only supports the x86
CPU architecture. After concluding in-depth research on XP embedded, it was
added to the list of viable operating system for the project.
Another operating system that was investigated for its viability on this
project was the other windows embedded operating system, Windows C.E. NET.
A real time embedded operating system with a small footprint, strong connectivity
features, and low licensing cost, make it a desirable operating system for devices
such as Voice over IP phones, residential gateways, and handheld devices. It is
- 13 -
also known for its robust multimedia support. Research seems to suggest that it
is a good operating system for devices that needed a small footprint, efficient
power management, and remote management capabilities. The minimum
footprint size can be 350 KB, and it can support both x86 and non-x86
processors. One drawback to this operating system is that Win32 applications
and drivers must be modified to run on this platform. With this operating system,
there exist three major licensing levels: Core, Pro, and ProPlus. Each upgrade
not only adds more functionality to the operating system but also increases the
unit license price increases with functionality. The respective unit license costs
are: $3, $15, $20. A third operating system that was checked into for viability
was the Linux embedded operating system.
Using an embedded Linux operating system is often very appealing due to
the lack of royalties on the kernel and abundance of documentation since it is
open source. Yet the amount of work required to bring the Linux kernel up to the
feature level of other embedded operating system can be extensive. Although
the kernel is free to download, to add additional features such as Web Browser,
Media Player, or MPEG-4 Decoder, additional licensing costs must be accounted
for. Another concern highly debated in the industry revolves around the reliability
of the operating system. Some question the reliability of the operating system if
not created by a professional.
After researching about different viable operating systems, research was
also done on different programming languages. Although there exist hundreds of
different programming languages, our research was narrowed to five major
languages due to their current industry popularity and supportability. C, C++, C#,
VB.Net, Java were the five languages specifically researched with additional
research also done on the benefits of using the .Net framework.
The first language researched, C, was found to have a large usage base,
making it easy to find help and libraries. It also claims to be powerful even with
its simple core language. Since C files are compiled into stand alone programs,
there is no need for an interpreter. An additional feature of this language is the
allowance of unchecked access to computer memory using pointers. One
drawback to this language is that it is not strictly object oriented. It also has a
history of being difficult to port complex applications. Since it is provides the
programmer so much flexibility it comes with the price of only having a little
safety net.
The second language researched, Java, is a platform independent
language that is easy to learn if a programmer knows C or C++. One popular
benefit of this language is the automatic memory management and the large set
pre-built libraries. One of the drawbacks of Java is the relative speed at which it
runs. Although newer versions claim to have fixed this problem, the fact that the
system’s virtual machine must interpret the code still leaves a performance gap
between interpreted and compiled languages. Another drawback to this
language is the difficulty associated with compiling the application into a
standalone application.
The third language researched, C++, is found to be another commonly
used language that supports object oriented technology. Since it is a compiled
- 14 -
language, it easily creates standalone applications that are easy to port
platforms. An additional benefit of C++ is the availability of many different
libraries. One of the drawbacks of this language is that is relatively difficult to
learn. Although it is commonly easy to port platforms, programs that use
platform specific libraries make porting platforms more difficult. Additionally, in
theory C++ is slightly larger and slower than C.
The fourth language researched, VB.Net, is easy to learn, quick to
implement, and provides lots of built in functionality. It is considered a cleaner
language because it has no macros like in C or C++. An additional feature of the
VB.Net language is that all relevant values are boundary-checked. One of the
downsides associated with this language is that it is not as flexible with other
languages. It also runs slower than C/C++.
The fifth language researched was the new C# language. Since it was
developed with the .Net framework, it has all the advantages associated with the
.Net environment. It is an object oriented language where ease of deployment is
an additional benefit. The enforcement of static typing allows for safer, more
reliable, and more efficient code. It allows for the use of inheritance and both
method and operation overloading to increase flexibility. Generational or Mark
and Sweep garbage collection is another advantage of this language. One
potential drawback of this language is that it can have a steep learning curve if
the developer is trained in C or C++.
In addition to researching programming languages, the .Net frame work
was also researched to look at the advantages of using this new technology.
One advantage of using the .Net environment is that it increases productivity
because of the ease of use of the libraries. It also has a strong framework for
rich graphical user interface. Another benefit is that it allows the user to choose
among twenty different programming languages. It is also tightly integrates with
the operating system. One potential downside to using this framework is that it
can have a steep learning curve. Additionally, the choice of IDEs is limited.
Another problem area with the .Net environment is that getting older applications
to run in the new .Net environment may require a substantial amount of work.
Software Decisions
Once all the research was completed on the different options for operating
systems and programming languages, decision matrices were developed for both
choices.
When making our decision about the operating system, the constraints
that the system was measured against included license cost, ease of
maintenance, resources required, and development time. The decision matrix is
included as Table 1 in Appendix B.
The embedded Linux operating system rated well in the product cost, due
to its lack of core kernel license, and also rated well in the resource requirements
category because of its ability to create an operating system with a small
footprint. It rated poorly in the development time category because it will require
a lot of development time to get the operating system up to the specifications
necessary to run our applications. It also rated poorly in the ease of
- 15 -
maintenance category, mainly because current Grob employees only have
experience with Microsoft products, so having them maintain it afterwards could
be a considerable hindrance.
The XP professional operating system seems to sit at the extreme ends of
the spectrum. It rated highly in the ease of maintenance because Grob
employees are already familiar with this operating system. The other category it
was rated highly in was development time because it would require minimal
configuration of the operating system to get it up to the specifications necessary
to run our application. It rated poorly in product cost because it has the most
expensive license cost out of all the choices. It also rated poorly in resource
requirement because it requires the largest amount of resources to run on a PC.
The XP embedded operating system scores in the average or above
average range on all categories. It is rated as average with product cost
because it is more expensive than the other two options, but almost half as
expensive as XP professional. It rated above average in all the other categories
when compared with the other options in their respective categories.
The C.E.Net operating system scored around the medium in the variety of
categories. It scored well in the resource requirements category because it
requires a small footprint and minimal resources. It also scored relatively well in
the product cost category because it has a low core license cost. The
development time category received an average score because additional time
will be needed to bring the operating system up to needed specifications. It
scored below average on the ease of maintenance due to its more complex
development requirements.
Although all the consideration factors are important, a necessary
weighting needed to be applied. The development time is weighted most heavily
at forty percent because of time limitations. It was determined that development
time should be minimized as much as possible in order to focus more attention
and time on the actual application development. The product cost, weighted at
thirty percent, is the next important issue. Because this product will eventually go
into mass production, considering the economic factor of mass producing this
project is important. The ease of maintenance category is weighted at twenty
percent due to the fact that Grob employees will need to be able to maintain the
product once the prototype is developed. The resource requirements category is
weighted at ten percent because the size of the footprint and resources needed
has become less important in the hardware design.
After analyzing the ratings with the weights taken into consideration, it has
been determined that that XP embedded is the most desirable operating system
to use for this project.
After determining the operating system to use, the next decision to be
made was which programming language to use. The major categories weighted
were maintenance by customer, GUI ease of use, flexibility, and integration. The
decision matrix includes the C++, C#, VB.Net, and Java languages. From our
earlier research C was taken out of the decision matrix because it was
determined that this project would need an object oriented language, of which C
fails to fulfill. Under the “maintenance by customer” category, VB.Net was rated
- 16 -
the highest because our customer has experience with VB and is willing to
consider using the .Net framework. C++ was rated as average because Grob
employees have had some experience with this language. Java was rated
lowest because it is very different from any of the languages Grob has had
experience with. C# was also rated low in this category because Grob is not
familiar with this language but would probably be able to pick up easier than
Java. The maintenance by customer category is weighed at fifty percent
because it is essential that our customer be able to maintain and upgrade the
product after the prototype is implemented.
The next category, GUI ease of use, was weighed at fifteen percent
because it is important that our language of choice allows the developers to
create a strong graphical user interface. Both VB.Net and C# were rated high in
this category because of their inclusion in the .Net framework that proves a
strong set of libraries for building graphical user interfaces. C++ was rated low in
this category because of its lack of integrated graphical interface libraries.
The flexibility category was weighted at twenty five percent because it was
determined that a language that has a large amount of flexibility was necessary
for the various components associated with this project. Overall the languages
that were considered scored about the same on this category with C++ and C#
showing a little more flexibility due to the robustness of their language.
Integration, the last category, was weighted at ten percent because we
were looking for a language that could be integrated with other languages if
needed in the future. Java was not rated well in this category because it is
difficult to integrate Java with other programming languages. The other
languages under the .Net framework can be integrated together fairly easily.
After incorporating all the weights and ratings, it was determined that
VB.Net would be the best choice of the programming languages as shown in
Table 2 of Appendix B.
Software Development Flow
In order to tackle this large project, the software component of the project
was broken into ten components. Figure 2 contains the flow chart of the order in
which the project is expected to be implemented, divided into important
components. The hardware interface component consists of implementing digital
acquisition card drivers. During the calibration/diagnostics stage the data
acquisition card’s accuracy will be confirmed and adjusted as needed. The
Design User Interface stage will be the stage where the graphical user interface
will be developed. The Save Data stage will encapsulate the functionality of
storing the data in the correct format to the hard drive. The Configuration of
Functionality stage consists of setting actions associated with physical buttons,
setting input channels, and defining archival intervals. During the Graph Data
stage, the functionality for graphing the data points to the screen will be
implemented. The Archive Components stage is to be used to develop the self
maintaining hard drive storage functionality for archiving and deleting data.
During the interface to Archive Data stage, the interface to see the archived data
will be implemented in addition to developing the functionality that populates this
- 17 -
interface. In the Analyze Archive Data stage the implementation of pulling up old
archived data and displaying it on the screen will be implemented. During the
Implement Multiple I/O Cards stage, the application will be tested with alternative
I/O cards to verify driver functionality. According to Figure 2, the software
development is mainly a linear process with the exception of the hardware
interfaces being developed in concurrence with the application software
development.
Hardware Calibrate/ Implement multiple
Interface Diagnostic I/O cards
Archive Interface to Analyze
Design User Configuration of Graph Data Components Archive Data Archive
Interface Functionality Data
Save
Data
Figure 2: Software Flow Chart
User Interface Specifications
The user interface consists of three main parts, the main program that
shows the results for each part as it goes through the system, a part to customize
the functionality of the program, and an interface for the archived data. The part
to customize the functionality will allow the user to define things like how long it
takes to process the parts, the upper and lower bounds for the failure test, and
the method whether archived data is stored by day, week, or month. Under most
circumstances this will only be needed when the industrial PC is installed and
during diagnostics. It would most likely never be used by the factory workers.
The interface for the archived data will allow for easy searching and management
of already processed parts. This component would not commonly be used by the
factory workers, if at all.
The most important part of the user interface is the main display of the
part data. It will be running at all times and will be what the factory workers will
use. The most important parts of the display will be the graph and the pass / fail
“light” that will indicate if any of the data points collected for the current part fall
outside the user defined limits. Both of these will be quite large so they can be
seen from a good distance from the screen. The display will also include a scroll
box listing the data points for the current graph. Another scroll box will contain a
list of past failed parts since the display is constantly updating when each new
part goes through the system. Finally, an array of buttons will be on the screen,
ideally lined up with the buttons on the case of the industrial PC, so that they can
be activated either by the physical buttons on the case or the pointer device.
- 18 -
These will be used to handle user input from controlling the graph to opening the
customization or archive interface.
Project Scheduling
The Gantt chart governing our design project consists of six main
categories. These categories are the planning stage, developing the written
proposal, hardware summary, software summary, testing summary, and the
documentation and reporting stage. The core implementation of the project is
located in the sub-divisions of the software and hardware summary categories.
Since the prototype design process consists of two distinct developing phases,
the hardware and software implementation will be ran in parallel. This allocation
of time and efforts will maximize our efficiency. The hardware and software
summaries have been allotted fifty-one days and forty-six days respectively to
complete. At the conclusion of these stages, the hardware and software aspects
of the project will be merged.
Testing will consist of two main phases; debugging software and hardware
during their individual developments, and intensive testing of the system after the
integration of software and hardware. Debugging will occur frequently during the
hardware and software development and will be independent of each other until it
is time for integration.
Specific plans for the intensive testing of the complete system will be
developed during the debugging phase. In general, the system will be tested
using a simulated input, and if time and resources permit, will then be tested
using an actual sensor.
Software will begin with the development of the basic user interface. The
hardware interface will be programmed as soon as hardware development
allows. After the basics of the user interface are designed, interface for the
customization of the functionality of the program can be done. When the
hardware interface is complete, the system to save the data can be implemented,
followed by the system for monitoring the archived data. After both data saving
and the basic user interface are finished, the graphing functionality of the
program can be started. After the archive monitoring system is finished, an
interface for the archive can be implemented. Finally, if there is time to develop
additional software at the end of the project, the software team will develop a
system to analyze archived data.
The hardware development begins with ordering and then assembling the
required hardware. After the system is assembled, any additional hardware
needs of the IO card will be identified. If additional hardware is needed, it will be
designed, built, and integrated with the IO card. When this is finished, design
and creation of testing equipment will take place.
When both hardware and software are finished with preliminary
debugging, they will be integrated and tested. This process will be starting with
the testing equipment developed by the hardware team, and then finally with an
actual sensor from Grob if time and resources allow. Both teams will be involved
in the creation of the test plan and the final testing.
- 19 -
Budget
The cost consideration in engineering projects is needed for profit and
non-profit companies. Financial gains through productivity increases often
provide justifications for spending the time to engineer a product, rather than a
trial-and-error approach. Grob will be designing a prototype for the company and
delivering it upon completion of the project. The project undertaken by Grob has
to consider cost as a constraint for the implementation of the design. The entire
cost of developing the project includes the cost of the following hardware and
software components.
• Industrial PC
• I/O Card
• Operating System License
Program Budget
Non-Personnel Expenses:
Component Expected Cost
Industrial PC $3859
I/O Card $475
Operating System $90
License
Total Expense $4424
Personnel Expenses:
The labor of design will not be charged to the company, but donated by
the students. An estimate of the hours required per week per student is 6 hours
during the fall quarter, 12 hours during the winter quarter, and 12 hours during
the spring quarter-- totaling 1800 hours for the team. At a rate of $25 per hour,
this relates to $45,000 of design labor cost for the project. Additional overhead
costs such as the cost of development tools that will be provided by the schools
MSDN academic licensing should also be taken into account. The development
kit for XP Embedded costs $995. The development kit for the National
Instruments products costs $3295. The license for Visual Studio .Net 2003
retails at $1799. Finally, we estimate $9000 for the universities computing
resources that we are using.
Therefore, the total project expense is estimated to be $64,513.
The majority of this cost is the engineering time, $45,000, is donated by
the students. The University has covered an additional $15,089 in resources
available. Grob is covering the remaining $4424, the cost of the hardware for the
prototype.
- 20 -
Conclusion
With the goal in mind of creating and industrial PC to meet the
needs and specifications of Grob, our team has investigated many alternatives
and developed this proposal to meet the customer’s specifications. We have
created a plan of implementation including a timeline and system deliverables,
with specified hardware and software components of the system. The final
system will consist of an industrial PC with LCD display, 16-bit data acquisition
card, and a custom user interface which will allow the user to control the various
applications of the system. The programming will be done in the VisualBasic.net
environment on the Windows XP Embedded platform. Our software will give the
customer power to customize statistical operations so that our system will be of
use in a vast array of operations. We feel that this system will serve Grob’s
needs, and it is our hope that the system will be smoothly implemented into
Grob’s product line.
- 21 -
Appendix A: References
- 22 -
[1] bSquare. Windows XP Embedded or Windows XP Professional [Online]. Available:
<http://www.bsquare.com/licenses/embeddedoscomparisons/winxpembedpro.asp>.
[2] bSquare. Windows Embedded or Embedded Linux [Online]. Available:
<http://www.bsquare.com/licenses/embeddedoscomparisons/winlinuxembedded.asp>.
[3] bSquare. Windows XP Emdedded or Windows XP Professional [Online]. Available:
<http://www.bsquare.com/licenses/embeddedoscomparisons/winxpembedpro.asp>.
[4] bSquare. Windows Embedded or Embedded Linux [Online]. Available:
<http://www.bsquare.com/licenses/embeddedoscomparisons/winlinuxembedded.asp>.
[5] “Comparison of Various Programming Languages.” Online Posting. 23 Jul. 2004. Dev Shed
Forums. 15 Sept. 2004 <http://forums.devshed.com/t167726/s.html>.
[6] Voegele, Jason. “Programming Language Comparison.” 25 Sept. 2001.
<http://www.jvoegele.com/software/langcomp.html>.
[7] National Instruments. “PCI-6014 Datasheet.” 2004. Available:
<http://www.ni.com/pdf/products/us/4daqsc208-209_212-213_230.pdf>.
[8] Measurement Computing. “PCI-DAS6014 Datasheet.” 2004. Available:
<http://www.measurementcomputing.com/pdfs/PCI-DAS6013-6014.pdf>.
[9] IO Tech. “DaqBoard /1000 Datasheet.” 2004. Available:
<http://www.iotech.com/vrs/tdc/registration/pdfregform.asp?url=/catalog/daq/daqboard1000.html
>.
[10] Meilhaus. “ME-4656 Datasheet.” 2004. Available:
<http://www.meilhaus.com/pdf/e_mefoxx.pdf>.
[11] Sensotec. “LVDT Model JEC Datasheet.” 2004. Available:
<http://www.sensotec.com/pdf/jecjecag.pdf>.
[12] Industrial PC, Inc. “WS-855AWorkstation Datasheet.” 2004. Available:
<http://www.industrialpc.com/pdf/workstation/ws-855a.pdf>.
[13] Beckhoff Product Catalog. 2004. Beckhoff. Available:
<ftp://ftp.beckhoff.com/Document/IndustPC/C33xxCE_e.pdf>.
[14] General Electric. “GE Industrial PC System Catalog.” 2004. Available:
<http://www.geindustrial.com/cwc/products?pnlid=2&famid=25&catid=163&id=ovcomp>.
[15] Grob System, Inc. “Corporate Website.” 2004. <http://www.grobgroup.com/en/_grob.htm>.
- 23 -
Appendix B: Decision Matrices
- 24 -
Table 1: Decision Matrix for Operating System
Embedded XP XP
Weight Linux Professional Embedded C.E.Net
Product Cost 30% 5 0 3 4
Ease of
Maintenance 20% 0 5 4 2
Resource
Requirements 10% 4.5 0 4 5
Development
Time 40% 2 5 4 3
Total 2.3 3.5 3.7 3
Product Cost:
0 – Low Cost
3 – Average Cost
5 – High Cost
Ease of Maintenance:
0 – Lot of Maintenance
3 – Average
5 – Little Maintenance
Resource Requirements:
0 – Requires lots of resources
3 – Requires average resources
5 – Requires few resources
Development Time:
0 – Requires substantial development time
3 – Requires average development time
5 – Requires little development time
- 25 -
Table 2: Decision Matrix for Programming Language
Weight C++ C# VB.Net Java
Maintenance by
Customer 50% 3 2 5 1
GUI Easy of Use 15% 1 4 4 3
Flexibility 25% 4 4 3 3
Integration 10% 4 4 5 0
Total 2.3 3 4.6 1.7
Maintenance by Customers:
0 – Causes difficulty in maintenance for customer
3 – Average maintenance difficulty level by customer
5 – Easy maintenance by customer
GUI Ease of Use:
0 – GUI tools easy to use
3 – Average level of difficulty
5 – GUI Tools difficult to use
Flexibility:
0 – Very Flexible
3 – Average Flexibility
5 – Little Flexibility
Integration:
0 – Poor integration with other languages
3 – Average integration ability
5 – Easy integration with other languages
- 26 -
Table 3: Decision Matrix for the DAQ Component
National Measurement IP Tech Meilhaus
Weight Instruments Computing DaqBoard/1000 ME4656
PCI-6014 PCI –
DAS6014
Meets 30% 2 2 2 2
Specifications
Additional 10% 2 2 2 2
capture
channels
available
Price 20% 1 2 1 0
Company 20% 2 1 1 1
Reputation
Documentation 20% 2 1 1 1
and support
Total 1.8 1.6 1.4 1.2
Meets Specifications:
0 – Meets very few of the specifications.
1 – Meets some of the specifications
2 – Meets all specifications
Additional capture channels available:
0 – No additional channels available
1 – Only one additional channel available
2 – Multiple additional channels available
Price:
0 – High priced models
1 – Moderately priced models
2 – Low or economy priced models
Company reputation:
0 – Poor reputation
1 – Neutral reputation
2 – Excellent reputation
Documentation and support:
0 – Poor documentation and support
1 – Average documentation and support
2 – Excellent documentation and support
- 27 -
Table 4: Decision Matrix for Industrial PC Selection
Weight C3320 C3340 4513KP 4512KP
Industrial Rating 20% 2 2 2 2
Appearance 10% 2 2 2 2
Processing Power 10% 2 2 1 1
Storage Capacity 10% 2 2 1 1
PCI Expandability 5% 2 2 1 1
Price 10% 2 2 1 1
Memory 5% 1 1 2 2
Display 10% 0 2 2 0
Company Reputation 20% 2 2 1 1
Total 1.75 1.95 1.45 1.25
Industrial rating:
0 – Not IP65 rated
2 – IP65 rated
Appearance:
0 – Not sturdy and industrial in appearance
2 - Sturdy and industrial in appearance
Processing power:
0 – Less than a Pentium 3
1 – Pentium 3 or equivalent
2 – Pentium 4 or equivalent
Storage Capacity:
0 – Less than 20 GB
1 – 20 – 39 GB
2 – 40 GB or more
PCI expandability:
0 – Less than 3 PCI slots
1 – 3 PCI slots
2 – More than 3 PCI slots
Price:
0 – More than $5000
1 - $4000 to $5000
2 – Less than $3000
Memory:
0 – Less than 256 MB
1 – 256 MB
2 – 512 MB or more
Display:
0 – 10” display
1 – 12” display
2 – 15” display
Company reputation:
0 – Poor reputation
2 – Excellent reputation
- 28 -
Table 5: Decision Matrix for User Peripherals
Weight Touchpad Touch Screen Keyboard/Mouse
Ease of use for 20% 1 2 2
Maintenance
Reliability 20% 2 2 2
Cost 30% 1 0 2
Programmability 10% 2 1 2
Ergonomics 20% 1 2 2
Totals 1.3 1.1 1.4
Ease of use for Maintenance:
0 – Difficult to use
1 – Average
2 – Very easy to use
Reliability:
0 – Nor reliable
1 – Somewhat reliable
2 – Very reliable
Cost:
0 – Large cost
1 – Affordable
2 – Very low cost
Programmability:
0 – Difficult to program
1 – Some programming may be required
2 – No programming required
Ergonomics:
0 – Very difficult to use
1 – Somewhat awkward to use
2 – Very comfortable to use
- 29 -
Appendix C: Complete Gantt Chart
- 30 -
ID Task Name Duration Start Finish ep '04 Oct '04 Nov '04 Dec '04 Jan '05 Feb '05 Mar '05 Apr '05 May '05
5 12 19 26 3 10 17 24 31 7 14 21 28 5 12 19 26 2 9 16 23 30 6 13 20 27 6 13 20 27 3 10 17 24 1 8 15 22
1 Planning Stage 58 days Mon 9/13/04 Wed 11/10/04
2 Receive Problem Statement 0 days Mon 9/13/04 Mon 9/13/04 9/13
3 Research company background 1 wk Mon 9/13/04 Sun 9/19/04
4 Refine Customer Need (meet with Grob) 0 days Fri 9/24/04 Fri 9/24/04 9/24
5 Develop Specifications 2 wks Mon 9/20/04 Sun 10/3/04
6 Meet With Project Manager 0 days Thu 9/30/04 Thu 9/30/04 9/30
7 Submit preliminary specification 0 days Wed 10/6/04 Wed 10/6/04 10/6
8 Research Availaible Components 2 wks Mon 10/4/04 Sun 10/17/04
9 Develop Writen Proposal 3 days Mon 11/1/04 Thu 11/4/04
10 Proposal to Lance for review 0 days Mon 11/1/04 Mon 11/1/04 11/1
11 Proposal to Vemuru for final proofread 0 days Thu 11/4/04 Thu 11/4/04 11/4
12 Writen Proposal Turned in 0 days Fri 11/5/04 Fri 11/5/04 11/5
13 Oral Presentation of Proposal - 4pm 0 days Wed 11/10/04 Wed 11/10/04 11/10
14
15 Hardware Summary 65 days Fri 11/5/04 Wed 2/2/05
16 Order Hardware Components 0 days Fri 11/5/04 Fri 11/5/04 11/5
17 Receive Hardware Components 0 days Mon 11/29/04 Mon 11/29/04 11/29
18 Assemble Hardware 1 day Mon 11/29/04 Mon 11/29/04
19 Identify need of I/O Card 3 days Tue 11/30/04 Thu 12/2/04
20 Plan additional need of I/O consideration 2 wks Fri 12/3/04 Thu 12/16/04
21 Build any needed I/O conditioning hardware 2 wks Fri 12/17/04 Sat 1/15/05
22 Integrate coditioning hardware with I/O card 2 wks Sun 1/16/05 Sat 1/29/05
23 Create Testing equipment/process 2 days Sun 1/30/05 Mon 1/31/05
24 Implement/Debug Drivers 2 days Tue 2/1/05 Wed 2/2/05
25
26 Software Summary 46.5 days Mon 11/29/04 Sun 1/30/05
27 Hardware Interface 2 days Mon 11/29/04 Tue 11/30/04
Task Milestone External Tasks
Project: grob-10-17-04 GANTT CHART Split Summary External Milestone
Date: Fri 11/5/04
Progress Project Summary Deadline
ID Task Name Duration Start Finish ep '04 Oct '04 Nov '04 Dec '04 Jan '05 Feb '05 Mar '05 Apr '05 May '05
5 12 19 26 3 10 17 24 31 7 14 21 28 5 12 19 26 2 9 16 23 30 6 13 20 27 6 13 20 27 3 10 17 24 1 8 15 22
28 Calibrate/Diagnostic 1 day Wed 12/1/04 Wed 12/1/04
29 Implement multiple I/O cards 1 wk Sun 1/16/05 Sun 1/23/05
30 Design User Interface 1.5 wks Mon 11/29/04 Thu 12/9/04
31 Configuration of Functionality 2 days Thu 12/9/04 Sat 12/11/04
32 Save Data 1 day Wed 12/1/04 Wed 12/1/04
33 Graph Data 3 days Sat 12/11/04 Tue 12/14/04
34 Archive Component 3 days Tue 12/14/04 Fri 12/17/04
35 Interface to archive data 2 wks Fri 12/17/04 Sun 1/16/05
36 Software Development Complete 0 days Sun 1/16/05 Sun 1/16/05 1/16
37 Analyze archive data 2 wks Sun 1/16/05 Sun 1/30/05
38
39 Testing Summary 28 days Sun 1/16/05 Sun 2/13/05
40 Create Test Plan 1 wk Sun 1/16/05 Sun 1/23/05
41 Testing hardware 1 wk Sun 1/23/05 Sun 1/30/05
42 Debugging software 1 wk Sun 1/23/05 Sun 1/30/05
43 Hardware and software integration testing 2 wks Sun 1/30/05 Sun 2/13/05
44
45 Documentation & Reporting 64.5 days Sun 2/13/05 Mon 5/9/05
46 Create Final Report 1 wk Sun 2/13/05 Tue 3/8/05
47 Create Web site 1 wk Tue 3/8/05 Tue 3/15/05
48 Create Poster for Presentation 1 wk Tue 3/15/05 Tue 3/22/05
49 ASEE Poster Competion at Ohio Northern 3 days Thu 4/7/05 Sat 4/9/05
50 Poster Presentation 6pm-8 0 days Thu 4/21/05 Thu 4/21/05 4/21
51 Final writen report due 0 days Fri 4/29/05 Fri 4/29/05 4/29
52 Final Oral Presentation 0 days Wed 5/4/05 Wed 5/4/05 5/4
53 Final Website due 0 days Mon 5/9/05 Mon 5/9/05 5/9
Task Milestone External Tasks
Project: grob-10-17-04 GANTT CHART Split Summary External Milestone
Date: Fri 11/5/04
Progress Project Summary Deadline
Appendix D: Team Member Resumes
- 33 -
