Senior Design 1

Project Documentation

The Smart Digital Voltmeter

Sponsored by Commercial Lighting Enterprises Incorporated

Department of Electrical Engineering and Computer Science

University of Central Florida
Dr. Lei Wei

December 6, 2016

Group 19

William Brumby Electrical Engineering [email protected]

Gaston Mulisanga Computer Engineering [email protected]
Travis Ram Computer Engineering [email protected]
Vladimir Tsarkov Electrical Engineering [email protected]
Table of Contents
Page
1. Executive Summary………………………………………………………………...1
2. Product Description………………………………………………………………....3
2.1 Motivation…………………………………………………………………...3
2.2 Goals and Objectives……………………………………………………...3
2.3 Requirements Specifications……………………………………………..4
2.4 Quality of House Analysis………………………………………………...6
Table 2.1: House of Quality Trade-Off Table…………………….…..6
3. Research……………………………………………………………………………..8
3.1 Technological Foundation………………………………………………...8
3.2 Existing Digital Voltmeter Projects and Products………………………8
Figure 3.1: T.K. Hareendran’s Project……...…………….…………10
Figure 3.2: Mayoogh Girish’s Project……………….....…………....10
3.2.1 Different Types of Voltmeters………………………………….11
3.3 Voltmeter Measurement Technologies………………………………....17
Figure 3.3: Selector switch (independent ranges).........................19
Figure 3.4: Selector switch (dependent ranges)............................19
3.4 Safety Hazards and Protective Measures……………………………...20
Figure 3.5: Good vs Bad Fuses………………....…………………..22
3.5 Strategic Components and Part Selections…………………………....26
3.5.1 Processors and Operation Principles………………………...27
3.5.2 Wireless Transmission………………………………………....31
Table 3.1: Short-Range Wireless Communication Comparison.....38
3.5.3 Smartphone Operating System…………………………….....41
3.5.4 Digital Display………………………………………………......43
3.5.5 Diodes and Power Supply……………………………………..45
3.5.6 Safety Components………………………………………….....50
3.5.7 Electrical Measurement Components………………………...50
Table 3.2: Resistor Divider Network
Values………….....…..52
Figure 3.6: Range Selector Switch……………....……….....53
3.5.8 Test Leads…………………………………………………….....55
3.6 Parts Selection Summary………………………………………………...56
Figure 3.7: Parts…………………….……………………………..…..59
4. Standards and Realistic Design Constraints………………………...................60
4.1 Standards……………………………………………………………….....60
4.1.1 Hardware Standards…………………………………………....60
Table 4.1: Transient test Values for measurement
categories..…..63
Figure 4.1: Bluetooth® Core Protocols………….…………………..67
Table 4.2: ROHS Restrictions………………….………………….....69
Table 4.3: ROHS Categories………………………….……………...70
4.1.2 Software Standards………………………………………….....70
 4.1.3 Impact of Standards…………………………………………....71
4.2 Realistic Hardware/Application Design Constraints……………….….71
4.2.1 Microcontroller Constraints…………………………………....72
Figure 4.2: Example of sampling frequency (voltage vs time).......73
Figure 4.3: Analog-to-Digital Converter………………..…………...74
4.2.2 Economic and Time Constraints………………………….......77
4.2.3 Environmental, Social, and Political Constraints…………....77
4.2.4 Ethical, Health, and Safety Constraints……………………...78
4.2.5 Manufacturability and Sustainability Constraints……………78
5. Hardware and Software Design Details………………………………………....80
5.1 Initial Design Architecture………………………………………………..80
5.2 Initial Voltmeter Design…………………………………………………..80
5.2.1 Breadboard Testing………………………………………….....81
5.2.2 Issues and Corrective Measures……………………………...81
5.3 Updated Hardware Design……………………………………………....83
Figure 5.1: Updated Design Schematic Layout………………..…..84
5.3.1 Discovered Issues……………………………………………...85
5.4 Final Hardware Design…………………………………………………..86
5.5 Application and Final Microcontroller Design………………………….91
5.5.1 Application…………………………………………………...….91
Figure 5.2: Final Software Design Tree……………….........93
5.5.2 Microcontroller………………………………………………......93
Figure 5.3: Conceptual Flow of the Program...………….....94
Figure 5.4: Code Design FlowChart………...……………....96
5.6 Summary of Application and Microcontroller Design
6. Prototype Construction and Schematics
6.1 Breadboard Prototype Construction Plan 4pgs
Figure 6.1: Prototype Construction - Step One
Figure 6.2: Prototype Construction - Step Two
Figure 6.3: Prototype Construction - Step Three
Figure 6.4: Prototype Construction - Step Four
Figure 6.5: Prototype Construction - Step Five
Figure 6.6: Prototype Construction - Final Schematic
6.2 PCB Vendor and Assembly 4pgs
7. Prototype Testing Plan
7.1 Hardware Test Environment 3pgs
7.2 Individual Component Testing 3pgs
7.2.1 Resistors, capacitors, test leads and switches
Table 7.1: Resistor Testing
Table 7.2: Capacitor Values
7.2.2 Safety Components
Figure 7.1: Varistor Testing Power Supply
7.2.3 Battery and voltage regulator
Figure 7.2: Power Supply Circuit
 7.2.4 Diodes
Figure 7.3: Diode Test Circuit
Figure 7.4: Diode Test Simulation
7.2.5 MCU and Crystal
Figure 7.5: Crystal Oscillator Connection
7.2.6 LCD and Bluetooth® Module
Figure 7.6: LCD and Bluetooth® Implementation
7.3 Breadboard and PCB Test Procedure
7.4 Software Test Environment 2pgs
7.5 Software Specific Testing 2pgs
8. Administrative Content
8.1 Milestone Discussion 4pgs
Table 8.1: Project Milestones
8.2 Budget and Finance Discussion 4pgs
8.3 Sponsor Information
Table 8.2: Bill of Materials
Table 8.3: Software Budget
9. Product Operation
9.1 Using the Smartphone Application
9.2 Taking Voltage Measurements
9.2.1 Setup
9.2.2 DC Measurement
9.2.3 AC Measurement
9.3 Troubleshooting
9.3.1 LEDs and Display
9.3.2 Incorrect Measurements
9.3.3 Application Issues
10. Project Summary
Appendix A - Works Cited
1 Executive Summary
A digital voltmeter is a tool which is used to measure either AC or DC voltage for
a large range, usually from millivolts all the way up to kilovolts, depending on the
product. Using two conductive probes attached to specified input jacks, the user
can complete the circuit within the voltmeter itself by placing the probes in the
desired location of the circuit under inspection. From the probes, an analog
signal is passed through various safety components and circuitry in order to feed
an analog-to-digital converter (ADC). This digital signal is now fed into a
processor which computes and displays the data via LED or LCD display,
depending on the model. The device generally has two input jacks, one for the
positive lead and one for the negative lead which is the common ground to the
supportive circuitry. The largest divide between modern digital voltmeters is
whether the device is manual or auto-ranging, i.e. whether the processor can
realize the range of value you are attempting to measure and select the correct
path the signal must follow to be read correctly. Parameters such as these are
detrimental in determining the direction in which our voltmeter design will
progress.

The goal of this project is to develop a lightweight and compact digital voltmeter
which wirelessly transmits data and measurements to an accompanied
smartphone application. The data to be collected and measured by the meter will
be both AC and DC voltage. The final product should provide an ease of use far
greater than the current voltmeter technology where the user can record values
to observe later. This voltmeter should be a low power device that can run for
many hours (around 200 hours) so when the the product is used in the field there
is no concern of battery life. The user will be able to control settings such as
sampling rates and measurement ranges via the application. The device will also
include a small traditional display for times when smartphone access is
inconvenient or unavailable. The smartphone application will allow users to view
measurements over time and automatically generate plots of the data as desired.

This product is designed for engineers, electricians, hobbyists, and general

household usage. Although most users prefer digital multimeters, devices that
can measure several different electrical and physical parameters in a single
package, some users only need to measure voltage in specific applications and
want a device with high level accuracy and precision. We will be fully designing
and creating a device that will precisely carry out these operations and present
the data in a method not previously done before. The advantage to our device is
the ability to record, access, and view these measurements via a smartphone
application. The Smart Digital Voltmeter will give users the ability to record the
data hands-free and save it automatically for later evaluation. Having the ability
to control precisely what you desire to measure for any range is the reason our
device, despite its limited functions, is far more beneficial to the average user.
The device will conform to all safety and communication standards and
constraints as defined in section 4. It should be noted that even though our
product will conform to the safety standards (Category rating), we do not have
the time nor money to obtain actual safety certifications.
2 Product Description
2.1 Motivation

As electrical and computer engineering students, our team has used voltmeters
on countless occasions for lab work, as general electronic hobbyists, or at our
real jobs. What we have all encountered and can all agree on when taking
measurements is the inconvenience in logging data. Traditionally, measurements
that are taken are able to be read by the user via the LCD or LED display on the
device itself. The downfall to this outdated design is recording this information,
due to the fact that when the probes are removed, the data disappears. Most
modern voltmeters have the ability to “hold” the value displayed so the user may
release the probes and record the information. While this is useful and partially
identifies the issue at hand, the main problem still lies with the inconvenience in
removing the probes, logging the information, and re-probing the circuit for
another measurement. For example, let us consider a circuit designer who
designs and constructs a low-pass filter and wants to test the output of his/her
circuit. Ideally, an oscilloscope would be the desired tool to use in this situation.
Unfortunately, these tools are very expensive, but the measurements can still be
taken with a digital multimeter. The user would set the function generator on the
input of the filter to a low frequency, probe the output, and read a voltage. The
user must then remove the probes, record the value, set the next frequency, and
re-probe the circuit. This can become very tedious when narrowing down specific
parameters such as cutoff frequency, maximum overshoot, etc.

Our goal is to create a product that eliminates this inconvenience, along with
aiding in the large scale ability to decipher the data that would be easily done by
an oscilloscope. This is where the idea of data being wirelessly transmitted to a
smartphone application was born.

2.2 Goals and Objectives

The goals for our product are very simple. We want to design a basic voltmeter
that can be used virtually anywhere with the ability to send and record data
measurements via wireless transmission to a smartphone application where the
data can be further analyzed. Functionality of the voltmeter will be limited to AC
and DC voltage for a range specified in section 3. The measured voltage can be
viewed on a digital display located on the meter itself, along with being wirelessly
transmitted to the accompanied smartphone application. Users will have many
options for recording and analyzing the transmitted data via this application.
Some of these options include sampling rates, data plotting layouts, and many
more settings. Once measurements have been taken, the user will be able to
automatically plot the recorded data inside the application. Being able to take
multiple measurements at a sampling rate determined by the user will prove to
be the most valuable aspect of our product. The reason being our cheap and
compact voltmeter can mirror the most basic function of a large and expensive
oscilloscope, which is seeing the characteristic response of circuitry in nearly real
time. This is done accomplished through a powerful processor and strategic
design architecture which can sample input voltage at very high speeds. If users
desire a single quick and accurate measurement without the need of the
application, the digital display will be programmed in a way that mirrors modern
voltmeters. Additionally, if the user wants to see other voltage parameters such
as maximum, minimum, and average values over time on the smartphone
application.

The PCB, electrical circuitry, and digital display will be contained within a small
container which can be clipped to a belt or stood upright on a flat surface. The
probes will be light and compact with comfortable grips and safety guards to
ensure the user does not accidentally make contact with the circuitry being
measured. Because we are designing an instrument to be potentially used with
high current or high voltage measurements, safety is of the utmost concern.
Physical high-voltage isolation slots will be embedded into the PCB, along with
numerous safety measures such as fuses, varistors, thermistors, and diodes to
aid in the protection of both the product itself and more importantly the user. Our
final design will utilize high quality products such as ceramic fuses (in
comparison to cheap glass fuses) and high power resistors to ensure the safety
of both product and user in cases of transient voltage spikes.

2.3 Requirements Specifications

The following are the key design requirements and specifications of our smart
voltmeter:
● Digital display that will display the current voltage being measured, updated at
a rate specified by the user via the smartphone application
● Wireless capability of connecting the voltmeter to a phone or computer in
order to record measured data and will comply with that wireless technologies
standard.
● Two probes to measure positive and negative terminal.
○ Button on the positive probe to record measured values.
● Voltmeter will be lightweight and compete with current voltmeters.
○ Will weigh under 300 grams.
● The device will be operated as similar as possible to current voltmeters to
retain user familiarity.
● Our voltmeter accuracy will be within 1% that of current voltmeters.
● The cost will be as low as possible to maintain competition with other digital
voltmeters that do not provide wireless data recording.
 ○ Finished project is expected to cost under $100.
● The device will be low power in order to maximize battery life, since it will be a
compact wireless data transmitter.
● Dimensions close to 8 x 4 x 2.5”.
● The Android application developed will operate in accordance with the
standards set by Google to be available on the Play Store.

Although creating a digital Voltmeter may sound simple in theory, there are many
requirements and specifications that must be met very carefully if this design
were to compete with existing meters. Our voltmeter, as its name suggests, will
provide accurate AC and DC voltage readings. In order to introduce a product, it
must first stand a fighting chance against the current standard voltmeters that
have been around for many years. Therefore, it must at least match the existing
models to have a chance at success. Our smart digital voltmeter will be required
to maintain a familiar feel to existing digital voltmeters. That is, a regular user of
today’s voltmeters should be able to switch over to our smart voltmeter with
relative ease. Our design and utility should appear familiar and very similar in its
functions as the average voltmeter. If our product will be too difficult, consumers
will hesitate to make the transition to the smart digital voltmeter. Furthermore, the
buttons, probes, and functions (AC and DC voltage) should all look and feel
familiar.

In addition to function and familiarity, the smart digital voltmeter will be specified
to weigh, cost, last, and be as accurate as the average voltmeter. The smart
digital voltmeter will be required to have dimensions no larger than 8” x 4” x 2.5”
and weigh no more than 300 grams. Any larger, and we run the risk of the
voltmeter losing its portability advantage. Next, the smart digital voltmeter must
be affordable and cost less or equal to the competition. Because our idea of
recording voltage readings is close to the luxury category, consumers are not
likely to invest their money into a luxury product if it is too expensive. Another key
specification of the smart digital voltmeter is its battery life. The voltmeter will
operate on battery power, and thus should have low power consumption in order
to last as long as possible. With regular use, the smart digital voltmeter batteries
should last approximately as long as the voltmeters from competitors (~36
hours). Last but not least, and one of the most important specifications of a
voltmeter is its accuracy. We expect our smart digital voltmeter to be within 1%
accuracy of the average handheld voltmeter.

Now this is where the smart digital voltmeter shines and has the edge over
today’s voltmeters. Wirelessly transmitted data recording of measurements is
what will set the smart digital voltmeter apart from the rest. Likely to be an
Android application, our software will allow the wireless transfer of data from the
voltmeter onto an app on a smartphone. The user can then observe changes in
voltage readings and share this data with others wirelessly. In other words, the
application will serve as a small portable oscilloscope that can instantly share
data and results with colleagues.

2.4 Quality of House Analysis

↑​ = Positive Correlation
↑↑​ = Strong Positive Correlation
↓​ = Negative Correlation
↓↓​ = Strong Negative Correlation

Marketing Engineering Requirements

Requirements

Low Power Accuracy Low Dimensions Lightweight

Cost

Accuracy ↓ ↑↑ ↓ ↓ ↓

Low Power ↑↑ ↑ ↑ ↑

Ease of Use ↑ ↑ ↑ ↑

Low Cost ↑ ↓ ↑↑ ↑↑ ↑↑

Battery life Within 1% of

at least 200 standard ≤$100 8 x 4 x 2.5” ≤300 grams
hours voltmeters
Table 2.1: House of Quality Trade-Off Table

This device has potential to be used in many fields, such as engineering,

people’s homes or any areas that need measurement relating to voltage, current,
and resistance. On top of that, the device should be portable, easy to carry, and
lightweight. Another vital factor, is that it will be used in remote areas where it
cannot be connected to a power outlet for long periods. Because of the reasons
mentioned above, the device will be designed such that it uses as less power as
possible. A low power device goes hand in hand with lightweight device and easy
of carrying around. Which will be very appealing to our target users.
Since this device will have multiple target users. We have taken into account that
not all users will have technical expertise in design or circuitry, therefore for us
designers, it is very important that the voltmeter should be easy to use for all
users. The more user friendly, the more acceptable it will be. This correlates with
the price of the device. It makes sense to buy a device that is easy to use for
everybody.
None of the things mentioned above are important if the device is not working.
Our first objective will be to make the device work and do what it is intended to
do, that is measure AC and DC voltage. Not only must it work but it should
display the output as accurately as possible, within 0.1% of standard voltmeters
in the market. This will give us a competitive edge over other voltmeter makers.
Overall our voltmeter will be a low cost device machine that gives accurate and
precise display of the measurements taken. To make it economical and practical,
it will be small, portable, and use low power consumption.
3 Research
Traditional analog voltmeters have been around since the 1920s, and the
conversion to digital technology emerged in the mid-1950s. That being said,
there is superfluous material to research to gain insight into the process of
designing and implementing a modern digital voltmeter. In this section we will
briefly discuss the background of the electrical technologies used in these
products, along with the pros and cons of existing products. We will then discuss
the strategy behind our parts selection and give a general idea on how our
product will be implemented using these components.

3.1 Technological Foundation

As mentioned above, the first multimeter, which could measure voltage,

resistance and current, was invented nearly a century ago. This device, called
the Avometer, operated using moving coils, precision voltage resistors, rheostats,
and switches and sockets to control range selection. It also incorporated a
“universal shunt” resistor to measure seven ranges of current. Later models
developed in the 1950s included the ability to measure AC current and voltage.
Analog multimeters such as this one fared well in voltage and current
measurements but had limited ability when measuring large resistances. Along
with electrical limitations, analog multimeters are generally bulky and non-mobile.
Like the rest of the world around this time, the conversion from analog to digital
also struck accord with products such as the multimeter. While not as accurate in
its early stages, the digital multimeter offered users a more compact and overall
usable product. Digital technology provided a platform for advanced functionality
in multimeters along with other improvements such as power consumption.
These advancements are the reason we decided to incorporate digital
technology in our device.

3.2 Existing Digital Voltmeter Projects and Products

In 2013, the engineering department of University of Petrosani in Romania

proposed an application that also added extra functions of a digital multimeter.
Their project was similar to ours in that it allowed a digital multimeter to connect
to a computer and gather information from the multimeter via an application. This
allowed for the user to observe and record data in real time. Data was recorded
in a chart and in a table with appropriate time markers as measurements were
performed. It is important to note that the key element of a digital multimeter is an
analog digital converter (ADC). This allows every value of voltage to be
represented by a string of binary code that is much more resistant to noise than
in analog multimeters. Therefore, digital multimeters are more accurate than the
analog counterparts. Although this engineering department of University of
Petrosani had the same goals as our smart digital voltmeter, rather than
designing their own multimeter, they simply used existing inexpensive models of
multimeters and wrote software to integrate a digital multimeter with a computer.

On June 4, 2015, the editorial team from All About Circuits provided a project on
how to make a simple digital voltmeter using an Arduino and 16x2 liquid crystal
display (LCD). This project uses the advantage of Arduino’s onboard analog to
digital converter as well as several analog input pins that connect the positive
and negative terminals of the probes. Arduino’s ADC is a 10-bit converter
resulting in 2​10 or 1024 output values ranging from 0 to 1023. By using Arduino’s
5V reference voltage, the team was able to calculate voltage present at the
analog input. Additionally, this project applied a voltage divider in order to
measure voltages above that of the reference voltage. The editorial team
connected the input voltage that is desired to be measured to the analog input
pin. The Arduino board then takes this analog value and multiplies it by the
reference voltage (in this case it is 5V), then divides the product by 1024 to
calculate the actual voltage value. After calculating the voltage, the measure
value is recorded on the LCD screen.

Because of the limiting factor of the Arduino voltmeter being able to measure
voltages of less than the 5V reference voltage, the editorial team from All About
Circuits applied a voltage divider. The input voltage needed to be divided so that
the voltage that actually went into the Arduino is 5V or less. For this project, the
team used a 90kΩ and a 10kΩ resistor to create a 10:1 ratio voltage divider. In
doing so, voltages up to 50V could be measured.

Another Arduino project by T.K. Hareendran created a digital voltmeter via

Arduino that allowed voltage readings between 0V to 30V. Because Arduino’s
analog inputs can be used to measure DC voltage between 0V to only 5V,
Hareendran used voltage divider to decrease the voltage being measured by the
Arduino input. Code in the Arduino software is then used to calculate the actual
voltage being measured by the hardware. The following is a schematic showing
the voltage divider in action:
 Figure 3.1: T.K. Hareendran’s Project.

As can be seen from the above schematic, the two resistors used for the voltage
divider are 100kΩ and 10kΩ. With these values, it is possible for the Arduino
board to read voltages from 0V to 55V. The voltage being fed into the analog pin
on the Arduino board is V​in [ (10k) / (100k +10k) ] which is approximately 0.09V​in​.
Because this is the input voltage divided by 11, this means that the reference
voltage of 5V multiplied by this coefficient of 11 yields 55. Therefore, 55V is the
upper bound of what this Arduino voltmeter can measure. However, according to
Hareendran, it is good practice to label a voltmeter such as this one a “0-30V
Digital Voltmeter” in order to add a “safety margin.”

Another significant portion of our project consists of Bluetooth® data

transmission and reception. On the “Do it Yourself Hacking” website
(DIYHacking.com), Mayoogh Girish created a simple project focusing specifically
on Arduino Bluetooth® Interfacing with basic Bluetooth® functionality. In Girish’s
project, he allowed the user to control a light emitting diode (LED) on the
Arduino board using a smartphone. The following flowchart shows how Girish’s
project functions:

Figure 3.2: Mayoogh Girish’s Project.

As can be seen from the above flowchart, this sample project can be broken
down into three main parts: An Android smartphone, a Bluetooth® transceiver,
and an Arduino board with an LED. The HC 05/06 Arduino Bluetooth® Module
that is used in this design work on serial communication. In other words, only one
bit can be transferred at a time. Based on this project, the Android app is
designed to send serial data to the Arduino Bluetooth® module when a button is
pressed on the app on the smartphone. The Arduino Bluetooth® module at the
other end receives the data and sends it to the Arduino through the TX (or
transmission) pin of the Bluetooth® module, which is connected to the RX (or
receiving) pin of the Arduino board. Then, the code written into Arduino checks
the received data. If the received data is 1 or HIGH, then the LED turns ON. If the
received data is 0 or LOW, then the LED turns OFF.

3.2.1 Different Types of Voltmeters

There are several different types of voltmeters available that we will discuss and
they each have a subcategory of DC Voltmeter, AC Voltmeter, or two in one.

1) Permanent Magnet Moving Coil (PMMC) Voltmeter - This type of Voltmeter is

an analog voltmeter that uses two
stationary magnets to generate a
magnetic field with a wire coil in
between the two magnets. These
magnets are typically made up of
alcomax and alnico materials in
order to provide high field strength
and increase reading accuracy of
the voltmeter. The wire coil moves
freely between the two permanent
magnets shown at the left. This
coil is wound in many turns of
copper wire and is then placed
onto a rectangular piece of
aluminum which is connected with bearings to allow it to rotate. The higher the
voltage that is being measured, the larger the current will be that is passing
between the two meter terminal connections, and as a result, the higher the
current, then the higher the voltage the needle will point to on the scale.

There are three main disadvantages of this type of Voltmeter. First, it is only
capable of measuring DC voltage. If we apply AC current to a PMMC, the
direction of current will be reversed half the time and as a result the torque will
also be reversed, yielding an average torque of zero for the entire AC clock
cycle. A second disadvantage of PMMC voltmeters, can be errors of the
permanent magnets. Over time and due to temperature effects, magnets may
lose their magnetism, resulting in inaccuracies of voltage readings. Third,
temperature affects the resistance of the moving coil. Normally, the temperature
coefficient of copper wire in the moving coil is 0.04 per degree celsius rise in
temperature. Due to lower value of temperature coefficient, the temperature rises
at a faster rate and causes the resistance to increase. A significant amount of
error is caused due to this.

As for good news, PMMC has its advantages in power consumption. This type of
device is very low power and can last a very long time. Additionally, PMMC
voltmeters make it very easy to measure quantities and the scale is very easy to
read and is known for its high accuracy.

2) Moving Iron (MI) Voltmeter - Moving Iron Voltmeters consist of two different
types, repulsion type and attraction type instruments. These types of instruments
are similar to the PMMC voltmeter due to the fact that they are all analog
voltmeters. However, both MI Voltmeters have an advantage over the PMMC
voltmeter because they can measure both DC and AC voltage.

● Repulsion Type MI - The basic idea behind the repulsion type MI Voltmeter is
to use two separate iron strips in parallel with similar magnetic properties
(same north and south poles). Due to the similar magnetic properties, the two
plates will repel each other. This repulsion force depends on the strength of
the electromagnetic field in the coil which surrounds the two plates. The iron
strips will keep their repulsion independent of the current direction of the coil
because the magnets will remain similar in polarity. Because of this, AC
readings using this device are available. Next, one plate is placed in a fixed
position so that it cannot move, while the other plate is connected at one end.
In other words, both poles of strip 1 are fixed, while only one pole of strip 2 is
fixed. This allows repulsion to take effect and for the unconnected pole end of
strip 2 to move farther away from strip 1, proportional to the electromagnetic
field in the coil and thus the current through the coil. Lastly, a pointer is
connected to the free end of the iron strip 2, and a scale is located near this
pointer to indicate the voltage reading based on the current running through
the coil. This is possible because the force of repulsion is proportional to the
current, which is proportional to the angle displacement of the pointer.
Repulsion type MI instruments are advantageous for being cheap, having low
friction errors, and having the ability to measure both AC and DC voltage.

● Attraction Type MI - Similar to the repulsion type MI, an electromagnet is

used. The two ends of the coil represent the two probes or reading terminals
for the voltmeter and are attached to a source. A current then flows through
the coil and turns the coil into an electromagnet. A plate of iron is then placed
in the magnetic field produced by the coil. Naturally, the iron plate will be
attracted by the flux of the coil and move towards the coil a distance
proportional to the strength of the magnetic field of the coil, which is also
proportional to the amount of current flowing through the coil. Next, a pointer
is attached to the iron plate and as the plate moves so does the pointer. A
 scale can then be fixed near the pointer to show voltage readings because
the position of the pointer is proportional to the current in the circuit.

3) Electrodynamometer Type Voltmeter - Electrodynamometer type voltmeters

were created because they have the same calibration for both AC and DC
readings. This type of voltmeter is calibrated with DC, allowing us to measure AC
without requiring additional calibration steps.

Let us focus on the structure of the electrodynamometer type voltmeter. This type
of voltmeter consists of two coils, one moving coil and one fixed coil. This is
similar to the idea in the repulsion type MI voltmeter, but coils are used instead of
iron plates. The moving coil will be used to move the pointer with the help of a
spring. A spring is used to minimize errors caused by the earth’s gravity. The
current flowing through the moving coil will be limited to avoid heating. To do this,
a very high resistor is placed in series with the moving coil. The spring will be
placed on a pivoted spindle, allowing free movement of the pointer to indicate the
correct value on a scale. This electrodynamometer type voltmeter has a uniform
scale because the moving coil moves linearly over a range of 40 to 50 degrees.
In this type of voltmeter, the moving coil acts and is referred to as a pressure coil.
This is because the moving coil is connected across the voltage and the current
flowing through the pressure coil is always proportional to the voltage.

As for the fixed coil, it is divided into two equal parts which are connected in
series with the load. Consequently, the load current will flow through these coils.
Dividing the fixed coil into two parts is done so that the voltmeter can carry a
considerable amount of electric current and measure more values than just that
in the micro- or milli-amp range. This gives the fixed coils the name of the current
coils of electrodynamometer type voltmeters. Initially, these fixed coils were
designed in order to carry current of approximately 100 amperes. However,
today’s voltmeters are designed in order to carry current of approximately 20
amperes. This is done to save power and accommodate the need for mobile,
compact, and long-lasting voltmeters.

Electrodynamometer type voltmeters have the advantage of using a uniform

scale. However, the scale does contain a limit for measurement. This type of
voltmeter is also a great choice because it can be used to measure both DC as
well as AC voltages due to the fact that the scale is calibrated to measure record
both values. Conversely, electrodynamometer type voltmeters have certain flaws.
This instrument may suffer from errors in the pressure coil inductance, errors due
to pressure coil capacitance, mutual inductance effects, errors resulting from
improper pressure coil and current coil connections, errors due to Eddy currents,
errors caused by vibration of a moving system, errors resulting from temperature
changes or high temperature, and errors due to stray magnetic field. Magnetic
fields outside of this electrodynamometer type voltmeter circuit can skew the
results that the pointer and scale represent.

4) Rectifier Type Voltmeter - Rectifier type voltmeters, as their name suggests,

implement a rectifier. A rectifier is basically a collection of diodes that allow
current to pass only one way. Diodes provide zero resistance if it is forward
biased (or if the current flows in the favored direction) and very high resistance if
it is reverse biased. This important property of diodes is used to rectify voltages.
The term rectify in electronics means converting an AC signal into a DC signal.
This allows a rectifier type voltmeter to measure DC and AC voltages.

The rectifier type voltmeter is simply a rectifier used in conjunction with another
voltmeter such as a permanent magnet moving coil voltmeter (PMMC). A rectifier
makes up for the disadvantage of the PMMC not being able to measure AC
voltage. When a DC source is connected to the rectifier type voltmeter, the
diodes act as an open gate and allow the voltage to pass and be read by the
voltmeter. As for the AC source case, the rectifier type voltmeter with a PMMC
will also work and yield AC voltage readings. The diodes from the rectifier will
limit the negative voltage or
current flowing from the
source, while allowing the
forward biased half of the
voltage or current to pass. As
a result, the new voltmeter
will not yield an average
torque of zero for the entire
AC clock cycle because only
the positive half will pass
through the rectifier and
provide the voltage reading.
The figure on the left shows
AC voltage without a rectifier
(top) versus AC voltage after
it passes through a rectifier
(bottom).

In addition, the rectifier limits the current drawn by the PMMC. This is important
to achieve because if the current exceeds the current rating of the PMMC, it may
damage it.

Rectifier type voltmeters are some of the most widely used instruments in the
industrial world simply because of its cost. The cost of electrodynamometer type
voltmeters is considerably high compared to rectifier type voltmeters, while the
accuracy is basically the same. Furthermore, rectifier type voltmeters are not as
delicate as others such as thermocouple instruments. Some additional
advantages of rectifier type voltmeters include, having a frequency range of
operation that can be extended to a high value, containing a uniform scale on the
meter, and having low operating value of current and voltages.

5) Induction Type Voltmeter - An induction type voltmeter has two fluxes that are
produced by two different alternating currents on a metallic disc. These two
fluxes alternate and create an induced electromagnetic field. The
electromagnetic field produced at one point then interacts with the alternating
current of the other side to generate torque. The metallic disc levitates above the
current coil due to Eddy current, and spins as a result of the electromagnetic field
interacting with the alternating current. This spin is proportional to the amount of
current and voltage of the load, and thus can be used in order to obtain a
measurement reading of potential difference between two ports.

Induction type voltmeters have several advantages. They are inexpensive

compared to moving iron type instruments, have high torque to weight ratio
compared to other types of voltmeters, and they remain accurate over a wide
range of temperatures and loads.

6) Electrostatic Type Voltmeter - Electrostatic type voltmeters use a static

electrical field in order to produce a deflecting torque. These kinds of voltmeters
are typically used in order to measure high voltages, but they can be used to
measure lower voltages as well. There are two possible options of how the
electrostatic force can act. First, two plates are oppositely charged in order to
create an attractive force between them. One of these plates is fixed, while the
other plate is free to move and is attached with a pointer. The free-to-move plate
is attached to a spring to keep it in parallel with the first plate. Because of the
attractive force between the plates, the free-to-move plate will move towards the
stationary plate until the moving plate stores its maximum electrostatic energy. In
the second option, there may be a force of attraction or repulsion (perhaps even
both) between the two plates. When the two probes are connected to a desired
element to be measured, the DC or AC voltage that we are interested in
measuring generates a torque between the two plates. This repelling force
rotates the movable plate, which in turn shifts the attached pointer along a
calibrated scale. This is a very simple type of voltmeter.

Let us discuss the advantages and disadvantages of electrostatic type

voltmeters. The first and most important advantage of electrostatic type
voltmeters is their capability of measuring both DC and AC voltage. The reason
for this is that deflecting torque is directly proportional to the square of the
voltage. Second, the power consumption of this device is low because the
current drawn by this type of instrument is very low. Another important property
and advantage of this voltmeter is its ability to measure very high values of
voltage. Electrostatic type voltmeters have a few disadvantages. This type of
instrument is quite expensive compared to other kinds of voltmeters and they are
also larger in size than its counterparts. As a result, these kinds of voltmeters are
mainly used in companies that require to measure large voltage readings and do
not have much domestic use because the average homeowner or user of
voltmeters does not have a need to measure absurdly large voltages. Additional
disadvantages of electrostatic type voltmeters include not having a uniform scale
and containing various operating forces that are small in magnitude.

7) Digital Voltmeter (DVM) - The digital voltmeter is different from all the previous
voltmeters discussed because it is the only voltmeter that obtains a
measurement digitally rather than an analog reading. In other words, a digital
multimeter directly indicates a measurement value rather than relying on
deflection. Digital voltmeters are excellent instruments because they completely
eliminate error due to parallax. Parallax is when a user attempts to read the
pointer on a scale. With digital voltmeters, any two users will not have two results
that differ by more than 0.01%. Similar to analog and digital clocks, digital
voltmeters have a high-speed reading advantage. The user simply takes one
quick look and instantly reads the recorded measurement on the digital voltmeter
rather than looking at a pointer and deciding which marker it is pointing to the
closest. Additionally, high speed readings of digital voltmeters can be stored in
memory for further analysis. This speeds up the recording process because none
of the analog voltmeters have this capability. Digital voltmeters have other
advantages such as being versatile and accurate, compact and cheap, having
low power requirements, increased portability, and the ability to measure both AC
and DC voltages.

For the purpose of this project, the team decided on choosing a digital voltmeter
design rather than any of the above analog voltmeters. Voltmeter input
impedance is the biggest difference between analog and digital voltmeters.
Because most digital voltmeters have 50 times more impedance than analog
voltmeters, digital meters are more accurate when measuring voltage in high
resistance circuits. In section 3.3, we will discuss the significant impact that the
input impedance of a voltmeter can have on the accuracy of your measurements.
It is important to select a digital voltmeter for our design to minimize this “loading
effect.” The loading effect a voltmeter has on a circuit is determined by the total
resistance of the measured circuit in relation to the impedance of the voltmeter.
The higher the input impedance of our voltmeter when compared to the circuit
that we are desiring to measure, the more accurate our measurements will be.
As a result, to maximize the accuracy of our readings, we are encouraged to
build a digital voltmeter instead of an analog design. Another disadvantage of the
analog voltmeters is that they are generally more bulky than the digital meters.
This disadvantage results from the required long needle, scale, and coil that must
be placed into an analog meter. Because the group is interested in creating a
more portable and compact voltmeter, we again choose the digital voltmeter over
analog due to this convenient and lightweight quality. Last but not least, one of
the most significant perks of using a digital voltmeter over an analog version is
the digital display. The display of a digital voltmeter is very easy to read because
it shows one simple value after setting the voltmeter to the correct range. In an
analog meter display, there is a scale with a pointer which shows the measured
value. Not only is it difficult to discern between the many different markings on
the scale, but the scale does not provide accurate voltage readings to several
decimal points like the digital voltmeter does. In addition, analog meter displays
are difficult to read because the user is required to close one eye and try to
determine the most accurate marker that the pointer is aiming to. After
considering the many differences between analog and digital voltmeters in
regards to convenience, portability, and accuracy, we decided to proceed with a
digital voltmeter design.

3.3 Voltmeter Measurement Technologies

In this section, we will discuss how a common digital voltmeter works. An ideal
voltmeter is a two terminal device that measures the voltage (or potential
difference) between the two terminals. Ideally, a voltmeter has an infinitely large
resistance. As a result, no current would be lost from the measured circuit, and
only the voltage will be computed because the two terminals would be in parallel.
A fundamental rule of circuit theory is that any two elements in parallel have the
same voltage. The design of the voltmeter abuses this law by connecting a very
high resistance in parallel with a circuit element, thus showing the voltage
recorded at that part of the circuit. Let us observe a
simple example to better understand the process
that voltmeters go through. On the left, we have a
thevenin equivalent circuit with a Voltage of 10V
and Resistance R​1 that can correspond to any
circuit that we are interested in measuring. In this
example, we are interested in measuring the
voltage across resistor R​1​. To do this, we connect
our voltmeter in parallel with resistor R​1​. Resistor R​2 symbolizes the high input
impedance of our voltmeter. Initially, the voltage across R​1 is 10V and the current
passing through R​1 is 0.01 Amps when R​1 is assumed to be 1kΩ. If the input
resistance of our voltmeter is approximately 10MΩ, which is a fairly popular value
for input impedance for a voltmeter, then we can calculate the expected current
draw from the circuit by our voltmeter. Since our voltmeter is connected in
parallel with resistor R​1 then our input impedance voltage should also equal 10V.
Therefore, the current flowing into our voltmeter is approximately 10V/10MΩ,
which then gives us 1µA (or one microamp). This value is 0.01% of the original
current that flowed through R​1 prior to connecting our voltmeter. Therefore,
connecting a voltmeter in parallel with a circuit element or multiple elements will
draw negligible amounts of power and will have virtually no effect on the circuit,
while at the same time providing an accurate reading of the potential difference
that we are interested in measuring.

Similarly, the circuit can be analyzed and compared prior to and after the
connection of the voltmeter to the circuit. When two resistors or impedances are
connected in parallel, their total resistance or impedance can be calculated by
the following formula:

In the example above, R​1 is equal to 1kΩ and R​2 is 10MΩ. Applying the above
formula for R​total we obtain approximately 999.9Ω. Again, it is proven that
connecting a voltmeter to a circuit leaves the circuit almost completely
unaffected.

Voltmeters have the advantage over ammeters because they do not require the
user to alter or break the circuit in order to obtain a measurement. The power
loss in a voltmeter is V​2 / R​V where V is the voltage that needs to be measured
and R​V is the resistance of the voltmeter. As a result, voltmeters should have a
high electrical resistance in order to draw a very small amount of current and
consequently consume a small amount of power from the circuit.

Analog inputs of a voltmeter usually have a limit of how much voltage can pass
into the port. For example, an Arduino board has a default analog reference of
5V. As a result, Arduino can only measure voltages between 0V and 5V by
default. These values can be increased by applying voltage divider and reducing
the amount of voltage that is inputted to the Arduino analog port. Additionally, the
reference voltage can be altered to give us more or less resolution. Let us
observe a similar example but with an analog voltmeter that applies voltage
divider to increase the range of measurement:
 Figure 3.3: Selector switch (independent ranges).

In the diagram above, the voltmeter circuit contains a range selector switch from
the positive probe. This range selector allows the user to choose which
measurement scale he or she wants to apply. Like in the above case, the user
can select between 1V, 10V, 100V, and 1000V ranges. The higher the range that
the user wants to measure, the higher the resistance is required in order to limit
the voltage seen by the analog input of the meter or analog to digital converter
(Digital Voltmeter case). However, because resistor values such as 999.5kΩ,
99.5kΩ, 9.5kΩ, and 500Ω are difficult to find, more common resistors are used,
and added in series such as the figure below:

Figure 3.4: Selector switch (dependent ranges).

The difference between the two circuits is that in the prior circuit, only one
resistor was used for each range. Conversely, in the latter circuit, multiple
resistors could be used in series and the higher the range, the more resistors
were used. This is simply done for convenience because more common resistors
are implemented in the circuit. Resistors with values 900kΩ, 90kΩ, 9kΩ, and
500Ω are much more likely to be located in the average resistor bin. However, no
noticeable difference will be seen from the user’s perspective whether the user
decides to implement the first circuit or the second circuit.

3.4 Safety Hazards and Protective Measures

Before selecting the safety measurements for our meter, we must identify all of
the possible hazardous phenomenon that could occur when using our product.
The most obvious threat are high voltage transients. According to Fluke’s “ABCs
of Multimeter Safety,” voltage spikes can be generated from motors, capacitors,
and power conversion equipment such as variable speed drives. Additionally,
lightning strikes on substations, power plants, and most commonly transmission
lines can also cause high voltage transients. These transients can even occur on
low-voltage power circuits and can reach values as high as many thousands of
volts. In a case like this, the voltage rating alone is not enough information to tell
the user how well the measurement device was designed to not break from these
high transient impulses. This is where the user depends on safety margins
already built into the voltmeter. Examples of safety measures and components
will be further discussed in this section.

Significant improvements in measurement devices such as the voltmeter were

made in order to combat transients. For instance, the nominal bus voltage of
electric commuter railroads was normally read at 600 Volts. Voltmeters rated at
1000 Volts were easily able to measure this nominal bus voltage. However, these
meters would break after only a few minutes of taking measurements while the
train was operating. After studying this mystery, a closer look revealed that the
train starting and stopping generated 10,000 Volt spikes! As a result, voltmeter
input protection needed to improve immediately.

The most dangerous and most injury-causing event from a transient is an arc
blast. An arc blast is more severe than an electric shock and causes more
electrical injuries every year. Arc blasts occur when transients ride on
high-energy circuits because these circuits can deliver very large currents. If a
transient causes an arc-over, the high current can sustain the arc and produce a
plasma breakdown. A plasma breakdown is an electric explosion that results
when the surrounding air becomes ionized and conductive. The way to combat
this dangerous issue is making sure that the measurement device with the
correct overvoltage category is used. Section 5.5.2 will discuss the four different
measurement categories in greater detail.

One device used for overload protection in voltmeters is fuses. A fuse is a very
thin wire that melts or vaporizes when current passing through it exceeds a
particular fuse rating. Fuses are available with current ratings from 1/500 Amps
to hundreds of amps. The thin wire can made from aluminum, tin-coated copper,
or nickel. Most fuses in electronic equipment are cylindrical in shape and are
glass or ceramic with a metal cap at each end. The current rating and voltage are
written on one of these two metal end caps. Fuses are placed near the start of
the circuit on a measurement device such as a voltmeter in order to monitor the
current going into the meter. If the current entering the probes of the voltmeter is
too high for the circuitry inside the meter to handle, the fuse will melt and prevent
further damage to the device. Although fuses are not very necessary because
the high input impedance of a voltmeter already limits most of the current that
can pass into the meter, fuses serve as an extra protection against overcurrent.
Overcurrent can cause electric components to overheat and cause a fire. It is
beneficial to incorporate a fuse into the voltmeter design because if worst case
scenario occurs, it is cheaper to replace a fuse rather than purchase a new
meter. Adding a fuse into a voltmeter also helps count for the user making a
mistake and connecting the meter’s probes where they should not be touching. If
lightning were to strike on equipment with a fuse, the user may open the device
and, upon examining the parts, see that only the fuse was damaged rather than
the power supply or other components.

There are two most common sizes for fuses: 1¼ by ¼ inch, and 5 by 20
millimeters. The first size is mainly used for automobiles. Both sizes are used for
electronic equipment, however the latter is more popular for electronics circuit
design. Electronics tends to prioritize small sized components in order to be able
to design small, low-power, and compact devices.

There are two basic types of fuses: fast acting and slow blow type. First, the fast
acting fuse, like its name suggests, opens very quickly when its particular current
rating is exceeded. This type of fuse is used for analog meters because their
input impedance is not as high as that of digital meters. As a result, if too much
current flows through an analog meter for even a very small amount of time, its
circuitry components can be destroyed. Second, the slow blow fuse has a coiled
construction inside the glass. Slow blow fuses are designed in such a way as to
open only when there is a continued overload such as a short circuit. The coiled
structure in the fuse helps stop the fuse from blowing on just a temporary current
surge. Below is an image that shows a visual comparison between a good fuse,
a slightly burnt fuse, and a damaged fuse.
 Figure 3.5: Good vs Bad Fuses.

Another method of testing a fuse instead of the visual inspection, is connecting

an ohmmeter to both ends of the fuse. A good fuse will either show an ohm
reading, or will read zero ohms. Contrarily, a blown fuse is technically an open.
Therefore, a blown fuse will not show any reading on the meter.

A second possible preventative measure for input circuitry protection is varistors.

Similar to how a fuse offers over-current protection, the varistor provides
over-voltage protection. Varistors do this by utilizing voltage clamping, similar to
a zener diode. That is, a varistor will trap extra voltage if the input voltage
becomes too high, thus protecting the delicate circuit components of the
voltmeter. The word varistor itself is an acronym, combining the words “variable”
and “resistor.” This means that the resistance of a varistor is dependant on the
amount of voltage flowing through it because the varistor changes its resistance
value automatically with the change in voltage across it. In fact, a varistor is also
known as a VDR, or voltage dependant nonlinear resistor. Varistors are made
from semiconductor material and have symmetrical voltage and current
characteristics, allowing them to be applicable to both AC and DC voltage
implementation. The main function of varistors in input circuit protection is to
combat transient surges. If lighting strikes near or onto a circuit which is being
measured by a voltmeter, transients will travel through the circuit and can reach
voltages up to several thousand volts.
As a preventative measure, varistors
are placed in parallel with the delicate
circuitry inside of the voltmeter (a
sample circuit can be seen to the left).
By doing this, under normal
measurement readings, the varistors
act as high impedances and force most
of that important current into the circuit
of the meter. However, in the case of a
transient surge, the high input voltage results in the varistors behaving as short
circuits, thus forcing most of this current to travel to ground instead of damaging
the significant and delicate electronic circuits and components.

There are two main types of varistors: the silicon carbide varistor (SiC), and the
metal oxide varistor (MOV). Both types of varistors begin conducting current at a
specific voltage and stop conduction when the voltage falls below a threshold
voltage. However, the MOV has two significant advantages over the SiC varistor.
First, the leakage current through the MOV’s zinc oxide material is very small at
normal operating conditions compared to the SiC varistor. In other words, metal
oxide varistors allow the meter’s circuitry to be more accurate and appear nearly
nonexistent. The second advantage of metal oxide varistors is their speed of
operation. MOV’s clamp transients much faster than SiC varistors do. Metal
oxide varistors are available in a wide range of varistor voltages ranging from
approximately 10 volts to over 1,000 volts AC or DC. It is also important to note
that the maximum continuous rms voltage rating of a varistor should be just
above the highest expected supply voltage. For example, if the highest
measurement that a voltmeter can record is 120 volts, then the rms voltage rating
should be around 130 volts rms. Therefore, if an input voltage is greater than 130
volts, then the varistor will kick in and create a short to protect the circuitry of the
meter from frying.

A third possible preventative measure for input circuitry protection is the zener
diode. A zener diode also acts like a voltage clamp, similar to the varistor.
However, a reverse biased Si small-signal diode breaks down at approximately
100 volts. This is much lower than a varistor. A zener diode is connected in
reverse bias because the breakdown voltage of a zener diode in reverse bias is
significantly higher than that of a zener diode connected in forward bias.
Additionally, once a zener diode breaks down, it usually does so permanently.
The zener diode breaks and acts as a short circuit with zero voltage across the
diode. They can no longer return to their normal working state like varistors.
Therefore, zener diodes may require additional maintenance compared to
varistors. Another difference between a zener diode and a varistor is that a
varistor is connected in parallel to an input voltage, while a zener diode connects
in series to the input voltage. As a result, if a zener diode were to break down
during the event of a transient surge, then the bulk of the transient voltage will go
into the delicate circuitry and further damage the voltmeter.

A fourth possible preventative measure for input circuitry protection is a

thermistor. A thermistor is a combination of the words “thermal” and “resistor.”
Therefore, it can be deduced that a thermistor is simply a temperature sensitive
resistor. Although all resistors are slightly affected by temperature, their
temperature coefficient is quite minimal compared to the special high coefficient
in thermistors. Thermistors can have negative temperature coefficients (NTC) or
positive temperature coefficients (PTC). In NTC thermistors, the resistance of the
thermistor decreases as the temperature increases. NTC type thermistors are
generally used when a change in resistance over a wide temperature range is
required. They are used more often for temperature sensors for the range -55
degrees celsius to 200 degrees celsius. NTC is most popular for having a quick
response, being reliable, robust, and have a low price. Conversely, PTC
thermistors experience an increase in resistance as temperature increases. PTC
type thermistors are prefered when a sudden change in resistance at a certain
temperature is required. Positive temperature coefficient thermistors have a type
of temperature controlled switch that allows for its resistance to instantly
increase. The most common switching temperatures for PTC thermistors are in
the range of 60-120 degrees celsius. They are most commonly used in
self-regulating heating elements and self-resetting overcurrent protection. PTC
thermistors are a good choice for input protection in a voltmeter circuit design
because it can help lower the input current if it reaches a dangerously high value.
Thermistors are ceramic semiconductors and are most commonly composed of
metal oxides.

A fifth possible preventative measure for input circuitry protection is the

wirewound resistor. A wirewound resistor is a passive electrical component that
limits current. It is basically a resistor that has a wire with a high resistivity
wrapped around an insulating core in order to provide resistance. This is the
oldest type of resistor that is still manufactured today. Wire wound resistor values
can be very precise because its resistance value is dependent on the resistivity
of the wire, the cross section, and the length, all of which can be accurately
controlled by the designer. However, wire wound resistors can influence the
current flow in an alternating current circuit, negatively impacting the results. This
is because of the natural capacitance and inductance of the wire wound resistor.
As a result, wire wound resistors are not ideal for high frequency operation within
a circuit.

There are three types of wirewound resistors: precision, power, and

potentiometer. In a precision wirewound resistor, the accuracy of resistance is
very good, staying within plus or minus 0.05% of the design value for a particular
circuit application. Power wirewound resistors are designed for very high power
applications. The range of this type of resistor is between 0.5 watts until over
1000 watts. The function of power wirewound resistors as an input protection
measure is to help dissipate heat. The third type of wire wound resistor is the
potentiometer wirewound resistor. Wire wound resistors can function as
potentiometers because of their durable construction. A potentiometer is
basically a resistor with three terminals. Two terminals are the same as a regular
resistor, but the third terminal is attached to a movable contact that varies the
amount of resistance.
Wire wound resistors are often used in circuit breakers or as fuses. In order to
function as fuses and breakers, a small spring is attached to one end of the
resistor using solder to hold the spring in place. If the current passing through the
resistor becomes too high, the solder will melt and the spring will pop up and
open the circuit. This will serve as circuit protection for the vulnerable voltmeter
components by stopping dangerously high current from passing through fragile
components and damaging them.

PCB safety is another concern for this project. As a result, PCBs incorporate
isolation slots to help separate nodes with high voltages. This is an additional
safety measure to help account for arc-overs. Arc over is a case where voltage is
so high that it passes through other even non-conductive materials and cause a
short in the circuit. Arc-overs can also cause carbonization or burning on the
PCB and this would be undesirable because a permanent short would be formed.
This would be irreversible damage and would require a new PCB to be rebuilt.
Arc-overs can also occur over air, so there is an option to include a
high-dielectric strength shield onto the PCB to counter this issue. Another
advantage of PCB isolation slots is that dust cannot accumulate on the empty
space. On a PCB with no isolation slots, dust can form on the board surface over
time and reduce its dielectric strength.

Because no tool by itself can guarantee safety, there are several safety practices
that the user should implement whenever he or she operates a voltmeter for
maximum protection. First, the user should work on de-energized circuits
whenever possible. However, if the user is not certain, then he or she should
always assume that the circuit is live and dangerous. On live circuits, it is crucial
to wear the correct protective gear. The user should use insulated tools, wear
safety glasses or even a face shield, wear insulated gloves, remove watches or
other jewelry, stand on an insulated mat, and wear flame resistant clothing
instead of regular work clothes. Next, the user must take careful precautions
when making measurements on live circuits. He or she should make it a habit to
hook on the ground clip first, then make contact with the positive probe. When
removing the probes, first disconnect the red or positive probe and then
disconnect the ground probe last. For additional safety, the user can minimize his
or her contact with the voltmeter by hanging or resting the meter on some object
or surface instead of holding it in their hands. This will minimize personal
exposure to the effects of transients. Another safety precaution is using the
three-point test method to make certain that a circuit is dead or has no power. In
the three-point test method, the user should first use the meter to test a known
live circuit. Second, the user will test the target circuit. Third, the user will test the
known live circuit again. This method will verify that the meter functioned
correctly before and after the measurement. Last but not least, a little old
electricians’ trick of keeping one hand in your pocket can be used. This technique
decreases the chance of a closed circuit to form across the chest and through
the heart.

The last line of defense for users against transients or high voltages is the human
body. The approximate body resistance under the skin from hand to hand across
the body is 1000 Ω. At approximately 30 mA, respiratory paralysis occurs, thus
causing a fatal situation for an average size human. Using Ohm’s law, this
means that only a 30 V voltage would be enough to cause death for a human.
Fortunately, the body contains another powerful natural resistor, skin. The outer
layer of dead skin cells protects the body and can help resist voltages up to 600
V from being fatal. However, this is assuming that the skin is not already
punctured, meaning the user has no cuts and scrapes, and the skin is not
covered in any conductive material such as water. A good voltmeter will eliminate
any transient dangers before these transient currents ever make contact with the
human. The safest meters contain double insulation, recessed input jacks and
test leads with shrouded input connectors, non-slippery test leads with finger
guards, and be made of high quality and durable non-conductive materials.

In the following section, we will discuss which safety components our team
decided to include in the smart digital voltmeter design and the thought process
behind why we chose these specific parts. Additionally, the standards section
4.1.1 will discuss the four different CAT ratings in greater detail and will expand
on the uses of each category.

3.5 Strategic Components and Part Selections

In this section we will describe the various parts that we have researched and
taken into consideration as part of our design. These include electrical
components, software platforms, and packaging options. We will then discuss the
pros and cons of each candidate component and come to a final conclusion on
their use in our product. In section 3.6 we will give a brief summary on our parts
selection.

When considering the various approaches to designing our meter, it is easy to

see that there are a many different paths we can take. To begin our parts
selection, it is necessary to define everything that the final product will entail. We
are designing a digital meter that will measure a voltage when two conductive
probes are connected together through the circuitry under inspection. This meter
will read an AC or DC voltage, translate this voltage for user readability via some
form of computation, and simultaneously display the value on a digital display
while sending the information via wireless transmission to a smartphone
application. In the application, the user will be able to read the measure voltage
and adjust various settings such as sampling rates and data plot parameters.
From this basic explanation of our product, it is easy to see our first area of
interest needs to be how our voltage will be read and computed.

3.5.1 Processors and Operation Principles

There are several components that can be integrated into our product for the
purpose of computing measurements and giving us the tools to create a
functional meter. The first option we considered was using a microcontroller as
the brain behind our project. A microcontroller, or MCU for short, is a small
computer on a single integrated circuit. The MCU contains a Central Processing
Unit, or CPU, some form of memory, and programmable input and output
peripherals. As stated above, our product requires voltage to be read via wired
probes into a component which can be programmed to compute these values
and translate them into a readable display. With this definition, it is easy to see
that a microcontroller would we a great option for our product.

Modern digital voltmeters have been optimized to run in the most efficient fashion
using specifically designed processors to handle various operations. Most of
these use what are known as chipsets. A chipset is a system of electronic
components in an integrated circuit that handles the processing of data and flow
of this data between the processor, memory, and peripherals. These integrated
circuits are usually found on the motherboard of electronic platforms such as
gaming consoles, home computers, appliances and many other hardware. They
are designed to perform specific tasks in the most efficient manner. Companies
such as Fluke design all of their multimeters to work with specific chipsets. These
chipsets are developed by numerous companies and quite often designed per
customer specifications. It is evident that this technology could be largely
beneficial to our product and a viable option for our processing requirements.

In deciding between using a microcontroller or a chipset, we needed to consider

the other design features of our product from our requirements specifications,
along with the standards and constraints listed in section 4. The first and
foremost constraint is the essentially the bread and butter of our product; the
wireless transmission. There are multiple options for wireless transmission of
data that will be further discussed in the next subsection (3.4.2). Before selecting
which path to go down in that aspect, we had to consider which option,
microcontroller or chipset, would be most compatible with wireless capabilities.
Microcontroller platforms such as the Arduino, MSP430, and Raspberry Pi are
largely known for their ability to easily interface with a multitude of peripherals
and accessories. For example, both Arduino and MSP430 have shields which
are produced and sold by various companies that have all of the components
and hardware necessary for Bluetooth®, WiFi, NFC, and GSM wireless
transmission. These shields are small PCBs that have been designed specifically
for these applications and are relatively simple to incorporate with the
microcontroller. Additionally, these microcontrollers have been around long
enough that users have created various software libraries that make the coding
process much simpler. This means that regardless of the path we choose for
wireless transmission, using a microcontroller would allow our hardware and
software design time to be very minimal.

For a pre-designed chipset, on the topic of wireless transmission, we realized

immediately that we would have to design and build our own module. Regardless
of the transmission type, we would need to purchase all of the hardware
necessary for that type of transmission. We were quick to realize the impact this
would have on our design time, cost, and the amount of extra space this
hardware will take up on our PCB. Even though the cons heavily outweigh the
pros in regards to wireless transmission, it is not an impossible task. This then
led us to think of design time would be saved by using a chipset when
considering the other hardware components such as input protection, voltage
division, and display modules. As mentioned earlier, companies produce these
chipsets specifically for voltmeter use which is the main selling point to
companies like Fluke that produce voltmeters. This allows the company
designing the meters to shorten their design time from 1-2 years down to 1-2
months! The companies manufacturing the chipsets have accompanying
datasheets that have diagrams and demonstration board schematics for
suggested use, with all of the components and their values already listed. If we
were to use one of these chipsets, we could potentially cut our design time for
the remaining circuitry by more than half. So even though we would need to
spend more time designing the wireless capabilities, we would in turn save a lot
of time on designing the voltage measurement and display components.

At this point in the selection process, the pros and cons of either choice were
very evenly weighted. In regards to size constraints, both options were well within
our requirement specifications. In terms of cost, both products with the required
additional components were roughly the same, and well within our budget. We
then did quick research and analysis comparisons on the power consumption
and battery options based on data we could find in available datasheets for both
options. For the various microcontrollers, all have generous power supply
options, being that the platforms are designed with pre-installed voltage
regulators. This means that we could power our microcontroller with voltages
ranging from 1.8V all the way up to 20V, depending on the microcontroller. For
us, this means that we can choose almost any power supply depending on our
battery life specification to power our meter. On the other hand, chipsets are
designed to work at very specific voltage levels, having various pins that require
different voltage as is the case in most embedded system applications. This
means that our design must be specific to whichever chipset we choose, and
even though listed in the datasheets, requires many more components for
reliable and accurate operation. In this sense, we determined microcontroller
would be the best choice, but we still had more specifications to take into
consideration.

The next important constraint we needed to consider was the electrical

characteristics of each option and how they relate to our design specifications,
mainly the accuracy of our meter in comparison to other products. Fluke is known
as one of the highest quality multimeter producers in the world. Their products
have been known to be as accurate as can be when it comes to digital
measurement. We therefore chose to reference our meter with a Fluke 17b
Digital Multimeter. Although this meter does more than voltage, the chipset used
should perform all of the necessary operations for our product. The chipset this
meter uses is a 4000 counts auto-range DMM IC, model FS9721_LP3 made by
Fortune Semiconductor Corporation. This 4000 counts is the resolution of the
ADC used in the chipset which can produce 3 and ¾ digits. While this accuracy
level depends largely on the external electrical components, we can achieve this
level of accuracy using the recommended demonstration design. If we were to
use a microcontroller, we would be writing all of the code and designing the
external hardware ourselves. This would mean our design needs to be optimal in
order to achieve the accuracy levels chipsets like these produce. Fortunately, we
can utilize additional components such as trim potentiometers to calibrate our
meter within the required accuracy in the final design. The downside to this is the
added PCB space and required user maintenance.

After considering all of the aforementioned options when choosing which

processing path we will go down, it seemed to be neck and neck between the
two possibilities. The last and arguably most crucial component to deciding
between the two was total project implementation. Modern voltmeters are
designed by experienced companies through thousands of man hours in all
aspects of engineering, not just electrical. For our final product to be functional,
we need to have a working system in which mechanical components and
electrical components work together in unison. One of the chipset factors we
have not yet considered are the mechanical requirements for the IC to operate
correctly. The demonstration board schematics listed in all of the available
chipset options show various switches in order to select different modes and user
options such as range selection, hold functions, and other switching operations. If
we were to use one of these chipsets and mirror our design on the recommended
layouts, we would need to incorporate all of these switches and design our
package to work with them properly. If we were to stray from the recommended
use, we would need to figure out how to implement these changes without
affecting the function of the IC. Regardless, the mechanical design of our product
to use this chip effectively would account for a large amount of time dedication.
Unfortunately, due to our largest project constriction which is time, we realized
that we do not have the man hours nor the resources to complete this project on
time using a chipset. On the other hand, microcontrollers are made for a wide
range of user design options through several available I/O ports. This largely
increases our switching options and design possibilities if our initial design ideas
were not able to be realized. Even though using a chipset looked to be the more
efficient option at first, mechanical switching implementation was the deciding
factor for our group between the two options.

Once we decided that we would be using a microcontroller as the brain behind

our meter, we had to consider the various options available and which devices
could realize our product needs. The first microcontroller we considered using
was the Arduino. Arduino itself is an open source electronics platform which
allows users to create projects by making the pins of the main microcontroller IC
easily configurable. This means that the input and output pins of the IC are
routed on a single PCB to large ports in which users can quickly place and
remove jumper wires without the need to solder. The pins on the IC used to
control power, clock signals, and switching are taken care of through external
components on the PCB. There are multiple versions of the Arduino with different
parameters such as the microcontroller being used, operating/input voltages,
clock speeds, memory, and input/output capabilities. Other microcontrollers such
as the families of MSP430 and the Raspberry Pi have platforms similar to
Arduino, differing mostly in the processors and the programming language. While
we could do an extensive analysis on the various MSP430 and Raspberry Pi
options, we did some quick research and decided to eliminate these options for a
few reasons.

Our first concern with these platforms is our own familiarity. Due to time
constraints, we had to consider what programming skills we already have in
order to minimize design time. While we have some experience using the
MSP430, none of us had ever used the Raspberry Pi. This microcontroller is
programmed to run with Python IDE, rather than C/C++ languages used on the
MSP340 and Arduino. The second factor in deciding between the MSP430 and
Arduino was available platform options. Arduino currently has 22 different
platforms with several different processors, layouts, and many other
characteristics to choose from. This means that we have several model options
to sort through and select which platform would best suit our needs. Similarly, the
MSP430 has over 500 microcontrollers available for purchase. In this aspect,
regardless of which we choose, both brands have sufficient options to choose
from in order to optimize our design. The final and deciding factor to take into
consideration was available resources. While the MSP430 and Arduino have
both developed user-friendly kits since around 2006, the Arduino has gained
much more popularity due to the IDE and programming simplicity. Because the
Arduino uses a very easy to use, high-level language and has numerous
available libraries, users have found Arduino to be the first choice in personal
project development. This translates to more resources for our group to access
when we inevitably run into issues with our design. After taking careful
consideration into all of the options, we decided as a group to use an Arduino
microcontroller platform.

For our project, we are not allowed to use a development board that the Arduino
platforms are built upon. Fortunately, Arduino gives a step-by-step procedure for
building most, if not all of the external components that power and control the
microcontroller. However, this procedure and build requires components that are
specific to the ATmega328P microcontroller. Before jumping the gun and
purchasing the required components, we must first review all of the additional
components so that we know what specifically our microcontroller needs to do. If
the ATmega328P is unable to meet our needs, we will need to consider other
microcontroller options and thus design the supportive circuitry ourselves. In the
following subsections, we will discuss the general design and layout of our meter,
and then we will walk through the strategic selection process for the remaining
components. Once we have a solid design, we will be able to determine if either
of the microcontrollers can be used for our meter. Sections 3.4.2 through 3.4.5
will outline our strategic selection process for the external hardware that our final
product will use. In section 3.5 we will review a few possible design architectures
to help narrow our microcontroller selection which will be finalized and
summarized section 3.6.

3.5.2 Wireless Transmission

Wireless communications began a new era in 1901, when M. G. Marconi

successfully established a radio link in Morse Code between a land-based station
and a tugboat (from Cornwall, England to St. John’s, Canada). Since this grand
discovery, rapid improvements have been made in the field of wireless
communication, allowing mankind to live in a new advanced digital age. In the
beginning of digital wireless communication, its use was restricted to the military
due to strategic requirements. Following Marconi’s wireless communication
model, the military’s wireless systems consisted of a base station with a powerful
transmitter that covered a certain geographical area. However, each of these
base stations were independent of one another, not allowing the bases to
communicate with each other. Conversely, today’s engineering advances in
digital communications improved wireless communication worldwide. Not only
are these base stations now connected and able to communicate with each
other, but the cellular systems of today contain low-power transmitters. Low
power resulted in low price, and low price allowed this technology to flow into the
public and civilian hands.

Wireless communication serves to be one of technology’s greatest contributions

to mankind. Wireless communication is the transmission of information over a
distance without the need for wires, cables, or other electrical conductors.
Transmission distance of wireless communication can range from a few meters
(like in the case of a television’s remote control) to thousands of kilometers (like
in the case of radio communication).

Wireless communication has several important advantages. First, wireless

communication was improved to transfer information very quickly to consumers.
Compared to wires, where the length of the wire creates a delay in signal
transmission and reception. Second, the wireless capability allows users to
access the Internet or specific devices anywhere and anytime without carrying
cables or wires with them to wherever they go. In addition, this helps users save
time by not having to travel to a certain location to transfer data. For instance,
doctors and other professionals at their location of employment can all be in
touch through wireless communication. A doctor can be instantly notified if he or
she is required to perform an emergency surgery. Another advantage of wireless
communication networks is that they are cheaper to install and maintain than
wired networks. Wired networks require long wires to be buried deep
underground, and a considerable amount of wiring is to be used in order to
connect many buildings such as houses in a grid.

The main disadvantage of digital wireless communication networks is their easier

susceptibility to hackers. It is very easy for hackers to intercept wireless signals
that are spread through the air. Consequently, it is very important to secure
private wireless networks so that shared information cannot be exploited by
unauthorized users. Security measures such as WEP, WPA, and WPA2 are
applied in many of today’s wireless networks.

WEP, or Wired Equivalent Privacy, is the most widely used Wi-Fi security
algorithm in the world. It was even formally validated as a Wi-Fi security standard
in September of 1999. WEP security initially provided 64-bit encryption and was
later increased to 128-bit after US restrictions on the export of various
cryptographic technology (64-bits) was lifted. Although 256-bit WEP encryption
was later introduced, the 128-bit WEP encryption remained one of the most
common implementations. However, even though the encryption algorithm
increased its number of bits, it was not enough to keep up with the even faster
increase in computer processor computation speeds. As computing power
increased, it became easier and easier to exploit WEP encryption. To increase
the public’s awareness of how weak WEP encryption was, in 2005 the FBI gave
a public demonstration where they cracked WEP passwords in minutes all while
using freely available software! Therefore, despite the attempts to improve the
WEP system, it still remains very vulnerable and systems that rely on WEP are
encouraged to be upgraded or replaced. The Wi-Fi Alliance officially retired WEP
in 2004.

WPA, or Wi-Fi Protected Access, was the Wi-Fi Alliance’s replacement to the
now vulnerable WEP. It was formally adopted in 2003. The most common WPA
configuration is WPA-PSK, or Pre-Shared Key. Compared to the 64-bit and
128-bit keys used in the WEP system, WPA used 256-bit keys. One significant
difference with the WPA was implementing message integrity checks. These
checks determined if a hacker had captured or altered packets passed between
access point and client. Another important upgrade was the Temporal Key
Integrity Protocol (TKIP). TKIP used a per-packet key system that was much
more secure than the fixed key used in the WEP system. Advanced Encryption
Standard (AES) later replaced TKIP. Although WPA was a direct upgrade to
WEP, the TKIP of the WPA system was designed to be easily rolled out via
firmware upgrades onto WEP-enabled devices. As a result, these WPA systems
with TKIP were exploited because they reused certain elements from the WEP
system.

WPA2 is simply called Wi-Fi Protected Access II. One of the most significant
changes between WPA and WPA2 was the use of AES algorithms and CCMP, or
Counter Cipher Mode with Block Chaining Message Authentication Code
Protocol. This CCMP replaced TKIP, which was a weakness of the older WPA.
The main security vulnerability of the WPA2 system requires the hacker to
already have access to the secured Wi-Fi network. While connected to the
network, the hacker gains access to certain keys and then attacks other devices
on the same network. Therefore, home networks or other small networks are
very safe with WPA2 security because it is easier to keep track of everyone that
is connected to the network. Large businesses, on the other hand, be aware of
this security weakness of WPA2. Wi-Fi Protected Setup (WPS) is used to
facilitate the linking of devices to modern access points. This WPS is part of both
WPA and WPA2, and remains as a vulnerability and potential access point for
hackers. Although, it may take anywhere from two to fourteen hours of sustained
effort with a modern computer for an attacker to breach a WPA or WPA2 security
with WPS enabled. As a result, it is recommended for large businesses and
companies with many different users accessing the same Wi-Fi to disable WPS
so that virtually all major threats of hackers are removed.

The following are some possible types of short-range wireless communication:

1) Infrared (IR) Wireless Communication - This type of wireless communication

applies infrared (IR) radiation. Infrared is electromagnetic energy at a wavelength
higher than that of red light. Generally, IR wireless communication is used for
shorter range communications as well as security control. In order for IR
communication to work, the transmitter and receiver should both be located
nearby and in line of sight of one another. In other words, there should be no
obstruction between the two. Infrared is often used in television remote controls
and security systems. On the electromagnetic spectrum, infrared radiation is
located between microwaves and visible light. As a result, infrared is able to be
used as a source of communication.
2) Radio Frequency (RF) Wireless Communication - Radio frequency is a
wireless electromagnetic signal that is mainly used as a form of communication
in wireless electronics. Radio waves are a type of electromagnetic radiation with
frequencies ranging from 3Hz to 300GHz. Frequency is basically the rate of
oscillation of the radio waves, so the higher the frequency, the more cycles it
undergoes per second. Radio frequency propagation is unique because it occurs
at the speed of light and, as a result, they do not require a medium such as air in
order to travel. Radio frequency waves even occur naturally from sun flares,
lightning, and from stars in space. We humans communicate with artificially
created radio waves that oscillate at various needed frequencies between 3Hz
and 300GHz. Radio frequency communication is used in many of today’s
industries such as television broadcasting, radar systems, computer networks,
and remote control.

Radio frequency measurement can be divided into three major general

categories: spectral analysis, vector analysis, and network analysis. First,
spectrum analyzers provide basic measurement capabilities of radio frequencies.
They are the most popular type of radio frequency instrument in many
general-purpose applications. The main advantage of spectrum analysis is being
able to view power vs frequency information. This will show what percentage of
power is consumed at which frequencies. Spectrum analyzers also have the
ability to demodulate analog formats via amplitude modulation (AM), frequency
modulation (FM), and phase modulation (PM).

The second category of radio frequency measurement is vector instruments.

These are vector or real-time signal analyzers and generators. Vector
instruments analyze and generate broadband waveforms. They are able to
capture time, frequency, phase, and power information from a signal. These
types of instruments are much more powerful than spectrum analyzers and
provide excellent modulation control and signal analysis.

The third category of radio frequency measurement is network analyzers.

Network analyzers are mainly used in order to make S-parameter measurements
and other characterization measurements on radio frequencies or high-frequency
components. Network analyzers are able to compare both the generation and
analysis on multiple channels. However, network analyzers perform at a much
higher price than spectrum analyzers and vector signal generators and
analyzers.

Radio frequency instruments tend to operate in the very high frequency ranges.
An added bonus resulting from high frequency ranges is immunity to some forms
of noise and impairments as well as the size of the antenna that is required to
receive such radio frequencies. The size of the antenna is usually ¼ the
wavelength.

3) Wi-Fi Wireless Communication - WiFi stands for Wireless Fidelity and is

synonymous with WLAN, or Wireless Local Area Network. WiFi, like most of the
other wireless communications, is based off of radio frequencies. However, the
radio frequencies of WiFi are completely different from walkie talkies, car radios,
and cell phones because WiFi operates in the Gigahertz range of frequency,
similar to microwaves. Car radios, for example, receive signals in Kilohertz and
Megahertz frequency range, which are AM and FM stations respectively. More
specifically, WiFi frequency is between 2.4GHz and 5GHz, while microwave
operates at 2.45GHz. This is why some people may experience interference with
their WiFi signal or in their internet connection when the microwave is cooking.

WiFi has many advantages. For example, the high frequency that WiFi operates
at allows the signal to carry more data. This makes it an excellent choice for
networks that transfer large files such as videos and online gaming. Data
transmitted at 5GHz can move up to 54 megabits of data per second thanks to
orthogonal frequency-division multiplexing (OFDM). This is a more efficient
coding technique that splits a radio signal into several sub-signals before they
reach a receiver. An added bonus of OFDM is that it greatly reduces
interference.

Wi-Fi wireless communication does have a disadvantage. Wi-Fi places all of its
eggs into one basket because the networks relies on a strong router. In other
words, it is also called a star network. If something were to happen to the router,
then the entire network is disabled as a result. If the router fails, or if too many
users connect to the router and use high-bandwidth applications at the same
time, then the users can experience interference or lose their connections.

4) Bluetooth® Wireless Communication - Bluetooth® technology is designed to

provide reliable, low-power, wireless communications over short distances. The
original version of Bluetooth® used Gaussian frequency shift keying (GFSK) in
order to transmit and receive data. GFSK is a form of modulation that is used in a
variety of digital radio communications systems. It is able to carry digital
modulation while still using the spectrum efficiently. This type of wireless
communication is an improvement from other forms of phase shift keying
because in the other forms the sidebands extend outwards from the main carrier,
resulting in an interference to other radio communications systems in the same
or nearby frequencies. Therefore, two different Bluetooth® devices could
interfere with one another resulting in noise. In a later version, Bluetooth®
introduced the Enhanced Data Rate (EDR) feature. This feature applied phase
shift keying (PSK) modulation techniques, which allowed for data rates that were
up to three times faster than GFSK. The main reason how multiple Bluetooth®
devices avoid noise interference from each other (although they are in the same
frequency band) is through a frequency-hopping technique called adaptive
frequency-hopping spread spectrum (AFH). Thanks to this special type of
spectrum, Bluetooth® devices avoid interference, or noise, by quickly switching
between 79 evenly spaced frequency channels from 2402 MHz to 2480 MHz.
This hopping occurs at 1600 times every second, so any data lost due to
interference is re-sent later over a different channel.

Although Bluetooth® is known mainly for its connections with headsets and
microphones, Bluetooth® wireless communication can be an excellent choice for
many other products. Benefits of Bluetooth® include, quick and easy wireless
connectivity as well as lower power consumption and greater portability in
comparison with with Wi-Fi.

5) ZigBee Wireless Communication - ZigBee is a wireless language that allows

everyday devices to connect to one another. The main advantage of ZigBee
devices is that they can even transmit data over long distances by creating a
mesh network of multiple ZigBee devices. Mesh network is unique from a tree
network or a star network because those networks require to have a manager
device or a priority device through which all data must pass. For example, Wi-Fi
is a star network because all information must pass through a router in order to
reach the computers connected to the Wi-Fi network. Therefore, if something
were to happen to the router and disable it, the devices on the same Wi-Fi
network would no longer be connected to the internet. By contrast, in a mesh
network any of the devices on the network can be connected to one another and
can pass information between and through devices. The figure to the left best
illustrates how devices can be connected and transmit data in a mesh network.
All circles symbolize a device in a mesh network, in addition to the lamp, the
switch, and stereo system. Initially, the black circles were also devices in the
mesh network, however, this
illustration shows how ZigBee
behaves under a situation in
which two devices were
damaged or ceased to work
for some reason. Because of
ZigBee unique mesh
networking, the switch can
take multiple paths to connect
and communicate with the
lamp, for instance. Before the
two circles were damaged,
the quickest path from switch
to lamp led through one of the
black circles. After these two
black circles no longer functioned correctly, the mesh network provided an
alternate path to retain communication between the switch and lamp.

ZigBee is mainly used for low data rate applications that need long battery life as
well as a secure network. Additionally, ZigBee has a defined data transfer rate of
250 kbit/s, which is excellent and more than enough for data transmission from a
sensor or an input device such as a voltmeter. ZigBee operates on 2.4 GHz
frequency which is available for unlicensed use anywhere around the world.
Consequently, a product developer can sell the same product using ZigBee
anywhere on the planet. ZigBee is a popular choice for smart homes, connected
lighting, and the utility industry. In smart homes, ZigBee provides home
automation (controlling doors, windows, garage), home security systems, and
appliances such as refrigerator, oven, microwave, etc. For connected lighting,
ZigBee is used in residential as well as commercial lighting applications. ZigBee
allows users to control LEDs, light bulbs, fixtures, remote controls, and switches.
In the utility industry, ZigBee is used in products that monitor, control, inform, and
automate the delivery and use of energy and water. Its main use is smart meters
because they have a low data rate and use nearly no power to maintain. ZigBee
controlled smart meters allow users to monitor their electricity and water usage.

6) Near Field Communication (NFC) - NFC is a short-range means of wireless

communication that applies magnetic field induction to connect two devices. RFC
is evolved from radio frequency identification (RFID) technology and operates as
part of a wireless link. Communication between devices is allowed when they are
touched together or brought within just a few centimeters of each other. NFC is a
peer-to-peer (P2P) wireless communication system, meaning it only allows for
two devices to connect and communicate by sharing data. To make up for the
short distance constraint, NFC may be accompanied by communication
technology such as Bluetooth® or Wi-Fi, which can be used for longer range
communication and the transfer of larger amounts of data. There are several
uses for NFC that will be covered. First, the user is able to use NFC to take
pictures with a cell phone (with a built-in camera) and touch an enabled
computer, television, or other cell phone to transmit the images for shared
display. Second, the user can download applications or games to a handheld
device or cell phone by touching a computer. Third, NFC can be combined with
another longer range wireless technology to transfer large files between two
devices such as a laptop and a desktop by bumping or touching the two together.
NFC has no need for an internet connection, is easy to use, and works fast.
Another significant advantage of NFC is that it does not require a pairing code to
link up two devices because instead it uses chips that run on very low amounts of
power. Therefore, it is much more power-efficient than other wireless
communication types. A common example of NFC chips is credit cards. As for
security, it is difficult for hackers to intercept data between two devices using
NFC because the devices are so close together and the attacker will also be
required to be located centimeters away from your device. The quick transfer
speeds of NFC are also advantageous to protect against hackers, giving them a
short window to intercept private data.

Now, each of these six short-range wireless communications will be compared to

each other in an effort to select a wireless communication system that will best fit
our design and recommendations for the smart digital voltmeter. The table below
will summarize the key advantages and disadvantages of each wireless
communication.

Table 3.1: Short-Range Wireless Communication Comparison.

First, infrared wireless communication will be discussed. Although infrared is an

effective means of short range communication, the dealbreaker was the
constraint of infrared technology having to remain in line of sight of connected
devices. This is a big concern for our digital voltmeter because we may want to
constantly record data in order to create a graph of voltage readings versus time.
This will be difficult with infrared technology because the recording device (in this
case an android smartphone) is required to maintain a constant visual connection
with the voltmeter and must remain in a line of sight. The user may accidentally
drop or move their smartphone, resulting in a loss of connection and
consequently will be required to restart the voltage recording from the beginning.
Another issue with infrared technology is its inability to function through objects
and walls. If the user is measuring and recording data from the digital voltmeter
and attempts to pass this information to their smartphone, this connection will be
lost if an object or someone were to walk through the line of sight link between
the voltmeter and smartphone. Consequently, if a plot of voltage versus time is in
the process of being recorded, data will be lost and the procedure will need to be
restarted. Therefore, this type of short-range wireless communication is not
suitable for our smart digital voltmeter design.

The next wireless communication in consideration will be Wifi. Wireless fidelity

communication offers a considerable 32 meter range for connecting devices,
which is over three times longer than most other wireless communications on our
list. Wifi also provides the fastest data transfer rate available from our list of
wireless communications, clocking in at approximately 250 megabits per second,
compared to infrared data transfer of 115.2 kilobits per second. However, the
main downfall of choosing Wifi for our smart digital voltmeter is the constraint of
Wifi hotspots. In other words, our device would only work if we could connect a
smartphone and the digital voltmeter to the same Wifi network. This would create
a big problem in our design because then our instrument would only be able to
transfer data to a smartphone if and only if both of these devices would be
connected to the same Wifi network or hotspot. In addition, many Wifi networks
are password protected, making it increasingly difficult to find an available
hotspot to choose from in order to transfer data wirelessly through Wifi from the
digital voltmeter to a smartphone. Another big disadvantage of selecting Wifi, is
the length of time required to connect two devices via Wifi. Both devices would
be required to input a password to connect to a password protected network and
then establish a connection between the smartphone and voltmeter. Because our
team wanted the smart digital voltmeter to operate at its full capability regardless
of its location, the Wifi wireless communication option had to be rejected in favor
of more convenient means of wireless data transfer.

The third wireless network that the group considered was Near Field
Communication (NFC). Although establishing an NFC communication between
two devices is very fast and easy, the main disadvantage of this type of
connection is that the two devices need to be in close proximity of each other.
This is a problem when it comes to connecting the digital voltmeter to a
smartphone because we would need to keep the smartphone a few centimeters
away from the voltmeter to maintain a stable connection between the two
devices. Otherwise we may face similar problems to the infrared wireless
communication. If the user shifts the smartphone slightly farther away from
voltmeter, the wireless connection between the two devices may be disrupted.
The connection would have two possible designs to combat this issue: 1)
Continuously hold the smartphone steadily and very close to the voltmeter, and
2) Abandon NFC and adopt a wired connection such as USB. In option #1, this
will cause great discomfort because generally both hands are already in use to
hold the probes and attaching them to the desired terminals to be measured.
This would be inconvenient for the user. In option #2, applying a USB cable
would go against the entire project idea and design of the smart digital voltmeter
with wireless capabilities. In conclusion, NFC wireless technology would not be
an optimal candidate for the smart digital voltmeter.

The fourth wireless communication that the group considered was ZigBee.
ZigBee shines when it is used to connect hundreds of devices together and still
made affordable due to the chip’s much lower power consumption. This type of
wireless communication specializes in connecting a very large number of devices
together, such as street lights, and transfer the data between each device as well
as all the combined data to a main device that the user can use to read all
information. Normally the range of ZigBee devices is under 50 meters, but this
range can be extended due to ZigBee special mesh networks, which allows for
many different paths of data transfer between multiple devices on the same
network. This is unlike Wifi networks where all devices must be connected to a
master device (the wireless router) in order to function. Although ZigBee is an
ideal means of communication for large networks, our smart digital voltmeter is a
single device connected to a single smartphone. Therefore, ZigBee would not be
used to its fullest potential if applied to our personal peer-to-peer (P2P) network.
However, possibly in a future project, ZigBee can be used in order for a professor
in a laboratory to view and grade students’ accuracy and results by connecting all
voltmeters in the room together and conveniently observing their progress from
the professor’s computer. Nevertheless, for this project, ZigBee is not the most
optimal option for wireless communications.

Last but not least, is Bluetooth® wireless communication. This option seemed to
have the least disadvantages when compared to how the team desired for the
smart digital voltmeter to function. For our particular project, Bluetooth® is an
excellent choice because of its versatility. Versatility, meaning that the
Bluetooth® module in the digital voltmeter will allow for our device to connect
wirelessly to any android smartphone in a very short amount of time. Setting up a
Bluetooth® connection between two devices is so simple that it may even take
only one simple push of a button and the user can begin utilizing the wireless
functions of the smart voltmeter. In addition, Bluetooth® is also completely
automatic, meaning that each Bluetooth®-enabled device can sense the
presence of another device within a physical range. A common network protocol
allows these devices to do this and share data. Bluetooth® devices can also
remember other Bluetooth® devices in case they are separated out of each
other’s range. This allows the devices to recover their connection in a matter of
seconds. Another advantage about this network protocol is that it reduces
interference from other Bluetooth® devices that also exchange data nearby. The
network protocol uses adaptive frequency-hopping spread spectrum (AFH). This
allows Bluetooth® to limit noise interference between multiple devices because
they quickly switch between 79 evenly spaced frequency channels from 2402
MHz to 2480 MHz. These switches happen 1600 times each second, so that any
data lost due to interference is re-sent later over a different channel. One of the
most important advantages of Bluetooth® is its compact quality. Bluetooth®
chips are very small, and more importantly, their short range function only
requires a tiny amount of radio power to work, which is less than a thousandth of
a Watt. As a result, Bluetooth® is the ideal short range wireless communication
for mobile battery operated devices such as smartphones and pocket voltmeters.
Therefore, Bluetooth® is the best option for short range wireless data
communication to use in conjunction with our digital voltmeter.

Because we will be using Arduino based circuitry/microcontroller, the group

decided to go with the Solu JY-MCU HC-06 Slave Bluetooth® Serial Port
Transceiver Baseboard Mini Module // Arduino Wireless Bluetooth® Transceiver
Module Slave 4Pin Serial. This is a simple four-pin Bluetooth® module that is
specifically meant to work with the microcontroller of our digital voltmeter. It has a
relatively low cost for a Bluetooth® module at $7.39 and is small (4.1 x 2.8 x 0.2
inches) and lightweight (0.3 ounces), which is excellent for our project weight
specifications.

3.5.3 Smartphone Operating System

The Smartphone market is saturated with a vast amount of options for

developing, each with it’s own advantages which makes considering these viable
operating systems a requirement in order to create the most effective and as
widely usable as possible. These different operating systems vary drastically
between resource allocation, development environment, data fragmentation, etc.
User outreach is also very important when selecting an operating system to
develop on, for instance if only 5 percent of potential application users prefer a
certain operating system and that operating system alone is developed for, then
95 percent of potential users are dismissed or unable to even access the
application, limiting use and outreach greatly.

As of the second quarter of 2016, Apple’s iOS has accounted for 12.9 percent of
the smartphone using population in the world. This operating system is the
second most popular operating system in the United States, largely behind
Google’s Android at 86.2 percent. iOS development is generally done by Apple’s
proprietary programming language known as ​Swift. ​Swift was released by Apple
on October 2014 and is an adapted version of the existing language Objective C.
Swift encompasses the key characteristics of Objective C while improving things
like error handling and minimal instruction requirement, although Objective C still
has its advantages such as faster compile times and wider library support. Apple
allows either language to be implemented in their development tool, Xcode,
making Objective C code and files able to be adapted to be recognized as swift
code. It is important to not that Xcode is only available on Mac OS X, so
developing for any Apple software through this Integrated Development
Environment (IDE) will require the use of Apple’s own operating system.

Android is the most popular smartphone operating system in the world as of the
second quarter of 2016. This means the outreach of the application will be
widespread if the digital voltmeter were to be distributed on a drastically larger
scale. The Android operating system uses the very common programming
language known as Java for its application development making it rather versatile
and user friendly if a user has developed and form of object oriented software in
the past. Android development also takes advantage of XML files, which are
used to easily manipulate graphical components of the application without
needing extensive knowledge of the syntax. Android development can be done in
multiple environments such as Eclipse (with the correct libraries added onto it), or
Android Studio which includes a built in simulator to test Android applications.
Android Studio has become one of the most popular IEDs available for Android
development, and is available for multiple operating systems such as Windows,
Linux, and even Mac OS X. One key advantage of Android versus most other
smartphone operating systems is the Google Play Store. Google’s Play Store
offers less restrictions and conditions in terms of applications allowable on the
online store, so developing an application is not hindered by meeting strict online
store policies, instead basic requirements and met and the application could
potentially be available in a matter of hours after completion.

Although arguably the largest software company in today’s technological world,

Microsoft Windows Phone has struggled to penetrate the market in terms of
smartphone operating systems. Windows phone users account for about 0.6
percent of the world’s smartphone user population. Windows phone integrates
Window’s personal computing operating systems such as Windows 10, Windows
8, etc. into the mobile operating system, connecting user data across multiple
platforms. This would increase usability in terms of cross platform usage,
however this adds to the intricacies of developing using the windows platform.
Microsoft windows phone can use many different languages in order to develop
applications such as XAML for graphical interface creation and C#, C++, or
Visual Basic for application behavioral code. Windows phone application
development is usually done using Visual Studio, an IDE created by Microsoft
and is available for Windows, Linux, and Mac OS X.

Linux has very recently entered the mobile operating system market with the
introduction of Ubuntu Touch. This operating system is a mobile version of the
very popular Linux distribution Ubuntu with some obvious adaptations to allow for
use in mobile devices. This operating system may be among one of the most
unpopular operating systems for mobile devices, as the users are generally
required to adapt their current devices in order to use Ubuntu Touch, which
requires a working knowledge of mobile phone software installation and special
permissions. Following in the characteristics of its Linux based operating system,
Ubuntu Touch is vastly customizable and open source, allowing and encouraging
community develop in order to further improve the operating system.
Development in Ubuntu Touch is very versatile, allowing QML or HTML5 to be
used when developing the graphical interface, and JavaScript, Qt, Python, Go or
some other programming languages to be used for the logical design of the
application. The completed application is shared similar to that of Google’s Play
Store, however the outreach is fractional of the Android operating system.
Ubuntu Touch is done by using the Ubuntu SDK IDE, which is only available for
Linux Ubuntu 14.04 and newer.

RIM has many different branches in application development, whether an

application is loaded onto a device with BlackBerry 10, BlackBerry OS, or a
Playbook the development process may be slightly variant. RIM’s Blackberry
accounts for about 0.1 percent of all mobile operating systems in the world.
Blackberry uses multiple SDKs based on the developing style the user prefers.
Blackberry allows users to choose between many different languages of
development also much like Ubuntu Touch. JavaScript, CSS, or HTML5 can be
used for the graphical user interface while the logical design can be done using
C, C++, ActionScript, AIR, Qt, etc. Even Java can be used to develop
applications with a very similar development scheme to Android. BlackBerry
allows development on many different IDEs as well such as FDT5 or Adobe
Flash Builder 4.7 to list options.

3.5.4 Digital Display

Now that we have selected how our data will be wirelessly transmitted to the
Android application, we will being looking that the other accessories that will turn
our meter into a complete and total package. In this section we will review and
select the digital display to be used in our final product. As in the previous
sections, we will discuss the various options and the strategic selection process
for each possibility.

There are several types of digital displays that can be used to display data. The
first option we took into consideration was an LED display. This type of display
uses seven light emitting segments to create one numerical digit. We can
purchase various sizes that contain several digits and can be programmed to
display a decimal point at any location. This type of display is common for
counters, alarm clocks, timers and other applications for clear and readable
values. We first considered this option because we believed it would draw low
power, and we were not initially concerned having an advanced digital display
being that the main use of our meter is to log data on a smartphone application.
The second option we considered was an liquid crystal display, or LCD. These
displays are used for televisions and computers, electronic products, and most
importantly for us, they are the most commonly used display for digital meters.
While there are other types of electrical displays, we had to quickly research their
ability to interact with Arduino platforms. Fortunately, both LED and LCD displays
have supportive software libraries that can be used to control all necessary
functions, and because they are the most common types of displays, online
resources for programming and controlling are readily available for our use. That
being said, we quickly decided our product would use one of these two options.

LCDs operate on an electrical level by blocking light rather than emitting light as
is the case for an LED display. This translates to a lower current necessary to
power the device which, in turn, translates to less power consumption and a
longer battery life for our product. LED displays must have a single digital I/O pin
for each of the seven segments used to display a single digit. This means that in
order to display a three digit value, over twenty output pins must be used!
Fortunately, companies have developed specific integrated circuits for powering
and controlling multiple-digit LED displays. One such example is the TM1637
LED Drive Control Special Circuit. This device uses CMOS technology which
translates into very low current draw and power consumption. There are also
accompanying Arduino libraries that can be downloaded for simple programming.
Additionally, companies such as RobotDyn™ have developed products that use
this driver in a small package to fully control a 4-digit LED display using only four
pins. These four pins are power, ground, and two I/O pins to control data and a
clock signal. This tells us that both LCD and LED displays are very comparable
as it relates to power consumption for our product. Before selecting our display,
we knew that our meter needed to display voltage measurements to at least the
tenth decimal place. Luckily, using conversions within the code, we could display
data for millivolts up to kilovolts on using the same number of digits. Knowing
this, we determined that only four decimal places were necessary to display any
range of voltage for our meter. Once we knew this, we could compare the size
and cost options between the two options. After sufficient research, we found that
both display options were nearly identical and well within our budget. We also
found that the size of either options were within a few centimeters, and both
compact enough to be used in our design.

Similarly to our selection process for the brain behind our meter, it seemed as if
both options had an even amount of pros and cons up to this point. The next
important factor to consider was the minimum number of I/O ports needed for
each option. As mentioned before, the RobotDyn™ LED Drive Control Special
Circuit needs only two digital I/O ports for full control. This is neglecting the
power and ground connections which will be common for arduino and supportive
circuitry because the regulated voltage output of the Arduino and common
ground pins will be routed on the PCB to power all devices. An LCD, on the other
hand requires several additional output pins to drive the display. Size options for
LCD displays vary from as little as 8x2 (8 characters, 2 rows) up to 20x4 and
larger. For our purposes, an 8x2 module would be sufficient to display the
voltage being measured for any range. Even though this is one of the smallest
packages, this device requires six digital I/O pins for full operation. Four pins
control the data being written to the module, one pin enables the writing
operation, and the last pin is used to select the rows and digits where characters
will be displayed. It is necessary to realize that our final product must also control
measurement range switching as well as AC and DC switching. The MCU must
be able to sense these operations for accurate computation within the MCU’s
code. That being said, our final design must use additional dedicated I/O pins to
let the MCU know which range and signal parameter (AC/DC) we are operating
in. In section 3, we found that in order to measure voltage ranges from millivolts
to around one kilovolt, we need at least four “R2” values in our voltage divider
network. This means that we need at least five I/O pins for sensing (one for
AC/DC selection). We knew that our design was subject to change and could
require additional I/O ports, so we wanted at least two more in case we ran into
issues. In total, our microcontroller would need 9 digital I/O ports for and LED
display, or 13 for an LCD.

While the difference in necessary I/O ports seems to be large, even if we were to
use an LCD, the ATmega328P has sufficient I/O pins, specifically 14, which
would be enough for our product with two extra pins for wiggle room. That being
said, we eventually realized that both of the displays would have nearly identical
functional impacts on our final product. This led us to consider the final
parameter to take into consideration which was how we wanted the to display our
data. Initially, we believed a single value, the voltage being measured in real
time, would be enough for our project. This is why we believed an LED display
would be sufficient, because it would also use less power and we could select a
smaller Arduino platform since it would use less I/O ports. But after our extensive
research and analysis, we realized that we were not limited by either of these
constraints and our display did not have to be limited to a single value. We could
give our users more measurement values, such as average and
minimum/maximum readings along with the real-time voltage being measured if
we were to use an LCD. Additionally, we could now program the measurement
units such as mVDC, mVAC, VDC, VAC, kVDC, kVAC, and any other text we
might decide users would like. This was the determining factor for our display,
and why we decided to use an LCD. It should be noted that because both of
these displays are very cheap and well within our budget, we decided to
purchase both in case we ran into trouble in the implementation of our project.

3.5.5 Diodes and Power Supply

Regardless of the microcontroller which we select for our final design, the
external circuitry will be nearly the same. We have determined thus far that our
platform needs at least 14 digital I/O ports for switching and digital display. We
have also confirmed that we need to use both of the Rx and Tx ports for
Bluetooth® communication. In following subsections we will determine any
remaining ports necessary for our final design by discussing the remaining
circuitry components and what our microcontroller needs to realize our final
design. We will first discuss diodes as they will be crucial to AC measurement
and power supply design. In section 3.4.6 we will discuss the input protection
circuitry and necessary components followed by the remaining components that
will be used for voltage measurement such as resistors, capacitors, and switches
in section 3.4.7.

Diodes are two terminal components that allows the flow of current in one
direction by making use of a PN junction. Ideally, this device has zero resistance
in one direction and infinite resistance in the other direction. For our device, this
technology will be necessary for various reasons. The first application of diodes
we must consider are those used for AC measurement. As we have described,
voltage measurement is done by using an analog-to-digital converter. This device
only produces digital values for a positive input voltage. This means that for our
device to measure AC voltage, we have to modify the signal before being input
into the ADC. To do so, we can design a full-wave or half-wave rectifier, or
modify the original signal without rectification. A half-wave rectifier removes the
negative cycle of a sinusoidal signal by blocking current in the reverse direction
with a single diode. Instead of removing the negative cycle, a full-wave rectifier
inverts this voltage to a positive cycle using four diodes in a bridge. The common
use for AC rectification is turning an AC voltage into a DC voltage. Therefore in
either type of rectifier, a capacitor is connected between the output terminals to
hold the voltage at a near constant level, only deviating by a small voltage swing.
A full wave rectifier produces a more stable output with a smaller swing which is
desirable for measurement accuracy. For our project, we designed three possible
measurement options to consider which are listed below before selecting our
diodes.

1) The first option would be to divide the AC voltage using the same network
of resistors used for DC voltage division, and then pass the signal through a
rectifier before being fed into the ADC. If we were to use a capacitor across the
input of the ADC, we can turn the signal into a DC voltage and measure the peak
values of the AC signal. Because the ADC would not see an sinusoidal signal,
we would need to sense the frequency separately so that the data can be plotted
on the smartphone application. This can be done using special pulse-width
modulation ports on the Arduino, but could have undesirable effects on the
voltage measurement circuitry. This method would work using either type of
rectifier (half or full wave).
2) The second option would be to remove the capacitor and feed only the
rectified wave into the ADC. Then, depending on our sampling rate, we could
read all of the values of the positive cycle and generate a complete waveform
from calculated averages. Unfortunately, our program would need to decipher
where the minimum and maximum values in order to calculate the frequency.
This could prove difficult with a fully rectified signal, but using a half-rectified
signal, determining the positive sequence would be much easier by isolating the
ADC values in between the strings of zeros produced by the rectified negative
cycle. Therefore, this option would only need one diode.
3) The last option would be to divide the AC signal and add a positive DC
offset. Because the voltage divider provides minimal deviation for each range, we
know roughly how much voltage will be dropped on the ADC. We can then add
half of this value as a DC offset so the bottom of the negative cycle would be at
zero volts. The DC offset would be removed for voltage calculations and we
would get the original waveform. Therefore, this option would not require a single
diode for operation.
Because we want to use the least amount of components in our final design to
minimize cost and size, we determined the third option would be the best choice
for our design. As mentioned before, this option could possibly have undesirable
effects which could hinder the accuracy of our meter. Therefore we decided to
purchase enough diodes in the occurrence of this design not functioning
properly. Because AC rectification has been a widely used technology for many
years, companies have designed products to take care of these operation using
only a single component. In the case that we must use option 1, we want the
most accurate DC output, meaning the smallest possible ripples, and would need
to use a full-wave rectifier. We therefore ordered a MULTICOMP W01 Bridge
Rectifier Diode, which contains a full-wave rectifier on a single component. This
component has a low maximum forward voltage drop of 1.1V and a forward
current of 1.5A which is more than sufficient for our design. If we were to used
option 2, we would need a single diode for half-wave rectification. We therefore
purchased a 1N5818 Schottky diode. This diode is perfect for AC rectification as
it also has low forward voltage drop and high current capacity.

The next possible application of diodes in our product are for the power supply.
Before we can discuss this application, we need to once keep in consideration
the possible microcontroller options. For the microcontrollers we are currently
considering using in our design, all have an operating voltage around 3.3-5V.
Because we will be implementing our own voltage regulation, we can design the
rest of the circuitry based on our power supply selection. Once we have selected
how our device will be powered, we can then discuss if diodes will be necessary
for our power supply.

At this point in the selection process, taking into consideration the selected
components to be used in our design, we have covered almost all constraints
that would possibly prevent us from being able to use the ATmega328P
microcontroller. Although we would ideally like to use the ATmega328P because
of the available step-by-step guides for mirroring the development board along
with PDIP package options which will aid in the breadboard design/testing, we
still have the option to use other microcontrollers and design the supportive
circuitry if need be. That being said, we will continue to examine and relate the
remaining microcontroller possibilities before choosing a path. Our first area of
interest for power supply was battery lifetime. We wanted a power supply that
would be above the microcontroller’s operating voltage has to offer a long life
span, while also being an appropriate size for the meter. After taking in these
considerations, we decided to use a 9V battery to power our device. This option
would allow our product to operate properly until the battery discharged to around
6V for the worst case microcontroller option. We believe this option will be
sufficient for our requirement specification of at least 200 hours.
Because we chose a 9V battery for our power supply, we had to consider the
connector which would be used in our product. Our two options were a solid
metal connector, or a wired connector. Because our final packaging design had
not been complete at this point, we decided to purchase both connectors
because of their cheap cost. Regardless of which connector we use, their
common denominator is the ability to ensure the 9V battery cannot be inverted
during installation due to their different sized positive and negative terminals.This
brings us back to the topic of diodes. A common use of diodes for power supply
is protection. This protection is for both the circuitry being powered and the
power supply itself. If power supply polarity is accidentally inverted, this can
potentially harm the microcontroller and supportive circuitry. Though it seems the
9V battery connector would prevent this from occurring, we may find ourselves
needing to use a different power source down the road. In this case, we need to
discuss how diodes can aid in the power supply and voltmeter protection by
looking at a few different options.

The first and most common configuration placing a diode in series with the
positive terminal of the power supply. If power supply, take a battery for example,
is accidentally inverted, an inline diode will not be forward biased and no current
will be passed. This causes the circuit to essentially be open and the power
supply is protected from being short circuited. Unfortunately, when the battery is
correctly installed, the diode is then forward biased and thus has a voltage drop,
usually around 0.8V, which lowers the voltage being supplied to the circuitry. This
is undesirable as it will shorten the battery’s usable lifetime along with
contributing to power loss. To help remedy this we can use a Schottky diode.
This diode is designed to have a low forward voltage drop, around 0.4V
depending on the specific diode through decreased resistance across the PN
junction. While this is better than a traditional diode, we can do even better. Our
third option is to use a P-channel MOSFET where the gate is connected to
ground, the drain to the positive terminal of the battery, and the source to the top
of the load. In this configuration, when the power supply is is connected properly,
the transistor is actively biased, meaning the voltage between the gate and
source (V​GS​) is sufficiently negative, and thus current flows and power is
delivered to the circuitry. When the polarity is inversed, V​GS is positive and
current will not flow from drain to source and the circuit will essentially be open.
Additionally, when the battery is inserted correctly, the resistance from drain to
source (R​DS(on)​) is very low, depending on the quality of the component. This will
therefore contribute magnitudes less power loss while also protecting the circuitry
and power supply. While we still plan on using a 9V battery which does not
necessarily need this extra measure, we decided to purchase a 60V P-channel
MOSFET made by Fairchild Semiconductors. This component has a maximum
R​DS(on) value of 70mΩ which is low enough so our 9V battery will not drop a large
voltage and the power loss will be minimal. Additionally, in case we run into
trouble with this arrangement, we decided to purchase a 1N5818 Schottky
Rectifier diode. This diode has a very low forward voltage, around 0.4V with
250mA current draw.

Before moving on to the remainder of our electrical circuitry for safety and
measurement, we will finish selecting the additional components needed for our
power supply. At this point in the selection process, we have slowly narrowed
down our microcontroller options to just two. These are the ATmega328P and the
ATmega32U4. While we have determined that both options have sufficient I/O
ports and also meet the rest of our requirements, there are a few differences that
separate the two such as built in functions for programming and serial
communication, as well as external power regulation requirements. We will save
the other comparisons for later and focus now on the power supply. The
ATmega328P requires a 5V regulated input, while the ATmega32U4 requires
both a 5V and 3.3V regulated inputs. The latter option would require an additional
voltage regulator along with two additional capacitors for operation. Although we
would like to use the fewest components for cost and size concerns, we have yet
to analyse the remaining differences as mentioned before and therefore we will
discuss the power supply design and necessary components for both options.

For both cases, we need to have a 5V regulated voltage to the microcontroller’s

V​CC port. To do do this, we will be using a LM7805 voltage regulator from Texas
Instruments. This product takes an unregulated input voltage from 7-25V and
outputs a regulated voltage determined by the test conditions such as output
current and power draw. From the LM7805 datasheet, for test conditions of V​I​=
7-12V, I​O​= 5mA to 1.5A and P​D ​< 15W, the device will produce a regulated 5V
output with a maximum swing of ​+ 50mV. The device requires two decoupling
capacitors, one from the input pin to ground and one from the output pin to
ground. These capacitors help to filter out high-frequency noise to produce a
smoother output. From Arduino’s tutorial on building your own development
board using the ATmega328P, they use two 10μF These three components
along with the 9V battery+connector and P-channel MOSFET described earlier
are the only components necessary for our 5V power supply to the board. This
setup will be used regardless of the microcontroller we select. If we are to use
the ATmega32U4, we have to include two additional components for 3.3V
regulation. From Arduino’s website, we obtained the schematics for the Arduino
Micro which uses the ATmega32U4. This platform has roughly the same 5V
regulation as described above, with an additional LP2985 voltage regulator and
1μF capacitor. The output of the 5V regulator is connected to the input and
ON/OFF pins of the LP2985 with the decoupling capacitor between the output pin
and ground. Although we are leaning towards the Atmega328P, we will still keep
these components in consideration and will determine if they are needed once
we make our final MCU selection at the end of section 3.
3.5.6 Safety Components

Now that we have identified usage of diodes for both AC measurement and
power supply protection, we will discuss the components used for input
protection to the meter. These components are of utmost priority when designing
the input to our meter, as they will be the barrier between dangerous electrical
phenomenon and the rest of the meter. We will discuss how these components
will protect both the product itself and more importantly the user.

3.5.7 Electrical Measurement Components

In this section we will describe how our microcontroller will be able to measure a
wide range of AC and DC voltages and what components are necessary for this
operation. We will begin by discussing the internal technology which will allow
our microcontroller to measure voltage and then we will discuss the various
design options which will allow this to happen.

Digital voltmeters operate on an electrical level by making use of an

analog-to-digital converter, or ADC. An ADC is a system which converts a
continuous-amplitude, continuous-time analog signal into a discrete-amplitude,
discrete-time digital signal. There are several different types of ADCs such as the
flash ADC, digital ramp ADC, successive approximation ADC, and many others.
Microcontrollers such as the ones used in Arduino platforms have built-in analog
comparators that serve many functions such as analog-to-digital conversion and
multiplexing. For our product, we are interested in using the ADC to convert our
input voltage into a digital signal which the microcontroller can decipher through
code and generate the original voltage value that we can display and wirelessly
transmit. Both the ATmega328P and ATmega32U4 use successive
approximation ADCs. This type of ADC begins by using a comparator to compare
the input signal to a predetermined reference voltage in sequence with a
successive approximation register (SAR) and a digital-to-analog converter
(DAC). Without getting into too much detail about the circuitry inside the system,
the takeaway for our purpose is that the component needs only to have an
analog voltage applied at the input for conversion. There is no current limiting
requirements necessary for the system to function correctly. This is very
beneficial in our case because the voltmeter must draw as little current as
possible in order to make an accurate measurement and not interfere with the
performance of the circuitry under inspection.

The ADC used in our two microcontroller options are both 10-bit and operate with
input voltages from 0 to a specific reference voltage which is used for the internal
comparators. This reference voltage (AV​REF​) can be set by the user to various
values, but the microcontroller automatically will automatically default to the
operating voltage which we have established will be 5V. The resolution of the
ADC is defined by the number of specific values it can produce. In our case, a
10-bit (N) ADC can produce 2​10 or 1024 distinct values. The resolution is defined
by the formula AV​REF​/(2​N​-1). If we are to use the default 5V reference voltage, this
gives us a resolution of 4.89mV. This means that when the input to the ADC
changes by 4.89mV, the ADC is guaranteed to produce a new value. If we were
to lower the reference voltage, let us say 1.1V for example, we can increase this
resolution to 1.08mV. Unfortunately, this means that our range of input voltage is
limited at 0 - 1.1V. Because our meter would be be useless if could only measure
this range, we have to make use of additional components so that we can
measure a large range without violating our input range of the ADC.

As we described in section 3.3, voltmeters must have a very high input

resistance as to not draw current away from the circuitry under inspection. With
this high input resistance in place and no current being drawn, the voltage at the
ADC input will be the same as the voltage being measured. This means that we
must use a voltage divider with a second resistor (R​2​) and measure the voltage
across this component. Using this method, we can select specific R​2 values that
will allow for a small voltage drop within the ADC input range for a large range of
input voltage. Even though using voltage division will increase our measureable
range, we cannot use a single R​2 value for the entire range as lower input
voltages will drop values on R​2 which are smaller than our resolution. Therefore,
we must design a network of resistors that will be used for division for different
ranges of measurement. This strategy was briefly defined in section 3.3. Before
we can begin calculating these values, we must first identify the range of values
in which our product should be able to measure.

Because our meter is intended to be used for electronic hobbyists, general

household appliance maintenance and even electricians, we want our meter to
be able to measure a steady-state voltage range of 1mV up to 1000V. As
described in the previous section on safety design, this 1020V is the upper limit
to what we can accurately measure as any higher voltage will be above the three
varistors’ clamping voltage and shunt the current before entering the IC. Although
the safety components add some resistance to the input, we will select our
nominal input resistance to be 10MΩ. This value will be the R​1 resistance for the
voltage divider. We will calculate the remaining R​2 values by first addressing the
ADC reference voltage as these must be calculated hand-in-hand. Both the
microcontroller possibilities have the option to externally set AV​REF to any voltage.
Because we want a high resolution, we will set this value to 1V. That being said,
each resistor in our divider network must only drop a maximum of 1V onto the
ADC. This gives us a resolution of 1/1023 or 0.9775mV. This resolution will
ensure that when measuring in the millivolt range, our meter will be able to detect
when the input changes by 1mV. It should be noted that when measuring in the
millivolt range (1-999mV), the voltage can be directly fed into the ADC. Because
our range grows larger for each resistor value, the precision of each range will
decrease by 1mV, even though the optimal value would be our resolution of
0.9775mV. Using the voltage divider equation with the values of Vout = 0.1mV,
R1 = 10MΩ, and Vin ranges being known, we calculated the R​2 values shown in
Figure 3.1 below.

Table 3.2: Resistor Divider Network Values

When the input voltage for the first range (1.00-9.99V) changes by 0.01 volts, or
10mV, the output voltage being fed into the ADC changes by 0.0009999V or
0.9999mV. This change is slightly larger than our resolution value of 0.9775mV
ensuring that we can measure down to 10mV of accuracy for this range. Similarly
for the second range (10-100V), when the input voltage changes by 0.1V, we will
see the same 0.9999mV change on the output meaning our meter can accurately
measure down to 0.1V in this range. And as is evident, for the range of
100-1000V we can measure changes on the input down to a single volt. This
design intuitively confirms that our product should be able to measure within
0.1% of the precision and accuracy that advanced modern meters are able to
measure which meets our requirements specification.

Now that we know the resistance values we need for measurement, we can
begin selecting our specific parts. Because these resistors must be very precise
values, we determined that the tolerance must be as small as possible. We
therefore decided to purchase 0.1% tolerance, through-hole resistors. Although
we would ideally like to have surface mounted resistors for minimizing space, we
had to acknowledge that we must test our designs on a breadboard.

As we mentioned previously, the R​2 values calculated above were based on our
nominal R​1 value of 10MΩ which we knew would be affected by the added
resistance of the safety components. Additionally, although we selected resistors
with the best tolerance of 0.1%, we knew that these values had to be exact for
our calculations to be correct. We therefore purchased a 13 value assortment of
trim potentiometers that we will use to compensate for error in both the R​1 and R​2
resistors.
Now that we have determined how many resistors we need for accurate
measurement, we can discuss how the range selecting will be implemented. To
be able to measure off of each of these resistors, we have to have a separate
analog input for each. Because we have four ranges, our microcontroller must
have at least four analog inputs. Both the ATmega328P and the ATmega32U4
meet this requirement with the former having six channels and the latter having
eight. While both can satisfy our requirement, it may prove valuable to have more
channels as we may run into issues using the same channels for both DC and
AC measurement. Though this can occur, we are still going on the assumption
that we will be using the same network for both measurements. That being said,
we must now consider the switching required to select the correct range.

After several different designs, we came to a final setup that would allow safe
and efficient range switching for accurate measurements. The range selector will
be a 4-bit, 2-position dip switch with all four switches connected together on one
side as shown in Figure 3.2 below.

Figure 3.6: Range Selector Switch

In this configuration, the user can simply select the desired range of
measurement by sliding a single switch. When no switch is selected, the circuit is
open and no voltage will be measured. Additionally, when no range is selected
there will be no connection to the ADC and the microcontroller will be protected.
The most ominous design flaw in this configuration is the possibility of selecting
more than one range at a time. This would effectively place two or more resistors
in parallel, which would produce an inaccurate measurement. Fortunately,
because we are using separate channels for each range, we can write code that
checks if more than one analog input channel has voltage applied. We can then
warn the user by illuminating an LED which would be placed near the range
selector switch. Alternatively, we could connect a piezo speaker to sound a
warning signal, or a combination of both LED and speaker. This part of the
design will be determined after testing and further consideration. While there are
many types of switches and methods we could use for range switching, we
believed this to be the most effective design as it relates to cost, size, and
functionality.

The final switch we will need for our design is for the power supply to connect to
the microcontroller. Because this will only function to turn the meter on and off,
we only need a two-position switch that will be placed in-line with the positive
terminal of the battery and the P-channel MOSFET for reverse-polarity
protection. We therefore ordered a 2P2T vertical slide switch from Microtivity.
While we are on the topic of power supply, we also chose to add a single LED
that will be powered by the 5V regulated supply in series with a 220Ω resistor to
indicate when the device is powered on. This resistor is to limit the current
flowing into the LED and therefore does not need to be a low tolerance
component. In addition, this LED will not illuminate if the 9V battery is installed
incorrectly.

The remaining resistors necessary for our project are needed for controlling the
reset pin of the microcontroller, the brightness and contrast of the LCD, indication
LEDs, and for the AC/DC selection switch. For the microcontroller’s reset pin, we
must pull this to 5V (Vcc) in order to prevent the chip from resetting. In addition to
this pull-up resistor, we will connect a small momentary switch to take advantage
of the reset pin when writing new code and debugging. The pushbutton switch
will pull the reset pin to ground causing the chip to reset. For this operation, we
are going to use a 10kΩ resistor and a small momentary push-button that we
have on-hand. For the indication LEDs that will be used for debugging the
prototypes, we will be using 220Ω resistors to limit the current flow through the
diodes. These indication LEDs will be connected to the Rx and Tx pins of the
microcontroller to indicate when data is being wirelessly transmitted. Additionally,
we will be using a permanent indication LED for warning the user when multiple
range selection switches have been connected. If we choose to use a piezo
speaker for warning instead, we will not need a resistor as this component
operates through a pulse-width modulation (PWM) digital I/O pin alone. The last
resistors that will be used in our product are the ones to control brightness and
contrast of the LCD. We will use a 10kΩ potentiometer to determine the optimal
levels and then use appropriate standard resistors to set these values. As they
are not known at this point, we cannot determine their specific values until
testing. As with the pushbutton for the reset pin, we will connect another switch
with 10kΩ pull-up resistor that will be used to switch between AC and DC
measurement modes.

Now that we have determined all of the necessary resistors for our project, we
will now look at the capacitors necessary for remaining operation. As we
mentioned before, we need two decoupling capacitors for the 7805 voltage
regulator. We therefore purchased two 10μF ceramic capacitors rated at 20V
which will be sufficient for our regulated 5V supply. If we are to use the
ATmega32U4 which requires a 3.3V regulated power supply, we will also need a
decoupling capacitor. We therefore purchased a single 1μF tantalum capacitor in
the case that we use the ATmega32U4. The final capacitors that will be
necessary for our microcontroller are the capacitors needed for the
microcontroller’s clock signal. Both of the microcontroller options operate using a
16MHz crystal oscillator and need capacitors to regulate the frequency of the
crystal. The capacitance values needed depend on the load capacitance of the
specific oscillator used. We purchased a COM-00536 16MHz Crystal from
SparkFun which specifies the load capacitance to be 20pF. From Arduino’s
website, we obtained the open-source schematics for the Arduino Uno and the
Arduino Micro which use the ATmega328P and ATmega32U4 respectively. Both
of the schematics recommend using two 22pF capacitors, one from each pin of
the 16MHz clock to ground. From the crystal’s point of view, the two capacitors
are in series giving an 11pF capacitance. The remainder of the load capacitance
is taken care of from stray capacitance on the microcontroller’s pin. Although the
frequency of the crystal may not be exactly 16MHz, so long as the value is close
to 16MHz, the IC will operate correctly.

3.5.8 Test Leads

A test lead is the combination of a conductive probe, cables and terminating

connectors. There are several different types of test leads for circuit
measurement such as voltage, current, and oscilloscope leads. Digital
multimeters and voltmeters use simple voltage test leads with pointed probes,
some having the option to attach alligator clips. Oscilloscopes on the other hand
can use various complex and sophisticated leads for specific applications. Since
we are designing a meter to only measure voltage, we need only simple voltage
leads. We now must consider the various probe body options as they relate to
safety and intended use. The probe body is composed of the conductive probe
and insulated housing. For all options we will consider, the link between probe
body and meter will be the same. The probe body will attach to a single
conductive wire and then the male terminal connector which will mate with a
female connector and subsequent traces on the PCB. The two most common
probe types are pointed metal tips and alligator clips. Because the intended use
of our meter is for quick measurements, we decided to use pointed metal tips as
these are most maneuverable and inexpensive. With the probe type selected, we
now had to consider the safety features and what will be necessary for our
design. Because our meter will rated as Category III, our test leads must also
conform to this rating. Though the meter ratings are based on ability to handle
various levels of transient voltage spikes, the parameter that governs our test
lead’s rating is the current handling capabilities. In the cases of high voltage
transients, power supplies being shorted, component failures, or even user error,
large amounts of current can be sent through the test leads. Although there will
be numerous input protection components and procedures within the meter’s
design, it is necessary to ensure our leads can handle large amounts of current
without failure. Thin gauge wires will heat faster when current is passed through
in comparison to thicker wires. A hot wire in-turn could cause the probe body to
heat up enough, or even melt, possibly resulting in severe burns. We therefore
need to ensure our wire which connects the probe to the PCB is thick enough to
prevent this.
The second safety measure we have to consider while selecting our leads is the
external design. Fortunately for us, Fluke has developed a safety guidelines
document titled, “ABCs of Multimeter Safety.” In this document, Fluke lists four
general safety guidelines when purchasing meters as they relate to the test
leads. The first check is having double-insulated leads. The second is ensuring
the meter has recessed input jacks and test leads with shrouded input
connectors. The third check is to ensure the test leads’ probe body has non-slip
finger guards so that the user does not slip and come into contact with the
circuitry being measured. The last measure is that the meter and leads are made
of high quality, durable, and non conductive materials. After considering this
advice, we decided to purchase a pair of red and black, 1000V test leads with
pointed probes manufactured by DMiotech. These leads have 4mm male banana
jack connectors which will be complemented by two PCB mount, recessed
female connectors. Although the leads themselves are not shrouded, the
recessed female connectors will ensure that user skin cannot come into contact
with the conductive metal on the meter.

3.6 Parts Selection Summary

Through large amounts of research and careful consideration of all aspects that
could influence the accuracy and functionality of our product, we slowly narrowed
down the possible processing technologies that could potentially be used as the
brain of our meter. We first made the decision to use a microcontroller which led
us to compare the various brands and models available. After considering the
languages used and available resources for troubleshooting, we decided to use a
microcontroller that can be programmed using the Arduino IDE. As there are
several different MCUs used in the various Arduino platforms, we had to stop at
this point and consider the rest of our design. After selecting all of the
components that would be necessary to realize the remainder of the design, we
narrowed our microcontroller options to two models: the ATmega328P and the
ATmega32U4. Both of these options would meet the minimum design
requirements of our project, so it came down to which product would be both
cheaper to implement and easier to use on the breadboard. As the ATmega32U4
is a surface-mount device, we would have to solder individual pins to each of the
contacts in order to build and test our designs on a breadboard. For this reason,
along with the cost of the IC and supportive circuitry being cheaper, we chose to
use the ATmega328P. This is a PDIP package type which can be directly placed
on a PCB, and it can be mated with an adaptor socket for breadboard use. We
purchased this socket along with the remaining circuit components that are
needed to fully operate the device. These parts included an LM7805 voltage
regulator, two 10μF decoupling capacitors, a 16MHz crystal oscillator with two
22pF capacitors, and a USB to serial Breakout board from SparkFun to program
the chip with the Arduino IDE. Additionally, we needed to purchase a pushbutton
that would be connected to the RESET pin of the MCU that would pull the pin to
ground when we need to write new programs to the device.

After taking into consideration all of the possible remaining components for our
final product, we chose several different components for each operation as we
knew we could possibly run into issues with our potential designs. When
researching the safety measures that would be used for our product, we came
across several different technologies that could be used to protect the meter and
the user from overload conditions such as high voltage transients. After extensive
research for which technologies our product would need, we purchased the
following safety components: 340V varistors, 365V varistors, a thermistor (PTC),
fast-blow fuses, and a few 5W wirewound resistors. We also purchased Cat III
rated test leads for measurement which will be connected to the PCB using
female banana jack connectors. For the protection of the 9V power supply, we
purchased a 60V P-channel MOSFET and a pack of 1N5818, as we were unsure
of the final design at this point.

For the input voltage division, we needed several resistors and potentiometers to
allow for a large range of measurement. We knew that our input impedance
would have to be above 10MΩ in order to not compromise the the circuitry under
inspection. For this reason, we purchased several of the following resistors to be
used for the input measurement: 10MΩ, 9.5MΩ, 1MΩ, 100kΩ, 10kΩ, 1kΩ, 10Ω
and 1Ω. Because these resistors must be very precise for accurate measurement
and calculations, all of these components have a 1% tolerance. Additionally, we
were aware that in order to adjust these values for maximum accuracy, we
needed to be able to vary these resistors. We therefore purchased a set of trim
potentiometers with values from 100Ω to 1MΩ. The resistors used in the
remainder of the circuitry include: 10kΩ pull-ups, 220Ω resistors for the LEDs,
2kΩ and 4kΩ resistors for the Bluetooth® module, and the 100kΩ and 400kΩ
resistors used to set the reference voltage of the ADC on the AREF pin.

For the digital display, we decided to purchase two different types as we may run
into issues during implementation. The first device we purchased is a four-digit
LED display with driver chip installed. After further design changes, we decided
to purchase a 20x4 LCD which would be used instead of the LED display. Using
this option, we could display more data to the users such as maximum, minimum,
and average values when measuring in DC mode. In AC mode, we could display
the peak voltage, the RMS voltage, and the frequency of the measured signal. To
implement this display in our product, the only supportive component necessary
is a resistor to control the contrast of the display. As we have not yet determined
which value will be optimal, we will use a 10kΩ potentiometer until we decide
which resistor value will be used.
For indication purposes, we purchased an assortment of various colored LEDs.
We will use a single green LED that will be connected to the power supply to
indicate when the device has been turned on. To indicate the user of which
measurement mode, whether AC or DC, we will use two separate blue LEDs.
One will be labeled as AC, and one as DC and will illuminate accordingly when
the AC/DC selector button is pressed. The final LED is red and will be used to
warn the user when more than one range selection switch has been activated.

In order to transmit the measurement data wirelessly to the smartphone

application, we researched several options considering the pros and cons of
each. After extensive research, we decided to use Bluetooth® transmission as
this option will meet the minimum requirements and will be the easiest to
implement. We will be using the Arduino IDE to program the microcontroller
which contains specific libraries to aid in Bluetooth® transmission. We purchased
a Solu JY-MCU HC-06 Slave Bluetooth® Serial Port Transceiver. To use this
product, we only need to make the necessary connections to power, ground, and
the RX and TX ports of the Arduino. However, as this module operates on 3.3V
logic levels, we need to use a voltage divider with the two resistors mentioned
above.

For the Smartphone application, after consideration of the varying operating

systems discussed in the document, the application’s developing operating
system was finally decided to be designed using the Android operating system.
The Android operating system has one of the easiest processes of application
publication which is crucial for this project, as time is very limited. The operating
system’s application development process is also rather simplistic in terms of
language knowledge as Java is one of the most popular programming languages
and XML is a very visually based design language, with the GUI in Android studio
allowing components to be placed at the visual level including size changes and
position of these components on screen. Android is also the most popular
operating system used worldwide, which allows the application to be distributed
with the largest possible audience of users. Although this application is currently
only under consideration of design using the Android operating system, the
possibilities of expanding compatibility to other major operating systems is still
present depending on the success and overall user satisfaction of the Android
applications performance.
The remaining components needed for our product are the switches and LEDs.
For the range selection, we purchased a 4-pin 2-position DIP switch. For the
RESET pin of the microcontroller and the AC/DC selector, we purchased two
momentary pushbuttons as these will only need to be pressed once for
operation. The last switch is a 2-position slide switch connected between the
battery and the input of the voltage regulator that will be used to turn the device
on and off. We will be using a red LED to indicate when more than one
range-selector switch has been enabled as a warning to the user of incorrect
measurement. We will also be using a green LED that will be permanently
connected to the power supply to indicate when the device has been turned on.
The components that will be used in the final design are shown in Figure 8.

Figure 3.7: Parts

4 Standards and Realistic Design
Constraints
4.1 Standards

In the following section, we will cover all of the standards that govern the
voltmeter we are making. We will begin by discussing hardware standards. This
is a very important section, because we must design our device according to the
specifications of industry standards. These standards relate to the usage and
safety of the voltmeter. making sure our design is made to within the constraints
of the materials we are using is very important. A poorly design voltmeter or one
that does not follow standards can harm users when they use it like giving an
electrical shock that can cause in the human body being harmed (organs failing)
and even cause death in some situations. Our voltmeter needs humans contact
to make it work, this means that it needs to use the acceptable voltage that won't
shock the user. Another area we need to focus on is looking at the materials that
we be used in the voltmeter, we want to make sure the device’s materials are not
hazardous to the user and the environment. For each material we use and buy,
we shall check to see if it is compliant with environmental usage. We will then
discuss the standards necessary for software needed for our voltmeter. Basically
they are two parts for our software, the first part is the software that will run on
the microcontroller board we are designing. This software is closely related to the
hardware’s functionality, what this means is that the hardware mechanism must
feed into the microcontroller and the software from there will take care of the
data. the second part of the software is the application software that will run on
the phone tablet. this software is mainly for receiving the data from the
microcontroller and displaying to the user. A large part will spent on the
hardware software constraints because we must make the hardware working at
maximum capacity while locked within the standards. This part of the report will
begin with the hardware standards, then proceed to the software standards. will
work In section 4.2, we will discuss all of the possible design constraints that
must be taken into consideration for our design.

4.1.1 Hardware Standards

Since we are designing a voltmeter with the expectations of selling it to

consumers. It is very important that the voltmeter we are designing meets
industry standards and is compliant with regulations and constraints of our
market. the constraints are applied depending on what type of device we are
making, different electrical appliances have different standards. Since we are
making mobile electrical appliance, so it is normal that making our product
around mobile standards of electrical components. For example our voltmeter will
be mobile, so we need to design our system within standards of mobile
multimeters. another hardware example is safety, the two are closely related,
safety depends on the hardware used to make the voltmeter.

1. Mobility Standards
● There is not enough of information and official standardization on the mobility
standards. However we did some simple deduction and came to the
conclusion that our voltmeter must be as mobile as other devices of similar
nature. The closest device to our voltmeter are multimeters, as matter of fact
we can consider voltmeters as subsection of multimeters, being that the
voltmeter is more specific (only measures voltage) than multimeters. The
voltmeter is a subsection of multimeters therefore it is safe to say the
standards of mobile digital multimeters also apply to the voltmeter that we are
making. specifically we are designing a device that weighs the same as
other industry multimeters and possibly less since it will have less
functionality and components. The average weight is around 1 ounces. This is
based off sampled data collected from official electronic seller online. we
looked at the weights of ten multimeters which all have similar weights with
littler (between 0 to 1 ounces) and averaged their weights. Current
multimeters in the market is around ​1 OZ based on the average taken​, we
expect to make our product within the same range for the purpose of making
it easy to carry around. Our voltmeter has the possibility of weighing less
because it has less features on it (this could change).

Safety is very crucial when you dealing with voltages, the potential to be shocked
is high and likely to occur, so we had to research what safety standards apply to
voltmeters into safety standards as well. based on what we wanted in our
system, our standard were divided into two parts. hardware and software, we
shall begin with the hardware standards first followed by software standards
later. we focused on these main standards:

2. Safety Standards
○ Our voltmeter shall follow the standards of the ​“international Electronical
Commission (IEC)”. We shall use the ​IEC 61010-1 standard because it offers
a higher level of safety and the low-voltage (<1000V) requirement we will be
using, it is also the test equipment needed for the highest levels of protection
for our portable device. The hardware standard(s) that we will mainly be
concerned with are the ​“measuring categories”. ​The categories deal with
voltage operation that the devices need to work on. We use what is known
test instrument. It is rating system used to find the ability to withstand a
voltage spike, which are applied through a specified level of resistance. The
ratings are broken down into categories - CAT I, II, III, and IV. Before we talk
about the different categories and which one applies to our voltmeter. It's
important to talk about why we have such categories.
○ Voltage Spikes (unavoidable hazards) - multimeters are subject to much
higher voltage than the user anticipated when taking measurements. There is
a time when the user/ engineer has to measure a medium voltage appliance
in the range of 5000 V using a multimeter that can only measure 1000 V or
less.
○ Transients - are the most common voltages likely to happen and the ones
that users should be protected against. hazards likely to occur because most
likely users will have to measure voltage which are like to have high
transient. These are some of the dangers of transients and which are likely to
happen without warning.
■ Lighting Strikes - causes a transient on the power line, which in turn
strikes an arc between the input terminals ​inside the meter.
■ High Fault Current - possibly several thousands of amps– ows in
the short circuit just created. This happens in thousandths of a second.
○ these are the categories.
■ CAT I: ​rated test instruments are signal-level tools for
telecommunications and electronic equipment. Transient voltage risk is
limited, but still exists, due to the distances between equipment locations
and other equipment located between it and the primary electrical
supply. The IEC no longer specifies protection levels for CAT I
instruments. Under UL 3111-1, a Cat I 150V rated instrument must be
protected to 800V. Under IEC 61010-1 2nd edition, a Cat I 150V meter
could be protected only to 500V, as long as that information is in the
user's manual.
■ CAT II: ​rated test instruments cover the local level of circuits for fixed or
non-fixed power devices. This includes most lighting equipment,
appliances, and 120V or 240V equipment inside a building.
■ CAT III: ​rated test instruments can withstand the transient voltage range
found on most distribution circuits. These instruments are used primarily
on fixed primary feeders or branch circuits. They're separated from CAT
IV utility service or other high-voltage source equipment by at least one
level or through transformer isolation.
■ CAT IV: ​rated test instruments are designed for testing on the primary
supply source, which also includes 120V or 240V overhead or
underground lines that power detached buildings or underground lines
that power well pumps. The CAT IV rating covers the highest and most
dangerous level of transient overvoltage electricians encounter when
working on utility service equipment like exterior transformers.
 Measurement Working Peak Impulse Test Source
Category Voltage (DC Transient (20 Ohms = V/A
or AC-rms to repetitions)
ground)

CAT I 600 V 2500 V 30 ohms source

CAT I 1000 V 4000 V 30 ohms source

CAT II 600 V 4000 V 12 ohms source

CAT II 1000 V 6000 V 12 ohms source

CAT III 600 V 6000 V 2 ohms source

CAT III 1000 V 8000 V 2 ohms source

CAT IV 600 V 8000 V 2 ohms source

Table 4.1: Transient test Values for measurement categories

○ The table above breaks the voltages needed for each categories, It helps ‘us’
the engineers of the voltmeter to understand an instrument’s true voltage
“withstand rating”. the Test procedures take into account ​three main criteria:
steady-state voltage, peak impose transient voltage and source
impedance. ​This table gives us an understanding of what is going on:

■ Within ​a category, a higher “working voltage” (steady- state voltage) is

associated with a higher transient, as would be expected. For example,
a CAT III-600 V meter is tested with 6000 V transients while a CAT
III-1000 V meter is tested with 8000 V transients. So far, so good.
■ What is not as obvious is the difference between the 6000 V transient for
CAT III- 600 V and the 6000 V transient for CAT II-1000 V. They are ​not
the same. This is where the source impedance comes in. Ohm’s Law
(Amps = Volts/Ohms) tells us that the Ω test source for CAT III has ​six
times the current​ of the 1 Ω test source for CAT II.
■ The CAT III-600 V meter clearly offers superior transient protection
compared to the CAT II- 1000 V meter, even though its so-called
“voltage rating” could be perceived as being lower. ​It is the
combination of the steady-state voltage (called the working
voltage), and the category that determines the total voltage
withstand rating of the test instrument, including the all-important
transient voltage withstand rating.

○ Creepage and clearance - in addition to being tested to an actual

overvoltage transient value, multimeters are required by IEC 1010 to have
 minimum “creepage” and “clearance” distances between internal components
and circuit nodes. Creepage measures distance across a surface. Clearance
measures distances through the air. The higher the category and working
voltage level, the greater the internal spacing requirements. One of the main
differences between the old IEC 348 and IEC 1010 is the increased spacing
requirements in the latter.
○ As you shown above safety category was setup to protect the user against
transients, safety must be build into the test equipment. Our Task is to find
which category our voltmeter fits in. The real issue for multimeter circuit
protect is not just the maximum steady state voltage range, but the
combination of of ​both steady state and transient overvoltage withstand
capability​. When transients ride on high-energy circuits, they tend to be more
dangerous because these circuits can deliver large currents. If a transient
causes an arc-over, the high current can sustain the arc, producing a plasma
breakdown or explosion, which occurs when the surrounding air becomes
ionized and conductive. The result is an arc blast, a disastrous event which
causes more electrical injuries every year than the better known hazard of
electric shock.
○ For our voltmeter we are specifically going to use ​“CAT 3” ​because our
device will not be directly connected to mains (AC power source). We will be
using a portable power source likely a battery. consequently this come with
benefits, having voltmeter not connected to “mains” allows the device to be
protected, by making sure the it is connecting to a source circuit which
measures limit transient: another benefit is the “overvoltage” or “high voltage”
is calibrated to the right level, so the user doesn't have to worry about getting
an electric shock during the voltmeter operation.
○ Overload Protection - is another factor that has to be considered on top of
CAT III, it is simply a protection circuit that clamps high voltage to an
acceptable level. another addition is the the thermal protection circuit that
detects an overvoltage condition, it protects the meter until the condition is
removed and then automatically returns to normal operation the most
common benefit is to protect the voltmeter from overloads when it is in ohms
mode. In this way, overload protection with automatic recovery is provided for
all measurement functions long as the leads are in the voltage input
terminals.
○ In conclusion to “safety standards section”, it was important for us to imagine
the worst possible scenario where our voltmeter was going to be used, we
looked at the demographic and the environment it was going to be used in.
Once we finished that, we looked at the safety standards that applied to our
problem. Once we finishing making the voltmeter, ​we are going to test it
using the category we have chosen to see if it has met those standards
and passed, if it is passes it will be certified as safe and compliant with
regulation.
3. Serial Communication standards
○ In the world of embedded electronics and microcontroller, devices
communicate to each other through communication protocols, there are many
protocols (in the hundreds). but they can be grouped into two categories:
serial and parallel communication (NOTE: these communication use wired
cables to enable communication between two or more devices).
■ Parallel Communication
● parallel protocols enables the transfer of multiple bits at the same
time through buses of data (8,16, or more wires). the main advantage
of this protocol is how fast it is. compared to serial communication.
■ Serial Communication
● This is the most common form of communication between electronic
devices. Communication serially involves sending series of digital
pulse back and forth between devices at a mutually agreed-upon
rate. The sender sends pulse representing the data to be sent at the
agreed data rate (bits per second) and the receiver listens for the
pulses at that same rate. The devices both have clocks that
determine at which rate they will exchange information.This protocol
transfers data one bit at a time. they can operate on a single wire as
well. what happens during serial communication is that 1) a common
ground connection is set so that both devices have a common
reference point to measure voltage by. 2) one wire for the sender to
send data receiver on which is the transmitter line for the sender 3)
one wire for the receiver to send data to the sender on the receiver
line for the sender. For example suppose we have an exchange data
at rate of 9600 bps and we would like to transmit it to another device
the receiver will continually read the voltage that the sender is putting
out, and every “1/9600 th” of a second, it will interpret that voltage as
a new bit of data. If the voltage is high (+5V) it will interpret that bit of
data as a 1. If it is low (0V), it will interpret that bit of data as a 0. ​By
interpreting several bits of data over time, the receiver can get a
detailed message from the sender. at 9600 baud, for example, 1200
bytes of data can be exchanged in one second. This is how serial
communication works, while it send a lot of data it is not practical for
our needs.
■ Asynchronous communication
● In this protocol data is transferred without support of an external
clock signal like synchronous. It is good because it minimizes the
required wires and I/O pins but it comes at a cost. you must make
sure the data is transferred correctly or reaches the other end. ​one of
the most common types of Asynchronous communication
system is Bluetooth®. the benefits of using this protocol is that it
does not require synchronization of both communication devices. It is
 also cheap because this type of communication requires less
hardware. From now on we shall focus on Bluetooth® protocol.

4. Bluetooth® Protocol Standards

○ Bluetooth® is a technology that describes communication between devices
using short range radio frequency( RF) technology that operates at
2.4GHz. This technology is capable of transmitting voice and data at low
power. This is why we are using it in our project, because we want to transmit
the readings of the voltmeter to an Android application wirelessly. Bluetooth®
is one of the technologies that enables this process (it is the de facto standard
in the electrical and computing industry). This technology is safe, trustworthy
and very secure due to several layers of data encryption and user
authentication used. This is another appealing factor because it means our
data cannot be interfered with. We will be using Bluetooth® to wirelessly
synchronize and transfer data between the microcontroller attached to the
voltmeter board and the Android application.
○ Bluetooth® Profile: profiles are additional protocols that define how
Bluetooth® works. These are additional protocols built on top of the basic
available ones in the Bluetooth® standard specifications. Different devices
have different profiles. in order for two devices to communicate they must
have the same profiles otherwise they won’t work. the specification are
concerned with the use of Bluetooth® technology to support various
applications. Each profiles specification is mainly concerned with the
technology defined in the core specifications i which includes description of
which aspects of the core specifications are mandatory, optional and not
applicable. the main purpose of profile specification is to define a standard of
interoperability, so the different vendors who sell Bluetooth® enabled
components. that claim to support a given usage model will work together.
○ Bluetooth® Protocol Architecture - Bluetooth® is defined as a layered protocol
architecture consisting of core protocols, cables replacement and telephony
control and adopted protocols. It consists of five - layer stack consisting of the
following elements:
■ Radio - Specifies details of the air interface, including frequency, the use of
frequency hopping, modulation scheme, and transmit power.
■ Baseband - ​Concerned with connection establishment within a piconet,
addressing, packet format, timing, and power control.
■ Link manager protocol (LMP) - ​Responsible for link setup between
Bluetooth® devices and ongoing link management. This includes security
aspects such as authentication and encryption, plus the control and
negotiation of baseband packet sizes.
■ Logical link control and adaptation protocol (L2CAP) - ​Adapts
upper-layer protocols to the baseband layer. L2CAP provides both
connectionless and connection-oriented services.
■ Service discovery protocol (SDP​) - ​Device information, services, and the
characteristics of the services can be queried to enable the establishment
of a connection between two or more Bluetooth® devices.
■ RFCOMM is the ​cable replacement protocol included in the Bluetooth®
specification. RFCOMM presents a virtual serial port that is designed to
make replacement of cable technologies as transparent as possible. Serial
ports are one of the most common types of communications interfaces
used with computing and communications devices. Hence, RFCOMM
enables the replacement of serial port cables with the minimum of
modification of existing devices.
■ Bluetooth® specifies a ​telephony control protocol​. TCS BIN (telephony
control specification—binary) is a bit-oriented protocol that defines the call
control signaling for the establishment of speech and data calls between
Bluetooth® devices.
■ The ​adopted protocols are defined in specifications issued by other
standards-making organizations and incorporated into the overall
Bluetooth® architecture. The Bluetooth® strategy is to invent only
necessary protocols and use existing standards whenever possible. These
are the adopted protocols:
● PPP - The point-to-point protocol is an Internet standard protocol for
transporting IP datagrams over a point-to-point link.
● TCP/UDP/IP - These are the foundation protocols of the TCP/IP
protocol suite.
● OBEX - The object exchange protocol is a session-level protocol
developed by the Infrared Data Association (IrDA) for the exchange of
objects. OBEX provides functionality similar to that of HTTP, but in a
simpler fashion​.
■ Usage Models -

Figure 4.1: Bluetooth® Core Protocols

○ Serial Port Profile (SPP): This is the profile we will be using for our design,
it acts like RS-232 which sends a lot of data between two devices. It builds on
top of the serial communication mentioned earlier in that it uses less
equipment, transmits data asynchronously. Using this profile (SPP), a device
can send and receive information like RX and TX lines connected between at
the same rate. Two microcontroller can communicate back and forth
wirelessly effortly across a wide range (distance) as if they are connected with
wires. Another benefit of this protocol is that it used for integration with any
3rd party application that has SPP profile like an Android phone/tablet and an
arduino.

5. Environmental standards
○ Since our device is an electronic device. It falls under the standards of
electronic industry which includes passive components (resistors, capacitors,
inductors); semiconductor components (discretes, integrated circuits);
printed circuits boards (single and multilayer boards); and printed wiring
assemblies. Our voltmeter will include an integrated circuit (a semiconductor
device), thus it will encounter all the problems faced by integrated circuits
such as. We divide the standards of the multimeter based on the
components in the device as follows:
■ Semiconductors: are produced by treating the material with
dopants(boron and phosphorus), other materials used are silicon and
gallium arsenide. the process involved in manufacturing semiconductors
produces toxic chemicals which are harmful to the environment as well
as the user. during the manufacturing process such as crystal growth;
acid etch; doping and oxidation; diffusion and ion implantation; vapor
deposition all release chemicals which are dangerous to one's health.
thus it is necessary that we purchase and use semiconductor materials
that have gone through a proper procedure in the manufacturing
processes , the reason being simple. a material that's made from highly
quality standards is safe for both user and the environment. The biggest
problem with production process is the ​Air emissions that happens.
Chemicals that they use included the following: hydrogen, silane, arsine,
phosphine, diborane, hydrogen chloride, hydrogen fluoride,
dichlorosilane, phosphorous oxychloride, and boron tribromide, all of
which extremely dangerous. Once the production is complete, the
leftover include heavy metals, waste organic solvents etc. So the
production chain of the semiconductor is very crucial to us. when
purchasing these materials they must meet the Electronic Manufacturing
process for environmental guidelines. thus the production of semi
conductors should be carried out in a closed system.
■ Printed circuit board (PCB); these boards come in three parts, (single
sided, double sided, and multilayer boards). board production occurs
 when you produce patterns of conductive material on a non conductive
substrate (coating layer) by subtractive and additive process. the
production process releases chemicals as well like sulfuric,hydrochloric,
phosphoric, nitric and acetic; chlorine; ammonia; and organic solvent
vapors . waste produced after the manufacturing of PCBs are organic
solvent and heavy metals.On top of that, the process involves cleaning
and surface preparation of the base; electroless copper plating; pattern
printing and masking; electroplating and etching; all of which release
hazardous chemicals. therefore it is very crucial that PCB go through the
same guidelines of semiconductors. which is using chemicals or solvents
that will minimize the release of wasteful material.
○ Producers/manufactures of PCBs and semiconductors must go through
correct and safe processes. they must follow the Environmental assessment
(EA) process which are based on the country legislation in our case USA)
and the ​pollution prevention and Abatement Handbook ​as applied to local
conditions. the chemicals used must be justified to the EA which oversees
what goes into these materials and acceptable to MIGA.
○ ROHS Guide: stands for restriction of hazardous substances. these are
guidelines thats makes sure the materials used in electric components are
Lead-Free. These guidelines were set up to protect the environment and
users from inhaling harmful chemicals. anyone producing, selling, distributing
must make sure there products are compliant with the ROHS guidelines. the
specification are the following:

ROHS Restrictions
MATERIALS QUANTITY ALLOWED
Lead (Pb) <1000 ppm
Mercury (Hg) <100 ppm
Cadmium (Cd) <100 ppm
Hexavalent Chromium(Cr VI) <1000 ppm
Polybrominate Biphenyls (PBB) <1000 ppm
Polybrominated Diphenyl (DEHP) <1000 ppm
Benzy butyl phthalate (BBP) <1000 ppm
Dibutyl phthalate(DBP) <1000 ppm
Diisobutyl phthalate (DIBP <1000 ppm
Table 4.2: ROHS Restrictions.
 ROHS Categories
CATEGORIES DESCRIPTION
Large household appliance: refrigerators, washers, stoves,
1 AC
small household appliance, vacuum cleaners, hair dryers,
2 coffee makers, irons
computing & communication equipment: computer, printers,
3 copiers, phones
consumer electronics: TVs, DVD players, stereos, video
4 games
5 lighting: lamps, lighting, fixtures, light bulbs
6 power Tools: drills saws, mail guns,
7 Toys and sports Equipment videogames, electric
8 medical devices and equipment
9 Control and monitoring Equipment
10 Automatic dispensers: vending machines, ATM machines.
National Security use and military equipment
Large stationary industrial tools
spare parts for electronic equipment in the the market
before July 1, 2006.
Certain light bulbs and some batteries
Table 4.3: ROHS Categories.

○ Waste Characteristics: are the chemicals that are released once

semiconductors and PCB are no longer in the factory but are in the hands of
consumers. We will talk about the most wasteful chemicals released by parts
of the device we are making. this section will be based on the most wasteful
“types” produced by the voltmeter we are making.
■ Air emission:
■
○ WEEE: stands for waste From Electronic Equipment, they mandate or
demand that electronic products that have been discarded be recovered and
recycled. All applicable products (electronic components) must pass the
WEEE compliance and carry the “wheelie Bin” sticker.

4.1.2 Software Standards

Most of the software standard will be focusing on the communication between

the device and the Android application that will be used for displaying our
measurement output. They will be two parts to our standard, the first one will be
for the microcontroller and the second part one will be for the Android standard.
The microcontroller that we will be using is the MSP430, this embedded system
will have Bluetooth® capabilities in order for it to communicate with the Android
application. It will use Bluetooth® 2.1 which will be complete validated, certified
and production ready modules. The MSP430 will come pre integrated with
Bluetooth® software stack, it will include full hardware and software prototyping,
the software kit includes Bluetooth® software stack and serial port profile (SPP).
It shall follow the
● FCC/IC regulatory compliance
● FCC part 15 class a complaint
● IC ICES -003 Class A Complaint
● Bluetooth® core specification 2.1+

For the Android application, we want to make sure it communicates very well with
the microcontroller. That is why we shall use standards that work and
communicate with the hardware (MSP430 stack).
● Bluetooth® core specification.
● API interfacing with sensors on the microcontroller
● Transfer protocols such as serial communication

4.1.3 Impact of Standards

The impact of implementing these standards is that it will be able to make the
software on the Android interface and communicate with the microcontroller on
the multimeter. With proper communication the microcontroller should
seamlessly transmit data with the Android software and display the results. The
implementation of these standards across different platforms enables easier
communication between the two mediums.

4.2 Realistic Hardware/Application Design Constraints

As the name suggest , constraints are limitation that we have to work with. We as
engineers always have to work with what is available and work around obstacles
to achieve what we want. In our process of making the voltmeter, we shall be
work within the confinements of the electronic equipment that we will be using.
The confinements cover all the components from microcontroller to the circuits
board; as well how much time we need to o build our voltmeter. A lot of time will
be spent on the constraints of microcontrollers because of how crucial it is to
make for the success of this project. Also because of the difficulty will are
predicting when we make the connection between microcontroller and the
Android application. So it's very important that we have a full understanding of
the limitations we will be working on.
4.2.1 Microcontroller Constraints

The microcontroller is the piece that connects everything in our system. it will act
as the bridge between the integrated circuit board and the Android application.
without the microcontroller, nearly half of our project will not be able to work and
the goals set will not be reached or accomplished. However this device comes
with its own confinement that will must work with. After doing research and
looking at all the microcontrollers available, we decided to settle on the
ATmega328P microcontroller because of its great support for Bluetooth®
connectivity with other electronic components (our Android phone and custom
made application that will be used to display the output of voltmeter readings).
For example we need to know the clock speeds of the board, sampling frequency
of the data that will be transmitted to the Android phone and why it matters for
our design., these are some of the examples and constraints we have to work
with that we shall cover below.

● ATmega328P: ​Is the microcontroller that we shall be using for our project. It
is low powered device that ​operates at 5V​, it can receive input voltage in the
range of ​7-12V with capability to reach 6-20V. ​It has 14 for digital I/O pins,
6 provide PWM (pulse modular modulation) output and the other 8
pins/channel are for analog input. Most Arduino microcontrollers have an 8
channel 10 bit analog to digital converter (ADC), which is 8*10= 80 bits
transmitted total at any given time.
● An ADC: is used to convert an analog voltage that is continuously varying (
known as a signal) within a known range into a digital value . The analog
value is a representation of a real world measurement which is the voltage
measurement of our voltmeter. The digital values are collected through a
process known as ​sampling​, this is a process by which arduino board
reads/samples the signal a certain time per second which is known as the
sampling frequency. The graph below is an example of a analog signal. As
shown below the green line shows the analog signal rise and fall between -5
and 5 volts. It is a continuous line because that represents the measurement
of the voltmeter voltage. The blue dots on the graph are points in time where
the ADC is used to convert the current point of the analog signal into a digital
value. the number of times we take a reading per second is the frequency of
the sampling.
●
 Figure 4.2: Example of sampling frequency (voltage vs time).

● The ADC architecture : is made up of an 8 channel multiplexer. The

multiplexer combines the 8 analog pins into the single 10 bits ADC v.Only
one ADC operation can be carried at time. If more than one ADC (pin) is
being used, the reading/samplings are handled by the ​ADC unit​. Ths unit
handles which pin use being used , nor two pins can be used at the sample
time. if analog pin 0 is being converted, analog pin 1 needs to wait for pin 0
to finish before it can be converted and so forth until all the 8 analog pins
have been converted and then it starts all over again with pin 0. ​The AVR is
an 8 bit microcontroller. The ADC has 10 bit resolution. The result of the
ADC conversion needs to be stored in two registers. These are ADCH &
ADCL (ADC high & low).
● The good thing about this board is that it is user friendly, particularly when
mounting it to a breadboard which was a main reason why it appealed to us.
especially when working with The device will receive data from the
integrated circuit in the form of an analog signal (it has 8 analog input pins), it
will use the signal from the board and convert it to a digital signal at rate
(clock speed of 16MHz) defined by the microcontroller specifications.
●
 ●
Figure 4.3: Analog-to-Digital Converter

● The input “VIN” of the ADC must be between 0 volts and the ADC reference
voltage VREF which is default to VCC and on MCU which is 5 volts, although
this can be changed to any external voltage we desire using the AREF pin.
Vin is the measurement from voltmeter that is connected to one of the ADC
input channels. The ADC conversion formula is ​((VIN/VREF)*1023)​. if VIN=
0 volts (assuming VREF= 5V) then the converted value will be 0. if
VIN=VREF then the converted value will be 1023 which is the resolution.
● Operation Modes: ​The ADC has two operation modes. The first mode is
called the single operation mode. in this mode you have initiate each
conversion, once the conversion are done it is stored in the ADCH and
ADCL registers. You can begin this process setting the ADSC i.e ADC start
conversion but in the ADCSR. What it will do is that it will stay high while the
conversion is in progress and will automatically be cleared when the
conversion is complete. there is also an option to select which channel to
convert the signals. Once it done, you will get a notification conversion
completion event can be obtained via the ADC conversion complete interrupt
service routine (ISR). The mode is called the Free Running Mode. In this
mode we initiate the first conversion and the from there the ADC will
automatically start the next conversion. Like the single conversion mode, a
new channel conversion must be set in ADMUX before the next conversion
begins. ​be careful not to change the ADMUX just after a conversion starts.
This can lead to unspecified behaviour.
● The ADC has clock speed which is between 50 KHz to 200 KHz when 10 bit
resolution is desired. This is equivalent to 50,000 to 200,000 per second. Ay
faster than this range and the system will being to degrade which is why it
comes with a prescaler, this is a component which controls the ADC clock.
Its special counting circuit which is used to reduce high frequency electrical
 signal to lower signal using integer division. the prescaler takes the basic
timer clock frequency of the arduino and divides it by some integer value
before feeding it to the timer. The prescaler values can be configured to the
desired number. But are fixed to values of powers of 2 (2,4,8,16,32,64,128).
For example, suppose the arduino is running at 64 MHz and we want to
reduce the clock speed, we will set the prescaler to value of 4 bits, then the
clock speed will reduce to 60MHz/4 = 16. The reason why we do this is if we
want to to sample analog signals at slower rate in order to calculate the data
appropriately. they can act sytheinzier from the circuit board to arduino and
then the android application. there is possibility that we might sample signals
so fast that they are not properly displayed because they change quickly.
However the problem with undersampling is they we might not get the
correct digital replica of the signal, so we shall over sample our analog signal
to get the correct analog signals. ​We shall use Nyquist Sampling to get
the correct sampling of the signal.
● Resolution: the ADC has 10 bit resolution. This translate to values between
0 and 1023. If the reference voltage is 5 volts, this means that the smallest
detectable change in voltage on the input pin is 5V / 1023 = 0.0049 volts or
4.9 mV.
● Once the Signal is in the microcontroller, it will be converted to a digital
signal. This will happen during switching i.e. switching regulators in the
circuit that will flip from “ON” to “OFF” voltages that translate to 1 or 0.
Essentially what it does is, as the power voltage is leaving the circuit board
and about to enter the microcontroller , the voltage at terminal output is
changed to digital signal made up of 1’s and 0’s.The reason why it will be
converted to a digital signal is because this domain is easier to work with on
the microcontroller and also easy for transmission of data packets to the
android phone ( the phone does not have an analog or FM receiver/ tuner.
Current technology use Bluetooth® technology which allows transmission of
data wireless. As mentioned earlier both components need to have
Bluetooth® capabilities in order for the transmission to occur, this means that
both devices must have chip called DSP which stands for digital signal
processor. This is a special chip handles signal processing.
● Once the signal is in the MCU’s data storage area. It has to be transmitted to
android phone. The data is in located in the flash memory of the arduino in
the form of data packets (10 bits streams of data). basically they are 3 types
of memory pools in the arduino boards.
○ Flash Memory - also known as the program space, this is where the
arduino sketch is stored.
○ SRAM (static memory random access memory) - is where the sketch
creates and manipulates variables when it runs
○ EEPROM - is where the memory space that programmers use to to
store long term information.
● Flash and EEPROM memories are nonvolatile. This means that the
information stays when the power is OFF . Meanwhile SRAM is volatile and
will be lost when the power is recycled. The is not enough SRAM in the
arduino, so we as programmers have to make sure we do not use all of it.
This can happen when use long storage data variables when we storing
small values. It is important to free memory after we are done doing a
specific task so that there is room for the next instruction. When the arduino
runs out of memory the program may fail in an unexpected ways. It might
appear to run successfully but it will run strangely. To see if this is
happening, we check out the sketch and try commenting out or shortening
strings and other big data structures. If the program run successfully after
commenting out, this means we ran out of SRAM. the SRAM is too small to
handle big computations. In this case, we move or shift the data to
android/phone which has bigger memory pool and perform the calculations
there in real time. This will reduce the load on the Arduino and we won't run
out of the SRAM. Another option is to transfer the data and store it on the
arduino flash (program) memory using the ​PROGMEM​ keyword.
● Like mentioned previously the data is either in the SRAM or flash memory. At
this point we need to transfer the data (digital signal) to the android phone
and then it has be displayed in the android application we are building. We
have to find a mechanism to transfer that data and display it in human
readable form. In the arduino memory the signal is just streams of 0’s and
1’s (binary data) in packets of 4,8,16…..128. These streams of binary data
now need to be transferred through a Bluetooth® channel. we accomplish as
follows;
○ Bluetooth® serial term module to connect to the arduino (JY-MCU).
○ create a sketch (just another name for the program we writing) and
upload it to the arduino and run it there.
○ the first thing we do is set up the the connectivity between the arduino
board. We going to to pair the arduino with the android phone. we will
have a check-connectivity, a way of verifying that a connection as has
been established between the two platforms. one way of doing this is 1)
have the arduino flash the LED lights on the board, 2) display “Arduino
name” on the phone screen to show that the pairing was successful.If no
connection has been established, we shall reset the arduino
mechanically on the board and start the whole process again. the
benefits of doing such is that whatever is in the volatile memory will go
and have a fresh startup. This step is very important, if no connection
then we cannot even get the reading from the data.
○ Data Transmission Via Bluetooth® - Data transfer depends on the
protocols, these protocols are designed based on how much power the
devices produce and use. The devices arduino and phone roughly have
power consumption of (5V*46.5 mA = 232.5mW and 1 KWh
respectively). the Bluetooth® module has two mode of operation.
 Another important factor we have to consider is that the android
phone might have different baud rate to that of the phone
i. Command Mode:​ in here we send At commands to it.
ii. Data mode: This is the default mode for the Arduino.In this mode
we can transmit and receives data to another Bluetooth® module.
Based on the data sheet, at 16 MHz it transmit 1 Mbps.
● Baud rate: 9600bps of data: 8 bits, stop Bits: 1 bit, parity.
●
○ Display the Data - depending on which configuration we used (whether
we calculated the distance)

4.2.2 Economic and Time Constraints

A lot of time will be spent on making sure the Android application communicates
with the microcontroller on the voltmeter. Finding a way to make the protocols to
make the two machines to work together will be the biggest obstacle. The reason
for this is because these two machines are used for different purpose but they
can work together thanks to protocols that enables two devices to work together.
so a lot of time will be spent on finding the correct microcontroller that will
communicate with the Android device. A lot of time will be spent on making sure
the sensors work well and are able to transmit the data to the Android device and
in turn the device is able to communicate back to the microcontroller on the
voltmeter.

Another area where we will spend a lot time is on the testing of the circuitry to
make sure we are able to measure the voltage accurately and precisely; the
microcontroller and its properties like the sampling rate and frequency it will
transmit the data, If we are not able to make measurement because of the design
of the circuitry or unable to transmit the data to the Android application, then we
will have to redesign the circuitry and then retest it again until we get the correct
measurements.

Thus it is possible we might spend more money on the circuitry and

microcontroller if our initial assumptions are wrong

4.2.3 Environmental, Social, and Political Constraints

Since this is a portable device, it will be used outdoors a lot. we want the device
to be environmental friendly and usable. It should be used by for its intended
purpose and no nothing more. Since our device uses CAT 1 standards, it will be
limited to what is specified in that category which handles electrical environment
with high power available and higher energy transients.
The device shall be made with reusable material and reduce emissions of
greenhouse substances. It shall be designed in accordance to the certification
agencies which evaluate that the system(s) implemented by us is safe to both the
user and the environment environmental factors such as storage temperature,
humidity, air density and electromagnetic radiation can affect uncertainty, thus it
is imperative that the device must receive clean power.

On top of that, we expect users to use the device in the right environment and for
the right purposes or intentions. This means we do not want anybody to use this
device to harm others or modify it. We intend this device to be used by
everybody but safely.

4.2.4 Ethical, Health, and Safety Constraints

Besides designing a working multimeter, safety is also very important for our
customers. A device that is safe to use in the hands of our customers is very
important to us. We do not want any hazards nor electoral shocks happening.
This is why we have taken greater measure to make sure that our device is very
safe, stable and reliable. To do this we will be following industry standards in
accordance to our machine. We shall implement the “category 3” for our
measurement category because it is not connecting to MAINS.

● One of the hazards likely to occur is electrical shocks from the device.
Despite it being low power consumption device, there is a possibility of it
electrocuting the user. That is why we have to make sure the device has
an insular cover to prevent the user getting an electrocuted. Voltage
spikes are likely to occur in situation where a user is measuring electrical
systems that use high energy transient from transmissions or high energy
circuits. One standard that we will be using is ​“IEC 1010 standard”​. We
will use the guidelines to test against high energy circuitry.
● Transient protection is another important area against against steady
state and transient over voltage. High energy circuit with transient ride
can deliver high current voltage which can cause an explosion.
● Overvoltage installation ​is another important aspect that must be taken
care of. for our device we shall implement “CAT III”. which insure that the
low energy voltage is derived from high energy high energy circuits.
● Overload protection ​against overcurrent might be prevented when
adding high impedance of volts/ohms. It is achieved by protecting circuit
that clamps high voltage to acceptable level.

4.2.5 Manufacturability and Sustainability Constraints

As mentioned in section 2.4, this multimeter will be low powered device, that will
be reusable, use eco friendly parts. The device should last long during operation
(around 200 hours). It should be easy to assemble to make mass produce. The
device should use the best parts at low prices since it will be used different
purposes. It should be kept under correct weather conditions such as
temperature, humidity, Air. This device will be reliable, durable and affordable.
5 Hardware and Software Design Details
In this section, we will discuss in detail how our product design began, the issues
and concerns that came about, and how we came to our final design. We
discussed in section 3 how we chose the parts to be used in our product, and
here we will discuss how they will be implemented together to meet the
requirements specifications defined in section 2.

5.1 Initial Design Architecture

As discussed in section 3, there are vast amounts of information to base our

initial design on. From the basic pocket multimeters to advanced Fluke meters,
there are many paths to choose in designing our product. From the beginning of
our project, we needed to make a decision on if we had the knowledge and
experience to achieve our goals on both an electrical engineering and computer
engineering basis. To determine this, we needed an initial design to give us an
idea on where we stood. After many hours of researching various products and
designs, we decided to take a stab on the breadboard and attempt to take a
basic voltage measurement using the Arduino Uno.

5.2 Initial Voltmeter Design

Our first design was based off of the general properties of the analog-to-digital
converter used on the Arduino Uno. This component is 10-bit, which gives a
resolution of 2​10​, or 1024, discrete values. For our requirement specification to be
met, we need to be able to accurately measure voltage to the hundredth of a volt.
This component operates on the Arduino’s 5V reference voltage which gives the
voltage resolution to be 5V/1024 = 4.88mV which is sufficient for our product
requirement.

Once we had established that the microcontroller could give us the resolution we
required, we had to address what voltage we needed to measure. We decided to
strip the ends of a 12V DC wall adapter and use this for our initial measurement
test. The only issue this decision presented was the input limitations of the ADC.
An ADC, like any integrated circuit, operates within a range of input voltage. The
ADC on the Arduino Uno is contained within the microcontroller itself, the
ATmega328P, which is set to a default reference voltage of 5V. At this point,
being that this test is a only a proof of concept, we were not concerned with
being able to measure a wide range of voltage. Through the registers inside of
the ATmega328P, we can alter the reference voltage that the ADC operates on.
This will be altered in our final design to optimize the desired measurement range
and maximize the resolution of our product. Our power supply output, being 12V,
violates the input range for this component to operate correctly and must
therefore be dealt with using a voltage divider. We wanted to be able to safely
measure a range of around 30V for our initial design, so using the voltage divider
equation, we chose resistor values of 4.7kΩ for R1 and 1kΩ for R2. These
resistance values give us an ideal input range of 0-28.5V to the divider and a
0-5V input range to the ADC.

Before we could begin implementing the basic circuit, we needed to take

measurements of the components we would be using and testing. Using a store
bought Craftsman digital multimeter, we measured the voltage across our
makeshift power supply to be 12.03-12.04V. We then measured the actual
resistance of R1 and R2 to be 4739Ω and 988Ω respectively. These values are
critical for our design to operate correctly, as the conversion from digital values
out of the ADC back to the original input voltage is done through software. We
then connected the circuit on the breadboard and began writing the code using
the Arduino program.

We began our code by declaring variables such as the value read on the output
of the ADC, our resistor values, and other temporary values to be used in the
calculations. The program reads the output of the ADC and multiplies this value
by our resolution. This gives the actual voltage that is being read on the input of
the ADC. This value is then placed into the voltage divider equation using the
actual measured resistance values stated above. We then utilized the Arduino’s
serial functionality to read, calculate, and display the value on our computer’s
serial monitor. The code then hits a delay and goes on to repeat. We added a
few other measurements such as the average voltage, time elapsed, number of
samples, and maximum/minimum values.

5.2.1 Breadboard Testing

Once our program was complete, we probed the output of our power supply and
found that our barebones design was correctly reading voltage within 0.01% of
the store-bought meter. This was well within our accuracy requirement of 0.1%.
We then added a second voltage divider in series to vary our “power supply”
voltage to be measured. For the range of 0-12V, the design worked flawlessly
and within our accuracy and resolution requirement specifications. From this
initial test, we confirmed that we have the tools necessary to carry on with our
project and construct a functional voltmeter.

5.2.2 Issues and Corrective Measures

The first noted issue with our design is our range of measurement. As stated
previously, our range for measuring was 0 to 28.5V with this set up. To meet our
requirements specification of a range from 0 to 1000V, we needed to redesign
our voltage divider to allow such a range. On the same note, our second major
design flaw arose; the input impedance of our meter. Our highest concern
besides user and product safety is to not disturb or affect the circuitry being
measured in any way. We want to simply read a voltage without causing any
changes to the circuit being measured. We also want to ensure that our meter is
accurate for voltage measurements on any electrical circuitry, whether it be a
power supply or a component in a system. The input impedance of our voltmeter
is the key factor to addressing this issue.

In order for the meter to correctly measure a voltage, we want the smallest
amount of current to flow through our terminals. In order to do so, we have to
make the input impedance of our meter as large as possible. Most modern
voltmeters have an input impedance of 1MΩ (Category I and II) and upwards of
10MΩ for higher rated meters. The ADC itself has a very high impedance,
around 100MΩ on the ATmega328P, but this does not ensure the safety of the
device. To ensure the safety of both the user and our meter, we have to create a
very high impedance between the probes. We can accomplish this by adding
several large resistors in series to the input of the ADC.

To address the range of our input voltage, we need to discuss the range
switching of the voltmeter itself. Because the voltage drop being fed into the ADC
depends itself on the input voltage, we cannot simply use the same two resistors
to get an accurate measurement for both small and large voltages. To work
around this issue, we have to program the microcontroller to select the correct
path for the range of voltage in which we intend to measure. Modern meters use
complex designs between the physical dial and the PCB to select the correct
path, along with multiplexing inside the main IC. The Arduino Uno contains six
analog input pins which are directly fed into the ATmega328P which we can use
to our advantage. By wiring the circuitry into the different input pins, we can
create a network of resistors for our different voltage ranges. In our case, the
ATmega328P has internal multiplexing technology that selects the correct
channel to be used.

The next issue to be addressed is the optimization of the ADC. Using the default
reference voltage of 5V means that we can safely measure from 0 to 5V with the
analog input pins. As mentioned before, this gives a resolution of 4.89mV,
meaning our measured voltage is no more or less than this value. The problem
now lies the accuracy requirement of 10mV for our meter which is discussed
further in the next section. As stated previously, we want to be able to safely
measure a wide range of voltages, ideally up to 1000V. Using a network of
resistors for the different voltage dividers for different ranges, we can control the
inputs to the ADC to be on a much smaller range. By doing so, we can adjust the
reference voltage of the ADC to match this range and give us a much higher
resolution. This is explained in further detail in section 3.
The final and most important issues to address are the safety hazards of our
current design. Input protection to the device is absolutely critical for both the
protection of our product, but more importantly the safety of the user. Because
this test is only a proof of concept and we were only measuring 12V, protective
circuitry other than the voltage divider was not necessary. In our final design
however, the safety features implemented will be the most important and crucial
part of our project. Our safety design and components are outlined in section
3.4.6.

5.3 Updated Hardware Design

Because the initial test was prior to our research and parts selection, our second
design had to be vastly changed. Once our parts had been ordered, we drew the
schematic for our updated design which can be seen in Figure 5.1. As shown by
the schematic, we selected our nominal input impedance (R​1​) to be a single
10MΩ resistor which we knew would provide us with a high input impedance. For
the input protective circuitry, we used a total of four components. The first
component we knew our meter would need was a single large varistor. This
component would act as an open circuit during normal operation, but would
clamp down when the voltage being measured was above the varistor’s clamping
voltage, shunting the current through back through the negative test lead. To
handle this large current, we chose to use a PTC thermistor and a fast-blowing
ceramic fuse. The resistance of the thermistor would rapidly increase due to the
heat produced by large current surging through the component. The fuse at the
front of the circuit would blow once the current limit was reached to open the
circuit and protect the meter and the user. Additionally, we included a
high-power, wirewound resistor that would absorb the large amounts of energy
created by the overload current.

In our proof-of-concept design, we used the Arduino’s 5V operating voltage as

the reference voltage on our ADC, giving us a resolution of 4.89mV. This was
fine before because our voltage divider values were so low that a 10mV deviation
on the input provided a voltage drop within our resolution. Now that we have a
much larger input impedance and thus a larger range, we need to compensate
by maximizing the resolution. The ATmega328P, as mentioned previously, has
the option to set the ADC reference voltage (AREF) to any desired voltage. In
section 3 we calculated our R​2 values based on a reference voltage of 1V. In
order to maintain our range with these R​2 values, we had to set the AREF pin to
1V. We therefore connected voltage divider from VCC to AREF with R​1 and R​2 to
40kΩ and 10kΩ respectively to give a 1V input to AREF.

In this design, we had four resistors in our voltage divider network to have a
range that could potentially measure up to 10kV. Each of the resistors were
connected in parallel with a 4-pin DIP switch between the resistors and ground
as can be seen in Figure 5.1 below.

Figure 5.1: Updated Design Schematic Layout

The node in which each of the resistors were connected fed into a single analog
input. For measuring in the millivolts range, the user would not select a range
switch as the voltage would be directly fed into the single analog input.
Additionally, we had a separate 2-pin DIP switch in which both of the switches
were tied together for selecting AC or DC measurement. One of the switches
re-directed the raw input voltage to an AC rectifier circuit which would then send
a DC voltage to the resistor network to divide the voltage to a scalable input to
the ADC. The second switch connected 5V to a digital input on the
microcontroller which would set a flag in the code and jump to a separate loop
that calculates the true RMS voltage. With these switches connected together,
we could ensure the user would get an accurate measurement for both AC and
DC modes.

We also decided to include a digital LED display would be connected to four of

the digital output pins of the Arduino. Although not displayed in the schematic,
this design also included the Bluetooth® module to be connected to the RX and
TX pins of the Arduino for wireless transmission of data.

5.3.1 Discovered Issues

After further research and scrutinization of our updated design, we found several
flaws and potential issues that would hinder the quality of our product. The first
issue we discovered came upon simulating the measurement using a computer
simulation software. The design appeared to be working correctly for each of the
range selectors when only one range was selected. The problem with this design
is when none of the switches had been selected. In this case, the voltage would
be directly fed into the ADC. We saw that this design could potentially feed large
voltages into the IC which would damage the product. This led us to the
discovery of our second major flaw. Because we were using a single analog
input, the microcontroller had no way to realize which range the user had
selected. This means that the voltage computation would use the same
parameters for all ranges which would result in incorrect measurements. To fix
this issue, we realized we would need four separate switches that would need to
be engaged simultaneously with the specific range selector. This means that we
would need to use four additional I/O ports on the microcontroller.

The final range selection issue was the possibility for the user to select more than
one range switch at a time. In this design, if two or more switches were selected,
the resistors would be connected in parallel which would result in inaccurate
measurements and conversions. To remedy this, we could potentially use a
switch that only allows one button to be selected at a time. Unfortunately, this
would prevent us from implementing the option mentioned above of connecting a
second 4-pin switch for the MCU to know which range we are operating in.

The second major issue we noticed was the correlation between our potential
measurable range (up to 10kV) and the ability of the safety components to allow
this range. After further researching several models of varistors, we could not find
a component with a large enough clamping voltage to allow such a range of
measurement without activating. To fix this issue, we would have had to use
around 27 smaller varistors (365V) in series with one another! This would not be
realistic as the large number of components would take up huge amounts of
space on our PCB and increase the cost of our meter dramatically. We then
realized we had to reconsider the realistic range in which our meter could
measure without activating the varistors.

On a related note, we also realized a flaw in our input resistance. Although our
chosen 10MΩ resistor would provide enough input resistance, we had to address
that this was only the “nominal” input resistance, meaning that the safety
components would inherently add to the total resistance as seen from the input.
This would then affect the calculated voltage while also changing the actual
measurement ranges for each resistor in the divider network.

The next issue arose with our AC rectification design. In this design, our AC
voltage would be fully rectified and then converted to a DC voltage using a
capacitor across the output. As the rectification would occur before the voltage
division, we realized that the capacitor used at the output would need to be
specific to the range of voltage it would be rectifying. This means that in addition
to our range selector switches, we would need to implement a second network of
capacitors for each range. With this new roadblock, along with the necessary
addition of the previously mentioned 4-pin DIP switch for the MCU to be able to
calculate measurements correctly, we quickly realized how many additional
switches we would need to implement this design. While brainstorming options to
fix this issue, we discovered another major design flaw with the AC rectification.
One of the most important features of our product is to allow the user to view AC
waveforms on the smartphone application after being wirelessly transmitted from
the microcontroller. For the application to be able to plot this data, it has to know
the frequency of the signal. In this design, we fully rectify the signal before being
input to the ADC so the microcontroller had no way to determine the frequency of
the measured signal. As a possible corrective measure for this flaw, we found
that the Arduino IDE has a function that can measure input frequencies using
one of the PWM pins on the microcontroller. If we were to utilize this feature, we
would have to tap off of the input to the ADC which could possibly alter the
electrical characteristics of the signal being measured due to current leakage and
noise on the I/O pins.

The final largest issue to this design was the use of the Arduino platform. We had
completed this design while still under the assumption that we would be allowed
to use a premade development board from Arduino. After further investigation,
we discovered that we are not permitted to use such products as we have to
implement all of the hardware ourselves. This meant that we would now have to
design and implement the input power supply and voltage regulation, along with
the crystal oscillator and programming circuitry.

5.4 Final Hardware Design

After taking into consideration the multiple design flaws of our updated design,
we came to our final design which would correct all of the aforementioned issues.
To begin this section, we will discuss the measures taken to address the
measurable range, the range selection process, and how these measures
contributed to the re-designed safety features. We will then discuss our solutions
for AC rectification issues and end the section with additions that needed to be
made for the purpose of powering the MCU without an Arduino development
board.

The first issue we had to correct was the measureable range of our meter. We
realized that because our meter is only intended to be used for low voltage
applications (<1000V), we did not need a range above this value. We decided to
remove the 1kΩ divider which left us with only three resistors in our network. This
action now meant that the largest measurable range was 100-1000V. To address
the issue of a potentially large voltage being directly fed into the ADC, we had to
reconsider how we selected our range. Our first idea was to implement a
5-position switch between the nominal 10MΩ resistor and each of the range
switches. In the top position, the circuit would be open and no voltage would be
fed into the ADC. With this design, we would now have to use a separate analog
input for each range. While this seemed to be a disadvantage at first, we quickly
realized that this solved a few of the other discovered issues. By using a
separate analog input for each range, we could now write code that would detect
which input was being used. Therefore, we can eliminate the need for a second
row of switches to tell the MCU which range we had selected as it can now be
sensed automatically. Additionally, with this design, we can write a loop that
checks to see if more than one of the analog inputs is detecting a voltage. If more
than one input has a voltage, as would be the case when more than one range
selection switch is enabled, we can send the user a visual or audible warning
using either an LED or a piezo speaker.

After we had decided this was the route we would take for the range selection
process, we began to search for the 5-position switch that could realize our
design. What we quickly realized was that these types of switches were very
large and not intended to be used for PCB applications. After once again
reconsidering our switching options, we found that the 4-pin DIP switch from our
previous design could still be used. We found that by placing the switches in
between the 10MΩ resistor and each of the dividing resistors, we could now have
the entire circuit open without the need of a fifth position path. This is also due to
the fact that we eliminated the 10kV range and resistor. The final switching
circuitry is shown in section 6.

With the range selection sorted out, we can move onto the issues related to the
protective circuitry components. Now that we have decided our measurable
voltage range would be a maximum of 1000V, we can address the necessary
components to protect from high voltage and current overloads. The first
component we will address are the varistors. We found varistors rated with a
clamping voltage of 340V. By connecting three of these components in series,
we can create a total clamping voltage of 340*3 = 1020V. This will work perfect
for our design as is it will conform to the Category III meter rating, along with
minimizing the cost of the meter. Additionally, this will save much more space in
comparison to the previous option of nearly 27 varistors. The thermistor
implemented in the second design will remain for this design as it is necessary
for protection against high amounts of incoming current. Additionally, the
wirewound resistor will remain in this design in order to dissipate the large
amounts of energy that can be produced if connected to high voltage and high
power supplies. The only remaining change in safety components between our
previous design and this design is the elimination of the HRC fuse. With the
thermistor and high power resistor in place , there is no need for the fuse as the
PTC will take care of essentially opening the circuit in the case of a high voltage
transient.

Now that we have corrected all of the potential issues with the range selection
and safety components, we can discuss the issues associated with AC
measurement. After considering the several flaws in our second design, we came
up with a few possible options to take care of these issues. We knew that we had
to somehow measure the frequency of the incoming signal in order to be plotted
in the smartphone application. Like we mentioned previously, we could
potentially tap off of the input before the rectification and voltage division, but this
would mean that the peak of the AC signal must be at or below 5V, which is the
maximum input to the digital pin that would count the pulses. This means that we
would have to first divide the voltage before feeding the input into both the ADC
and the digital pin for frequency measurement. Unfortunately, for the digital pin to
recognize the upper portion of the waveform, it must be at least 3.3V which is
above the maximum output of the voltage dividers (1V). After exhausting all
options for this dilemma, we concluded that we would not be able to both rectify
the signal and measure frequency at the same time. As measuring the frequency
of our signal is detrimental to the application of our product, we decided to knick
the idea of fully rectifying the signal and converting it into a DC signal.

After making this decision, we came up with a few other possible designs. The
first option we considered was to not rectify the AC signal in any manner.
Instead, the waveform would first be divided using the same resistor network for
DC measurement before entering the ADC. Unfortunately, the ADC can only
measure from zero to its reference voltage, which is 1V. This means that the
negative cycle of the signal would not be read by the ADC. We would therefore
have to add a DC offset to the signal to place the entire signal within the ADC’s
measurable input range. This would mean that the peak-to-peak voltage of the
divided signal must be a maximum of 1V, which in turn means that we would
have to apply a 0.5V offset to fully place the signal between zero and 1V
considering the maximum of each range. In this option, the range of
measurement for each range would be cut in half. For example, if we are
measuring in the millivolts range, the amplitude of the signal would be limited to
only 500mV (0.5V) instead of 1000mV (1V) because of the added offset. This is
the same scenario for each of the ranges. The 1-10V DC range would now
become 0.5-5V. The 10-100V DC range would now become 5-50V and so on.
This now cuts the overall limit of AC measurement in half, which is still a flaw in
the design, but it could still work for our project. We spent a large amount of time
researching how to implement the DC offset, and after exhausting all designs, we
decided this was not the way to go. The first issue came about when attempting
to add the offset using a voltage divider where the input is +5VDC and the output
is 0.5V. In order to not draw any current away from the signal, the resistors used
in this divider have to be very large. The “R​2​“ resistor is connected to ground on
the other end, and therefore changes the resistance of the first “R​2​“ in the input’s
voltage divider because they are now effectively connected in parallel. Even
using upwards of 100MΩ for the offset resistors affects the input divider. To fix
this problem, we considered trying to separate the measurement ground from the
logic ground. Unfortunately, we found that the ADC must use the measurement
ground as the logic ground for the ADC to operate correctly. Knowing this, we
then knew we would have to re-calculate the ranges of input that we could
measure.

Because we are feeding four different analog inputs, we would need four “offset
dividers”. Additionally, before the offset is added to the signal, we have to remove
any previous DC offset from the incoming signal using a capacitor so that our
measurements will be correct. Unfortunately, if we were to connect a capacitor
before being fed into the ADC, then DC measurement mode would be completely
eliminated! This means that we would have to connect another set of
two-position switches that would have to be selected along with selecting the
range, one position feeding directly into the ADC, and the other being passed
through the capacitor and offset circuitry before being fed back to the ADC. We
thought of possibly using separate nets for the AC measurement, but because
our microcontroller has only 6 analog inputs, four of which are used for DC
measurement, we would only have two selectable ranges for AC measurement.

This design option had quickly snowballed into a large mess of switches and
paths that the user would need to select before even taking a single
measurement. Although we could spend large amounts of time scrutinizing our
design and optimizing it for accuracy and usability, we had to accept that the
physical implementation needed for this would require far more time than we
had. It would have been a huge advantage for users to be able to see the entirety
of the waveform exactly as it is seen from the ends of each probe, but we had to
acknowledge the lack of time we had to design all of the physical characteristics
for this to work. After reaching this decision, we had to move onto our second
option which is to simply divide the voltage and feed the AC signal directly into
the ADC. Although the ADC cannot convert a negative input, it produces an
output of zero. Because the incoming signal voltage has been divided already,
the maximum, or technically minimum, negative voltage drop would be -1V. This
voltage is small enough that it will not harm the ADC nor affect its operation in
any way. Then by using strategic programming techniques, we can take control
the ADC to take a very short sample at its maximum sampling frequency to
collect a quick snippet of the signal; ideally 3-5 cycles. Each value in this sample
would be stored in temporary variables which we can access and manipulate.
Because the ADC will output zero during the negative cycle of the signal, several
strings of zeros will be created and stored. We can then create a loop that
narrows down all of the positive values between these strings and essentially
recreate the signal. From this data, we can find the peak value of the signal and
then compute the true RMS voltage. Additionally, by counting how many strings
of zeros are present when the sample was taken at a known sampling frequency
and predetermined length of time, we can calculate the frequency of the signal.
While this method will surely prove to be a very challenging, we knew that it was
possible and that we had to make it work as it was the last remaining option. The
only hardware left needed to implement this method is a single pushbutton switch
that must be pressed to tell the microcontroller which mode we would like to
operate in.

Once we knew that we had to implement all of the circuitry that is used to power
and control the microcontroller, we quickly ordered the necessary parts and took
them into consideration for our final design. The components necessary are the
LM7805 5V regulator, two 10μF decoupling capacitors, a 16MHz crystal
oscillator, and two additional 22pF capacitors for the crystal. We ordered these
components, along with a P-channel MOSFET that would be used to protect the
battery from accidental polarity inversion. Also necessary for correct operation,
we implemented a pushbutton switch that is connected to the reset pin of the
microcontroller. This will allow us to reset the IC when loading new programs.
The next change that we made was replacing the LED display with an LCD. As
we have enough I/O pins to spare, we knew that this type of display would be
better for our design. For example, we can now display multiple values of
measurement such as maximum, minimum, and average voltage when in DC
measurement mode. When in AC mode, we can now display to the user the
frequency, peak voltage, and RMS voltage.

Last but not least, we included four LEDs. A green LED will be connected to the
power supply to indicate when the device has been powered on. A red LED will
be illuminated when more than one range switch has been selected as this would
result in miscalculation. The final two LEDs are blue and will be to indicate which
mode of measurement the user is operating in, either AC or DC. Each of tehse
four LEDs will need a 220Ω resistor to limit the current flowing through them.
5.5 Application and Final Microcontroller Design

5.5.1 Application

In designing an Android application, guidelines and methods for design,

development, and distribution are supplied by the official Android development
resources available online. The application will take advantage of the established
standard that Android has set in terms of operating system updates. The
application is designed to work on 85+% of phones operating on the Android
operating system to maximize utility and compatibility. The application should be
able to run on multiple devices of varying screen sizes and screen resolutions
that encompass the vast majority of Android handsets. Android has performed
many changes to its operating system since its inception. One of the more drastic
changes is the shift in the graphics of the entire operating system towards a
“material design” theme, completely overhauling the home screens, menus,
icons, etc. Material design as summarized on the Android developers website is
“a new design metaphor inspired by paper and ink that provides a reassuring
sense of tactility”. The application will be developed to attempt and recreate this
“material design” visual style in order to be inclusive of the Android operating
system’s visuals.

The development of an Android application is a fairly rigorous process due to the

nature of applications, the scope of operation that applications could potentially
cover, and the process at which applications are created themselves. The
development tool Android Studio combines 2 different types files, .java and .xml
in order to create the visual and logical appearance and behavior of the different
activities within an Android application. Activities are basically commands or
changes the user indicates the application make based on the input provided.
Some activities are default within the Android operating software such as home,
back, and recent apps. These activities will continue to operate as normal for this
application, however all other activities will be required to be created within the
application.

Activities within Android will change based on altering the position of the screen.
This poses a design concern because a change of activity restarts the creation
process in that activity within the application, resetting all previously calculated
and presented data. The application will be able to maintain functionality in both
landscape and portrait without resetting data so the user can monitor voltage
trends continuously. The user will have the option of resetting the data presented
to analyze new data, otherwise the previous data will be stored and visually
supplied by graph or numerical value.
During the original design phases of the Android
application (See right image) the intent was to display
varying categories of information including voltage,
current, and resistance using both a graphical and
numerical representation which would be able to be
changed based on the user's preferences. This
original design would give the user the option at the
home screen to choose the desired measurement
required.

As the design process furthered, the project had been

modified to allow for only voltage to be a
measurement possibility, this means that current and
resistance would be removed from original application
design shown above and replaced with a similar
menu, however this menu would only allow for AC
(Alternating Current) or DC (Direct Current)
measurements. Because AC can be represented clearly on a graph depending
on the circuit measured, the application will plot the data points measured by the
voltmeter and display them in a way similar to an oscilloscope as pictured in the
image to the left. (Please note that the image shown is representing a common
sin wave as a form of example of the final applications behaviors). The
application will also allow the users options of data representation, either the
graphical representation as the example shows or a numerical value. Both of
these representations will be able to be output or saved so the user can keep the
measured values as record for further analysis or reporting. This action will we be
done from an external source, i.e the physical voltmeter, and the intent will be
transmitted wirelessly in order to trigger the data recording event.

As design of the overall voltmeter has advanced, the application has undergone
even more slight changes in order to ensure optimum efficiency of both the
hardware and software resources involved. As previously stated in the design
section of this document, the voltmeter will physically allow changing of the
measurement values of the given circuit. The voltmeter will also allow the user to
physically switch between AC and DC voltage measurement. The application will
be able to recognize these hardware changes by setting flagged values on the
Arduino microcontroller and sending the values to the connected mobile phone.
The communication between the application and the voltmeter will be
client-server relationship, where the client is the Arduino microcontroller and is
managing the measurements done on the circuit and calculating said data into
digital values that can be sent wirelessly. The server of this relationship is the
mobile device where the application is loaded onto it. The application will read
the incoming data and do basic calculations in order to present the user with the
appropriate values based on the settings selected by the user, i.e Current type
and numerical place value. The AC measurement will present the measured data
using the android GraphView library.

The android GraphView library is a third party library specifically created to allow
datasets to be represented graphically without manipulating existing android
libraries. The graphview library will be imported to the project and the data points
will have to be set dynamically by the values received from the voltmeter. The
application will also be able to read the number value whether it is in millivolts,
volts, etc. and change the graph headers accordingly.

Below is an image of the final software design tree representing the logical paths
taken throughout the use of the application:

Figure 5.2: Final Software Design Tree

5.5.2 Microcontroller

The first part of this section deals with the flow and structure of the program. The
second part of this section deals with internal flow of the code in the MCU, we
shall pay more attention to the MCU control structure since it very vital because
we need display what we are measurment.

Section 1: this is how the program will go from once the the voltage has been
measured. ​in the diagram below, the measurement of of voltage is
represented as sinusoidal wave for simplicity and visual representation​.
Now with that said let’s talk about how this will work. The voltmeter will measure
the voltage from a source and then at the same time it will convert the voltage it
to a digital signal in conjunction with the arduino. The arduino will perform the
necessary computation that we need and want. When it has done this it will
display them on the LCD then transfer the new calculated information to an
android smartphone via bluetooth

Figure 5.3: Conceptual Flow of the Program

Section 2​: In this section we will go more in depth and talk about the code
design for the microcontroller. The algorithm design is the piece that connect
everything in our project. It is the network that bridges the electric circuitry, the
arduino, and the android application to make a fully functional wireless android
application.
● Power On : This is where everything begins, we power the circuit and the
arduino, as soon as the system is running, we should get an indication that it
is ​“ON” when microcontroller starts blinking. When the system does not blink,
we should recheck everything on the board. One way of doing this is to turn
ON/OFF to reboot the arduino MCU. The reason why we are verifying that the
MCU is on is so that the code is running smoothly on a fully functional and
connected integrated circuit.
● Declare Variables and set up Flag ​- This is inside the program the code, it
is in the first part of the code before we begin anything. We declare the
variables we are going to need and use. These include the following: vmax,
vmin,vrms, vin,vout etc. We are going to include flags in our code, they are
 meant to raise signal when an event occurs, for example suppose we
measuring voltage but we are reading more than one input and all of them are
greater than 0, this will activate or raise a flag notifying us to remeasure the
voltages to make sure we are performing the correct tasks. Flags are meant
to check if something is wrong without crushing the code. they are like a test
condition and also used to store the state of the program .
● AC/DC : This is the part of the code where we check whether measuring DC
or AC. Basically we are going to have the program run infinitely and check to
see which mode we are in. Once we have determine which mode we are
working with, Read in the 4 analog inputs.. We check to see if more than
more two analog readings have measurements of more greater than 0. This
is where we activate the red flags if at least to analog input are greater than
zero. Then we have to reread again until we get the correct measurement.
● Voltage range ; ​after we have checked the correct input voltage readings,
the next step is to voltage the voltage ranges that the voltmeter is measuring.
These ranges have not yet been determined. basically the circuit will control
the voltages ranges then feed them to the program. The program will look at
the recieved the input voltage and send it to the correct switch statement. The
ranges are stored in arrays in the arduino memory which faces the ADC. This
is when sample the analog signal to digital in order for it to be stored, we can
only store the signal if it is digital format which is a sequence of bits. We have
not yet decided the mechanism of how we are going to achieve that . we are
still in the process of determining where exactly we are going to sample the
the signal. but for now we assume, we are going to sample before we store.
We are going to sample according the Nyquist sampling formula because we
have to get the most accurate replication of the voltage signal in digital form.
● Calculation ; This is where we do calculation of everything we are going to
need and display concerning the measurement of voltage. We are going to
use the resolution formula to convert the voltage to the value being used.
We are going to calculate the min, max, and averages and use take those
and display them on the LCD on the board. After they have been displayed on
the LCD we then Transmit the calculation we did on the Arduino board to to
the android smartphone through bluetooth. once on the data is on the phone
it get taken care of by the software on the bluetooth.
 Figure 5.4: Code Design FlowChart

5.6 Summary of Application and Microcontroller Design

The Android application will be primarily a monitoring application mimicking the

behaviors of an oscilloscope. The application will receive information transmitted
through the Bluetooth® module of the voltmeter and adjust its data output
accordingly. The options of data to be output will be the current type of the
measurement wanted such as alternating current (AC) or direct current (DC), the
measurement the values are taken at whether they are millivolts, volts, etc. and
the actual values measured from the voltmeter which can be plotted on a graph
for AC measurement or shown numerically for the DC measurement. The
android application will be able to function on most of the android software
versions in use today from the API 19, Kit Kat 4.4, and upwards. This application
will also reach all requirements to be published as an official application on
Google’s Play Store. The application designing process, much like most aspects
of this project has required fluidity in terms of final design and overall capabilities.
The application has changed quite a bit since the original creation to combat the
adapting nature of the project. The final design
of the home screen of the application can be
seen to the right, only allowing the user to
choose between AC or DC voltage. If AC
voltage is chosen, a graphical representation of
the voltage will be given by default, however the
user can choose to display that or the numerical
representation by using the radio buttons
marking the distinct screens. The numerical
representation will look similar to that of the DC
measurement, showing the current value being
measured while also keeping track of potentially
crucial information to the user such as the
minimum voltage, maximum voltage, average
voltage, etc. In the DC measurement ideally the
voltages will not fluctuate rapidly in order to
maintain accurate results. There will not be a
graphical representation option for the DC
measurement as the voltage should remain
constant or very close to constant. Forcing a
graphical representation in order to present a
straight line graphed at a specific value could
require more resources such as processing on
both the voltmeter and mobile phone hardware which is detrimental to the overall
operations of the system in place. Most of the data required for data
representation should sent from the actual voltmeter hardware such as degree of
voltage measured and potentially what measurement is being used such as AC
or DC modes. The application will simply interpret the data sent over the
Bluetooth® transmission and decode that data in a user friendly manner.

For the microcontroller,

Basically the code for microcontrollers is designed to handle the analog input
voltage that is being read into the integrated circuit. What the code does is to first
check to see if the arduino is properly working by looking. Then we declare the
variables that need to initialized like the baud rate, pinmode and such. when that
is done. we have while loop that checks to see if it is DC or AC , depending on
which path it takes, from there it reads in the voltage ranges then converts them
to digital signal which are then taken for calculation of the values we want. These
values are displayed on an LCD then transmitted to a bluetooth.

6 Prototype Construction and Schematics

In this section, we will cover the plan for breadboard testing and how this will be
translated to the PCB. The schematics for each step will be presented and we
will briefly discuss the plans for PCB layout. We will finish the section by
selecting a vendor that will construct our design. In section 7, we will outline the
specific part testing and results of each test and cover the strategic test
procedures for the breadboard and PCB.

6.1 Breadboard Prototype Construction Plan

Once each of the components have been individually tested to ensure each is
operating as expected, we will begin implementing the design using a
breadboard and the Arduino IDE. The full and final design schematic is shown in
Figure 6.6. This schematic shows all of the components and features that we
plan to implement. For the sake of breadboard testing, we will begin prototype
construction by assembling the microcontroller and supportive hardware
components as shown by the schematic in Figure 6.1 below.
 Figure 6.1: Prototype Construction - Step One

Although the power indication LED is not necessary, we will implement it in this
portion of the design as it will be a useful indication that power has been
connected correctly. Once the breadboard connections have been made, we will
conduct the various tests outlined in section 7.4. These tests will ensure that we
have full operation of the microcontroller and all features are working such as
serial communication and programmability. Once all tests have been completed
for the standalone microcontroller, we will implement the entirety of the input
circuitry as shown in Figure 6.2 below. In this setup, we will be testing the analog
inputs by measuring a wide range of AC and DC voltages while recording the
voltage drop on R2 through R7. Because the AC/DC selector switch will not be
installed at this step, we will have to manually switch between AC and DC modes
within software and ensure each mode is fully operational (Excluding wireless
transmission). We will compare these values and adjust the potentiometers to
create the exact resistance values calculated in section 3.5.7. To complete this
test, we will need a power supply that can generate large voltages such as the
ones found in the engineering labs at UCF. Once we have found the correct
resistance values, we will write a raw microcontroller code to read the analog
inputs as we test all ranges and feed the ADC outputs and calculated voltage
values back to the Arduino IDE through serial communication. Once we are
confident that the ADC is operating correctly for all ranges, we will connect the
LCD and attempt to write and display the calculated values.

Figure 6.2: Prototype Construction - Step Two

*Note: Power supply not pictured

The LCD connection shown in Figure 6.3 below will be made using the same
breadboard and connections shown in the previous steps. Once the LCD has
been connected, we will workshop the previous code to include Arduino’s Liquid
Crystal library and write the measured voltage values to each of the lines on the
LCD and ensure the values match the measured values and voltages.
 Figure 6.3: Prototype Construction - Step Three
*Note: Power supply not pictured

As the LCD, along with the pull-up resistors and AREF voltage divider are
connected to the microcontroller’s ground, we know that this will affect our input
voltage divider and thus measured values. That being said, we will repeat the
measurement tests completed in step 2 (Specific test procedure is outlined in
section 7.4). By slowly adjusting the potentiometers and subsequent variables in
the microcontroller code, we will continuously measure and revise our design
through the trim pots to match the accuracy achieved in the previous step. Our
margin of error at this point is to be within 1% accuracy of the digital multimeter
used in the lab. Once we have achieved this goal and are able to display each of
the calculated voltages, we will work to optimize the code before attempting to
implement the Bluetooth® module and the remaining switches and LEDs.

The next step in the prototype construction will be to include the AC/DC selector
switch, AC/DC indicator LEDs, and range selection warning LED as shown in
Figure 6.4 below. We will then include the necessary code modifications to
incorporate these new variables. The inclusion of these components will
inherently contribute to the overall current draw and noise presented on the pins
of the microcontroller. Knowing this, we will continue to repeat the thorough
measurement tests from step 2 and make adjustments accordingly.
 Figure 6.4: Prototype Construction - Step Four
*Note: Power supply not pictured

After we have made the above connections and are satisfied with all features
tested thus far, we can connect the Bluetooth® module as shown in Figure 6.5
below. After making the necessary connections, we will attempt to send the
voltage values to the S2 Terminal application found in the Google Play Store. In
this application, we will record the hex values received and ensure they match
the values sent from the microcontroller and thus our measurement values.

Figure 6.5: Prototype Construction - Step Five

*Note: The remaining components are not shown, but will be connected as in the previous steps

The final step in the prototype construction will be to finalize the smartphone
application and make a connection to our Bluetooth® module. As presented in
the following section, we will conduct a series of tests to ensure the full
functionality of the application. If all goes as planned, we will follow a strict testing
regimen that covers all features of our product and all possible scenarios of
measurement. Once we are satisfied with the results of these tests and our
prototype is fully functional, we will begin laying the plans for PCB and package
construction.

At this point in the prototype construction, we should still be within the 1% margin
of error for our measurement values. Should we not be able to meet this
requirement, we could potentially need to make changes to the input resistance
values by implementing various potentiometers. Once this requirement has been
met, we will begin optimizing the microcontroller code and application parameters
to meet our final requirement specification of 0.1% accuracy.

Figure 6.6: Final Schematic

6.2 PCB Vendor and Assembly

After much deliberation pertaining to the various PCB assembly companies

available to order from, the PCB vendor will be ExpressPCB.com. We decided to
go with this vendor because they provide free software to design the PCB and
quickly upload this design to their manufacturing division. This means the PCB
design process should be completed in a relatively timely fashion and multiple
PCBs will be ordered in order to combat any potential defects in specific PCBs.
This vendor was also recommended by previous engineering students.

For the transition from breadboard to PCB, we will use multiple resources when
designing the layout for optimal design. Once the PCB has been designed, we
will connect and test the components in the same order as done for the prototype
breadboard construction. We plan to use high quality solder and complete all
construction steps in a large room using extractor fans for ventilation purposes.
The construction steps for the breadboard prototype will be the same steps taken
when constructing the PCB. Additionally, at each step of the PCB assembly, we
will follow the strict testing procedures completed in the prototype testing so that
we may narrow down any issues quickly.

Because the PCB will be designed in a compact fashion, the measurement

components and microcontroller will be in close proximity. This inherently
presents multiple safety concerns for our product which must be addressed when
designing the layout. As we mentioned in section 3.4, our product will have to
take all possible hazards into consideration throughout the design process.
Through our extensive research we selected the electrical components to protect
both the meter and user from these hazards, but the safety measures cannot
stop at the component level. For added protection, we will design our PCB to
include isolation slots between high-voltage components and under the varistors
as well. Although unlikely, there is the possibility for large voltages to arc over the
electrical components in our meter which can compromise their functions which
could present dangerous and unsafe conditions. We have already taken this
potential hazard into consideration when we chose to use three lower voltage
varistors instead of a single large one. By expanding the spacing between these
components with the addition of isolation slots, the risk of high-voltage arc is
greatly reduced.

Other factors to consider when designing our PCB layout are reducing
electromagnetic interference, reducing noise on the sensitive I/O pins, and
possible undesired electrical effects caused from high voltage inputs. When
measuring large voltages, our product could potentially experience RF coupling
between components and wire traces if not properly designed. This potential
effect is generally caused through inductance on high current traces which
produce voltage drops that radiate throughout the PCB. Though this will be kept
in mind when designing the layout, measures such as the large nominal input
impedance that will limit the current draw into the meter will help in preventing
this undesired effect. As we have spent the majority of our project researching
and designing the meter for correct operation, we have not devoted much time
into the research for the PCB design. Though we have a general idea of the
parameters which must be considered when designing the layout, we still have
much to learn in Senior Design 2.
7 Prototype Testing Plan

7.1 Hardware Test Environment

Overall the voltmeter designed will be tested vigorously in order to assure not
only precision and usability, but also safety due to the possibility of high voltage
and/or current measurement by the user. Testing the voltmeter will include the
creation of various test circuits with predetermined voltages which are measured
using existing and calibrated multimeters. These measurements will be
compared to the digital multimeter in order to ensure quality and precision is
maintained consistently. 10 different predetermined circuits will be tested on both
the digital multimeter and the existing multimeter and compared. Safety is also a
major concern due to the nature of the use of the voltmeter. This meter may be
coming into contact with high voltage circuitry and in an instance where a circuit
is compromised or faulty which could be potentially dangerous to a user, the
meter must be able to withstand disabled operation and the transfer of electricity
to the unit via the safety components in the circuitry of the meter.

This will be tested by introducing a voltage greater than the rated voltage of the
digital voltmeter to ensure proper functionality of the safety precautionary
systems put in place. The meter will be first tested in controlled laboratory
environments in which temperature, humidity, and other external conditions are
minimized in order to ensure proper functionality. Based on the portable nature of
the voltmeter, after successfully passing tests conducted in the controlled
environment the product will be given similar circuitry to measure while in areas
of its likely use, such as outdoors, constrained environments, etc.

7.2 Individual Component Testing

To ensure that our meter would operate correctly, we had to be sure that all
components used were in working order. In this section, we will describe how
each part was tested and the results of each test. These components will be
tested in the order shown below.

1) Resistors, capacitors, test leads and switches

2) Safety Components
3) Battery and voltage regulator
4) Diodes
5) MCU and crystal
6) LCD and Bluetooth® Module
By testing the components in this order, we can ensure that each component is
operating efficiently. Additionally, we will be able to determine any possible
issues quickly as each step uses the components preceding them.

7.2.1 Resistors, capacitors, test leads and switches

The first components we decided to test were the resistors to be used in the final
design. These components are crucial to the success of our multimeter, so we
needed to ensure that the actual resistances were within the tolerance ranges for
each specific part. The list of all resistors to be used in the final construction are
listed in Table 7.1 below.

Table 7.1: Resistor Testing

Now that we have confirmed the resistors that will be used for the meter are
within their designed tolerances, we can move on to testing the capacitors.
Similarly to the above process, we tested each of the capacitors using the
capacitance mode of our store-bought meter. The test results can be seen in
Table 7.2 below.
 Table 7.2: Capacitor Values

As you can see from the table above, the capacitors are operating as expected.
The next section of testing involves continuity tests for the test leads and
switches. We began by first testing the resistance of each test lead from the
probe to the banana jack. Both of the probes had 2Ω resistance which is
satisfactory for our design. We then performed the same test on the
range-selector DIP switch, the power supply switch, and both push buttons. Each
of the contacts were 5Ω or lower which indicates all switches are operating
correctly.

7.2.2 Safety Components

To ensure all of the safety components were in working order, we had to perform
a variety of tests for each application. Unfortunately, some of the components
require large voltages and currents to test their functionality. These tests would
require high-voltage power supplies and would present multiple safety hazards.
Because this is only a proof of concept, we will have to trust the manufacturer’s
specifications and assume their values are correct. However when the time
comes to build the full meter and test its functionality, we will need a power
supply that can produce these large voltages. If we are unable to obtain a power
supply, we can construct a makeshift supply that amplifies and rectifies 120VAC
from a wall outlet. The AC signal would first pass through a 100μF capacitor
which will hold the peak of the 120V input which is 170V. The first diode prevents
the capacitor from discharging to ground which allows the voltage to be constant
at 170V. Similarly, on each positive cycle of the signal, the output capacitor is
charged to the peak voltage plus the first capacitor’s voltage which produces a
340V DC output. The schematic for this circuit is shown in Figure 7.1 below.
 Figure 7.1: Varistor Testing Power Supply

As mentioned in section 3.4, we chose to place three 340V varistors in series to

give us a maximum clamping voltage of 1020V. We can use this power supply to
produce 340V and test each of the three varistors. By placing the varistor in
parallel with the capacitor, the current should be shunted through the varistor
which we can measure with a store-bought ammeter.

Similarly to the varistor test, in order to test the high-wattage resistor which would
absorb the large amounts of energy produced by the varistors shunting the
current back through the negative lead, we require a power supply that can
produce a large current. While we have access to a power supply which can
potentially produce this large of a current, the safety hazards again take
precedent and we will not be able to confirm the actual power dissipation
parameters of this component. On the other hand, when it comes to the PTC
thermistor, we can connect this component to a power supply, raise the current
passing through the component and measure the resistance across the leads.
While this is safe in theory, in the event that the component is not to specification
or is possibly shorted, there is a chance of shorting our power supply and
damaging the equipment. Fortunately, we can perform a makeshift test by
connecting the thermistor to a store-bought meter and measure the change in
resistance as we hold the component increasingly closer to an open flame. We
performed this test and noted that the resistance at room temperature was
38.4Ω. After holding the thermistor close to an open flame for a few seconds, we
saw that the resistance increased to over 200Ω. This confirmed that the
component was operating as expected, although we cannot determine the limits
of operation.

7.2.3 Battery and Voltage Regulator

When testing the battery, all that we needed to do is connect the probes of our
store-bought meter across the positive and negative terminals to measure the
voltage. Doing so, we measure our battery to actually be 10.02V out of the
package. Although this is larger than expected, our voltage regulator should be
able to handle input voltage from 7-16V. This led us to the testing of our voltage
regulator. Using a breadboard, our two 10μF decoupling capacitors and our 9V
battery, we connected the 7805 voltage regulator in the circuit shown in Figure
7.2 below.

Figure 7.2: Power Supply Circuit

We measured the voltage on the output of the regulator to be 5.02V which is

sufficient for our design. To test the limits of this component, we connected a
power supply to the input and varied the voltage from 12V down to 4V. At 4.9V,
the output of the regulator dropped below 5V and followed the input. This
confirmed that the component was operating as expected and would work for our
device.

7.2.4 Diodes

The next step in our individual parts testing was to test the diodes that we
ordered for our project. As outlined in our final design, we eliminated the need for
a diode for AC rectification as proposed in section 5. The diode that would have
been used for power supply protection was replaced with a P-channel MOSFET.
Although we still plan to use this component, we wanted to confirm the
functionality of the 1N5818 Schottky diode that we would use as an alternative.
To test the rectifying diode, we connected a power supply to the circuit as shown
in Figure 7.3 below. Two tests will be performed; the first will be a DC sweep
from -12V to 12V with R1= 1kΩ and the second will be to change the value of R1
from 100Ω to 10kΩ at a constant Vin. The first test will be determine if the diode
is biasing correctly, and the second will vary the amount of current passing
through the circuit and will show us the range of voltage drop depending on the
load. We will use a digital multimeter to measure the voltage at Vout. From -12V
to 0V there should be no current flowing through the diode as it will not be
forward biased. Above zero volts to 12V the diode should be forward biased and
allow current to flow through the circuit. We ran a simulation of the first test and
got the result shown in Figure 7.4.

Figure 7.3: Diode Test Circuit

Figure 7.4: Diode Test Simulation

When the voltage rises above the threshold voltage, we expect to see the voltage
at Vout follow the input, only slightly lower as there will be a drop across the
diode. For the second test, we should see the voltage drop across the diode vary
slightly as more current is passed through the diode. We will take the appropriate
measurements for each test and compare the results with the datasheet.

The final diodes that needed to be tested were the indication and warning LEDs.
We connected a 220Ω resistor to the 5V regulated output of our power supply
and inserted the green LED. We confirmed that the LED was shining brightly and
not getting too hot. We performed the same test with the red and blue LEDs and
obtained the same result. This confirms both our design and that the LEDs are
functioning correctly.

7.2.5 MCU and Crystal

Now that we had an operating power supply, we had to test the remaining
components necessary to operate the microcontroller chip. We began by placing
the ATmega328P into the compatible breadboard socket and placed it on the
breadboard. Using our 5V regulated power supply, we made all of the necessary
connections to VCC and GND pins on the IC. We then connected the 16MHz
crystal and 22pF capacitors as shown in Figure 7.5.

Figure 7.5: Crystal Oscillator Connection

Lastly, we connected the pullup resistor from 5V to the reset pin of the
microcontroller. In order to confirm the operation of these devices, we had to
connect a laptop with the Arduino IDE to the circuitry. To do so, we connected
our USB to serial breakout board to the appropriate pins of the IC, power and
ground. We opened the Arduino IDE and confirmed that the program recognized
the device by writing a simple code to blink an LED on digital pin 13. We
connected a 220Ω resistor in series with a red LED to ground and confirmed that
the LED was blinking. This test proved that both the microcontroller and crystal
were operating as expected.

In order to further test the microcontroller, we connected a 10kΩ potentiometer

from 5V to the input of AIN0 which is an analog input to the ADC. We then
established serial communication to the board and through Arduino’s serial
monitor, we were able to feed in the output value of the ADC as we varied the
potentiometer. We confirmed that the output values of the ADC went from 0 to
1023 which was expected as the preset reference voltage of the ADC is 5V.

7.2.6 LCD and Bluetooth® Module

In order to test the the LCD and Bluetooth® module, we decided to first test their
functionality using the Arduino Uno we used in our initial proof of concept.
Because we could still have potential issues with our standalone microcontroller,
we thought it best to ensure the individual components were operating correctly
on an Arduino development board. We connected the LCD as shown in Figure
7.6. with a 10kΩ potentiometer between Vcc and Vo of the LCD which controls
the contrast. We then used the Arduino IDE to write 10 different example
programs provided in Arduino’s Liquid Crystal library. These example programs
make use of each function available in the library that can be used to control all
aspects of an LCD. After performing these example codes, we were confident
that our LCD was in perfect working order.

The final hardware we have to test is the Bluetooth® module. In order to do this,
we have to first connect the module to the appropriate pins of the Arduino. The
transmit (TX) pin of the Arduino will be connected to the receive (RX) pin of the
Bluetooth® module and vice versa. Because the Bluetooth® module operates on
3.3V logic levels, we have to make a simple voltage divider that will drop the 5V
signals being output from the MCU’s TX pin to 3.3V. The resistors we will use are
2kΩ for R​1 and
​ 4.7kΩ for R​2​. The remaining connections are simply power and
ground. We then have to create a simple code that sends data for the Bluetooth®
module to send to a smartphone application. Android has several free
applications that allow users to connect to Bluetooth® devices such as this. We
will be using an application called S2 Terminal for Bluetooth® as our application
has not been fully developed yet. The code we will write will send a variety of
data to the application, and the application will send the data back to the Arduino.
Upon successful transfer of data in both directions, we will have confirmed that
our Bluetooth® module is working correctly.

Figure 7.6: LCD and Bluetooth® Implementation

7.4 Application Test Environment

The Android application will be tested using multiple environments both

throughout and after the development process. This will be done using the same
device as the development of the application in order to remain consistent with
the development conditions for example Windows operating systems, Android
Studio versions, etc. As previously stated, the application will only be available
for the latest Android operating system, so mobile devices running Apple or Linux
operating systems, or even versions of the Android operating system that are
below the Android software release deployment 4.4 (Kit Kat) may not allow
optimal functionality. The Android development platform, Android Studio, also
includes a virtual Android operating system emulator which can be used to
simulate application operations in a controlled environment on multiple Android
operating system versions. This utility will be leveraged to ensure compatibility is
maintained across all platforms and versions of the Android operating system
listed as compatible with the software.

An image of the emulating utility is pictured to the below, presenting the amount
of possible virtual testing conditions able to be simulated. Many more versions
are available than the ones shown in the image, however not all will be required
to test as the APIs extend past 19, which is the corresponding API to Android 4.4
Kit Kat. The newest version of Android, Android 7.0 (Nouget) API 24 will be
tested to ensure functionality on the most up to date version of Android, and
testing will commence down the list of APIs in decreasing order until API 19 is
reached and confirmed functional.

In addition to utilizing the Android Studio provided emulating software shown

above, the application will also be downloaded onto multiple portable devices in
order to test for functionality in physical portable devices and the performance of
the application such as hardware and memory allocation, energy consumption
while running application, and overall reactivity to application given commands in
and out of the application. These portable devices will have varying Android
versions as well as hardware specifications such as RAM and hard drive space
in order to expose the application to various degrees of variables to test through.
These portable devices including specifications will be listed after testing has
begun.

7.4 Application Specific Testing

The digital voltmeter application involves many intricate programming techniques

to create, so testing this application will involve many steps from start up times
and GUI (Graphical User Interface) to measurement refreshing of rapidly
changing data and visual representations of the data being transferred by the
physical voltmeter. The application will first be tested based on the start up speed
and consistency of the application opening across multiple devices. Although
available hardware resources will impact opening and performance times, it is
imperative that the application can open without crashing across multiple testing
platforms to ensure widespread functionality and the lack of “bugs” in the
software development of the application.

The GUI and menus are the next important activity to be tested as this is how the
user interacts with the application and sets personal parameters to allow for a
streamlined and personalized experience while using the application. The GUI
will be tested by having data available to be changed by the user, such as units
of measurements, etc. in order to make sure the application is responding to the
user’s requests, as well as the links between buttons and launching new
“activities”. These “activities” refer to new screens or paths being launched or
created during a push of a button. If any of the buttons or lists are shown to be
nonfunctional, then an activity listener has failed and likely a crash or memory
dump will occur. This means all paths must be created, instantiated, and mapped
correctly to ensure optimal functionality. Although startup and user interface are
drastically important components of this application, the perhaps most important
component is the data collection and representation from the digital voltmeter
hardware to the application in the portable device.

The connection between the digital voltmeter and the portable device will be
achieved using a Bluetooth® wireless connection. This Bluetooth® connection
must first be granted permissions through the Android operating system in order
to receive data. The connection of the Bluetooth® devices will be tested across
multiple devices, i.e different mobile phones all using the application. Different
versions of the Bluetooth® wireless technology will be tested in order to ensure
backwards compatibility and the use of multiple mobile devices will present many
different operations required for pairing the Bluetooth® transmitters. Not only will
initial connection be an issue, but sending and receiving data is only present if
both the Android operating system and user allow this communications path to
be secured. Because the microcontroller will be used to compile and transfer
data to the application, predetermined circuit voltages will be used to test the
data transfer between the application and microcontroller.

The user will decide when data is transferred and represented on the application
so every test can be controlled based on the values chosen to be test values.
Finally the data received will be tested and compared for consistency and
accurate graphical representation. The data must be transmitted and received
first to ensure packages of information are able to be sent through the path
between the microcontroller and the application. The most accurate way to test
this data received is to compare the incoming numerical values measured to a
preexisting voltmeter that has already been mass manufactured and calibrated.
These results must compare accurately and consistently in order for the testing
to prove a fully functioning prototype of the digital voltmeter. After this testing is
confirmed, a final version of the voltmeter may be manufactured.

7.5 Breadboard and PCB Test Procedure

In this section, we will outline the specific test procedure we plan to conduct that
will ensure full functionality of the product. Upon the success of each individual
component tests, we will construct the breadboard prototype in the steps listed in
section 6. For each of the steps, the following tests listed below will be performed
in order once construction has been completed. Table 7.3 lists all of the
necessary equipment that will be used in the following test procedures.

Equipment Use Item Indicator

High-voltage power Producing DC voltage a

supply for measurement

Signal generator Producing AC signals for b

measurement

Digital Multimeter Continuity tests and c

comparison control

Oscilloscope Measuring AC signals d

for comparison
Laptop + Arduino IDE Programming and serial e
communication

USB to Serial breakout Programming and serial f

board communication
Table 7.3: Test Procedure Equipment

Step 1: Standalone Microcontroller

1. Visual inspection and continuity
a. Ensure all connections in Figure 6.1 have been made
b. Using item [c], ensure all point-to-point shorts are <5Ω
2. Power up and programs
a. Power the device and connect [e] to MCU through [f]
b. Connect LED and 220Ω to digital I/O D0 of MCU
i. Write program to blink LED with a delay
ii. Confirm correct operation
iii. Repeat for each remaining digital I/O
iv. Remove resistor and LED
c. Connect pushbutton, 10kΩ pull-up resistor to digital I/O D0 of MCU
i. Write example programs below and confirm expected values
1. Button
2. Debounce
3. StateChangeDetection
d. Connect 10kΩ potentiometer and 10kΩ resistor to analog input A0
between VCC and ground
i. Write example programs below and confirm expected values
1. AnalogReadSerial
2. ReadAnalogVoltage
ii. Remove potentiometer and resistor

Step 2: Measurement Circuitry

1. Visual inspection and continuity
a. Ensure all connections in Figure 6.2 have been made
b. Using item [c], ensure all point-to-point shorts are <5Ω
c. Using item [c], adjust R12 to create ~10MΩ between range switch
and positive lead
d. Using item [c] measure and record resistances of R2-R7
i. Make necessary adjustments to match values in Figure 3.1
2. DC measurement
a. Disconnect all jumper wires from MCU analog inputs
b. Select mV range using the Range Selector Switch
c. Using item [a], generate a 0V DC signal, connect output to test
leads of meter
 d. Using item [c], measure and record voltage at the end of the jumper
wire removed from A0 in step a
e. Sweep the voltage from 0-1V with 10mV increments and repeat
steps c and d
f. Confirm recorded values match expected values and make
adjustments accordingly
g. Select next range and repeat above steps, incrementing from
1-10V using 100mV increments
h. Select next range and repeat above steps, incrementing from
10-100V using 1V increments
i. Re-connect jumper wires to analog inputs
j. Write example program “AnalogReadSerial” to MCU
i. Repeat steps b through h while recording the output values
of the ADC being displayed in the serial monitor of [e]
ii. Confirm for each range, the ADC output values follow the
input from zero to 1023
3. AC measurement
a. Repeat steps a and b of the previous test
b. Using item [b], generate a 10mV signal at 1kHz and connect output
to test leads
c. Using item [d], record the waveform at the end of the jumper wire
removed from A0 in step a
d. Repeat steps e through i of the previous test
e. Write program to sample the ADC at maximum speed for ~0.2
seconds and print the values to the serial monitor of item [e]
f. Inspect the printed values and locate the positive and negative
cycles of the input signal
g. Determine the number of cycles recorded and calculate the
frequency of the signal
i. Confirm this value matches the 1kHz input
ii. If necessary, adjust the sampling frequency and length
h. Confirm peak values for each range have been detected
i. Repeat steps b through h at the following frequencies
i. 0.1 kHz
ii. 1 kHz
iii. 10 kHz
iv. 100 kHz
v. 1MHz

Step 3: LCD implementation

1. Visual inspection and continuity
a. Ensure all connections in Figure 6.3 have been made
b. Using item [c], ensure all point-to-point shorts are <5Ω
2. Repeat DC and AC measurement procedures to ensure the values match
the ones recorded with the LCD disconnected in Step Three
3. Using item [e], write a program to calculate the current voltage (V), the
average voltage (Vavg), the minimum voltage (Vmin) and the maximum
voltage (Vmax) when a 10V DC signal is applied at the input
a. Write the calculated values to the LCD
b. Ensure values displayed match the values generated at the input
4. Using item [e], write a program to calculate the RMS voltage (Vrms), the
peak voltage (Vpeak), and the frequency when a 10V AC signal at 1 kHz
is applied to the test leads using item [b]
a. Write the calculated values to the LCD
b. Ensure values displayed match the values generated at the input

Step Four: Pushbutton and LED implementation

1. Visual inspection and continuity
a. Ensure all connections in Figure 6.4 have been made
b. Using item [c], ensure all point-to-point shorts are <5Ω
2. Combine the AC and DC measurement programs written in the previous
section to include the AC/DC selector switch and indication LEDs
a. The program should now switch between modes when the
pushbutton is pressed (default to DC on powerup)
b. Write the program to the MCU and repeat the AC and DC tests
performed in Step Two
c. Ensure recorded values match the previously recorded values
within 1%

Step Five: Bluetooth® implementation

1. Visual inspection and continuity
a. Ensure all connections in Figure 6.4 have been made
b. Using item [c], ensure all point-to-point shorts are <5Ω
2. Establish connection between smartphone and Bluetooth® module using
the S2 Terminal application
3. Modify the above program to incorporate the Bluetooth® module
parameters
a. Repeat steps 3 and 4 of “Step Three” and write the calculated
values to the RX pin (connected to TX of MCU) of the module
b. Ensure the values received in the S2 Terminal application match
the transmitted values
c. Additionally, ensure the values in the previous step match the
voltages being applied at the input
8 Administrative Content
8.1 Milestone Discussion
The following is a table which includes a list of initial milestones for designing the
Smart Digital Voltmeter:

Task Deadline

Senior Design I (Fall 2016)

● Discussion of ideas and selecting a project idea. 09/06/2016

● Basic research on project components.

● Submit initial project document - Divide and Conquer 09/09/2016

● ½ hour meeting with Dr. Lei Wei to discuss initial project document 09/19/2016

● Update Divide and Conquer files 09/30/2016

● Research hardware components 10/11/2016

● Choose microcontroller and other components needed

● Order and test parts 10/20/2016

● Design prototype 11/01/2016

● Develop code

● Submit Table of Contents 11/04/2016

● Current Draft of Senior Design I Documentation 11/11/2016

● ½ hour meeting with Dr. Lei Wei to discuss project draft 11/16/2016

● Final Document Due! 12/06/2016

Senior Design II (Spring 2017)

● Build prototype TBD

● Test and redesign project TBD

● Finalize prototype TBD

● Peer presentation TBD

● Final report TBD

● Final presentation! TBD

Table 8.1: Project Milestones
For Senior Design I, our main goal is to finish the course with a working voltmeter
assembled onto a breadboard. This voltmeter is then expected to send data
wirelessly onto a smartphone via an Android application. We took small steps
throughout the semester to accomplish this. Like any project, designing the smart
digital voltmeter began with extensive research on hardware parts as well as
software that would be used. The majority of our research consists of studying
previous voltmeter designs and constraints. In addition, to make sure that our
product can compete with the leading voltmeters in the market, we will research
how to keep our digital voltmeter within 1% accuracy, low difference in cost, and
little or no decrease in performance.

The main challenge that the group faced is the time constraint of designing a
multimeter (voltage, current, resistance) in such a short amount of time. As a
result, the project was simplified to designing a voltmeter capable of reading AC
and DC voltages rather than a full three function multimeter. Voltage is, after all,
the most important and easiest component to measure.

8.2 Budget and Finance Discussion

Component Quantity Cost per Unit

9V - Batteries 2 $0.49

Microcontroller 1 $10.99

PCB 2 TBD

Tablet / Phone 1 (already own Tablet / Phone)

Wireless Adaptor 1 (already included in Tablet /

Phone)

Bluetooth® Module 1 $7.39

PTC Thermistor 1 $1.24

LCD 1 $1.99

Diodes 1-2 $0.34

Resistors 20 $0.08

Varistors 3 $1.10

High-Power Wirewound Resistor 1 $1.25

Trim Potentiometers 2-3 $0.35

Range Selector Switch 1 $0.47

Pushbuttons 2 $0.20

Power Switch 1 $0.13

LEDs 4 $0.10

Crystal Oscillator 1 $1.24

Capacitors 4 $0.10

P-channel MOSFET 1 $1.61

Estimated Total $35.24

Table 8.2: Bill of Materials

The main specification under the budget and finance discussion is to keep the
smart digital multimeter as affordable as possible. Therefore, the parts used will
be the cheapest we can find on the market. The above hardware budget table
includes multiple copies of parts and in essence includes more parts than
required. As a result, the final estimated total is expected to decrease below the
$59.39 estimated budget.

The above costs, as well as the components listed above, and even the
quantities of each component are expected to change. The hardware budget
table is a very rough estimate of what total budget we are expected to work with.

Software/License Cost

Android SDK FREE (open source)

Code Composer Studio FREE (Texas Instruments)

Table 8.3: Software Budget

The microcontroller and the application will both be developed and designed
respectively using free software such as the softwares listed above. These
softwares will be used for the majority of the software development of this
project, however if more software becomes necessary a free software will be
used and listed above as well.

8.3 Sponsor Information

Voltmeters such as ours are used by a wide range of users such as electronic
hobbyists, engineers, and electricians. Our group met with the owners of
Commercial Lighting Enterprises Incorporated located in Orlando FL to discuss
our project and their interest in possible sponsorship. This company has been
operating for 22 years and serves clients such as Duke Energy and the Orlando
International Airport. After pitching our idea, the owners LaChelle and William E.
Wright happily agreed to fund our project with a budget of $500. The contact
information for the owners and company are listed below.

Main number: 407-788-0075

Website: www.commerciallightingorlando.com
Address: 8130 N. Orange Blossom Trail Orlando, FL 32810
9 Product Operation
This section describes the full use of our final product. in Section 9.1, we will
describe to install the smartphone application and outline the details for full
usage. Section 9.2 will explain how to take a DC measurement and how to
interpret the results on both the LCD and smartphone application. Similarly, in
section 9.3 we will describe how to take an AC measurement and interpret the
results. We will also discuss the features and requirements for safe operation in
each section. Lastly, we will give some insight to troubleshooting in the case of
incorrect measurements or product deterioration.

9.1 Using the Smartphone Application

My Smart Voltmeter is the accompanying application that will allow users to view
and record voltage measurements with The Smart Digital Voltmeter. To begin
using your product, visit the Google Play Store on your Bluetooth® capable
Android device and download My Smart Voltmeter. Once installation has
completed, pair your Bluetooth® capable device with the digital voltmeter
hardware. Finally you may open the Smart Digital Voltmeter application and
select which voltage measurement you would like to measure, either AC or DC.
Now you can measure the circuit as normal by using the two terminals on the
voltmeter. These measurements will be displayed on the application based on
the measurement the user selected. If AC voltage is selected, the user can
choose between a numerical view or a graphical view.

9.2 Taking Voltage Measurements

9.2.1 Setup

Now that the smartphone application has been installed and a Bluetooth®
connection has been established, the next step is to setup your device and take
a voltage measurement. Begin by inserting a 9V battery (1) in the meter by
connecting the positive and negative terminals to the connector clip (2) and move
the power switch (3) from OFF to ON. Ensure that the green LED (4) is
illuminated meaning that the device is turned on. The next step is to insert the
red test lead (5) into the VIN jack of the meter, then insert the black test lead (6)
into the COM jack. These will be used to take voltage measurements of the
device under test (DUT).

9.2.2 DC Measurement

To take a DC voltage measurement, follow the steps below.

 1) Ensure that the blue LED under “DC” is illuminated. This means that the
device is in DC measurement mode and you are ready to begin taking a
measurement.
a) If the LED under “AC” is illuminated, press the “AC/DC Selector”
button (7). This is used to switch between AC and DC modes.
b) The blue LED under DC should now be illuminated
2) Select the range to measure by moving the appropriate slide (1, 2, 3, or 4)
from the bottom to the top (ON) position on the “Range Selector” switch
(8). Ensure that only one switch is up, as incorrect measurements will
result from more than one range being selected.
3) Using the test leads, gently probe the DUT from one point to another while
simultaneously looking at the LCD (8).
4) The LCD will display the real-time voltage (V), the average voltage (Vavg),
the maximum (Vmax) voltage and minimum (Vmin) voltage
5) Remove the test leads from the DUT and view the smartphone
application.
6) The application will display the values measured much like the LCD,
presenting the real-time voltage (V), the average voltage (Vavg), the
maximum (Vmax) voltage and minimum (Vmin) voltage.

9.2.3 AC Measurement

To take an AC measurement, follow the steps below.

1) Push the “AC/DC Selection” button and ensure that the blue LED under
“AC” is illuminated.
2) repeat steps 2-3 of section 9.2.2 as these will be the same for both AC
and DC modes.
3) The LCD will display the RMS voltage (V​RMS​), the peak voltage (V​PEAK​),
and the frequency of the signal.
4) Remove the test leads from the DUT and view the smartphone application
5) The application will allow two different forms of data representation. The
user and choose between numerical which presents all the data the DC
measurement does such as real-time voltage (V), the average voltage
(Vavg), the maximum (Vmax) voltage and minimum (Vmin) voltage. The
user can also choose the graphical representation which will plot the
values of the voltages measured, essentially acting as an oscilloscope.

9.3 Troubleshooting

If for any reason the device is not operating as defined in this section, please
refer back to the above subsections and ensure all steps have been completed in
the order in which they are listed. If the device is still malfunctioning or you
believe a measurement to be incorrect, please consider the following options to
help troubleshoot your device.
9.3.1 LEDs and Display

● Green power LED is not illuminating

○ Ensure that the battery has been inserted correctly
○ If still not illuminating, replace the battery
● Red LED under “Range Selection Warning” is illuminated
○ Return all pins of item 8 to the bottom positions
○ Select only one pin for the appropriate range
● No DC measurement values on LCD
○ Ensure DC mode has been selected with item 7 and blue LED is
illuminated under “DC”
○ Ensure only one range has been selected and the red warning LED
is not illuminated
○ If still no values are being displayed, this means that the voltage
being measured is negative
○ Remove both leads and switch positions on the DUT, not the meter.
○ If in the correct measurable range, the LCD should now display the
measurement value
● LCD displaying “Maximum Voltage!”
○ This means that the voltage being measured is above the selected
range
○ Remove the leads from the DUT, select the next highest range and
return the probes to the DUT
○ If still being displayed, repeat the previous step
○ If still being displayed when the 1000V range has been selected,
the voltage being measured is above the maximum measurable
level

9.3.2 Incorrect Measurements

If you believe that the meter is producing incorrect measurements, please first
refer to the previous subsection and ensure all steps and user requirements have
been completed. If this is the case, follow the steps below.

1) Use a proven voltmeter to measure the voltage.

2) If the value is off by a substantial amount, this means that either the safety
features have been activated and halted measurement, or a component
within the product has been compromised.
3) If this is the case, the product must be returned for repair
4) If the values are only slightly off, open the device packaging and locate the
small blue potentiometer (10)
5) Re-probe the DUT after recording the correct value taken in step 1.
 6) Using a small flathead screwdriver, very slightly adjust the potentiometer
until the voltage measurement matches the value recorded in step 1.
7) Reassemble the device and resume normal use

9.3.3 Application Issues

If you are experiencing issues using the application, please follow the steps listed
below.

1) Ensure that the correct application has been installed

2) Power on the meter and establish a Bluetooth® connection
a) If this is issue, this means that the meter is defective and must be
returned for repair.
3) DC mode not showing values
a) Ensure the meter is also in DC mode by checking that the
indication LED under “DC” is illuminated
b) If issues are still present, the product must be returned for repair.
4) AC mode not showing values
a) Repeat steps (a) and (b) above but the LED should be illuminated
under “AC”
10 Project Summary
From the beginning of this project, we had an overwhelming amount of ideas and
applications for our product. At the very first stages, we wanted to create a digital
multimeter that could measure voltage, current, and resistance. We wanted to
take away the hassle of reading and recording measurements by wirelessly
transmitting the data to a smartphone application where the user could save the
measurements and plot them against time and frequency to act as a sort of
mini-portable oscilloscope. After researching all of the different existing designs,
parts, and considering all of the constraints that would be against us, we decided
to implement only voltage measurement capabilities. As digital multimeters have
been around for many decades, the designs and technologies have been
thoroughly tested and revised for optimal operation. These devices are very
complicated and require a level of electrical, mechanical, and computer
engineering that we have simply not reached yet. Additionally, we had to
consider our largest constraint which was time, as we only had a few months to
fully design our product. In the end, we determined to make a voltmeter that can
measure both AC and DC voltages. To compensate for the lack of functionality
compared to advanced multimeters, we knew that we could use our engineering
skills to give the users more useful measurement information such as average,
minimum, and maximum values that you would not normally see. And of course,
because the smartphone application was the backbone of our project, we
decided to continue with this feature and include all the bells and whistles to
accommodate the meter.

Once we had established all what we wanted out product to do, we continued
with our extensive research and parts selection. We came across several
obstacles and flaws throughout the design process and made several changes
as the project continued, all the way up until the final design. Even still, as we
have not yet implemented the full design and tested all functionality, we expect to
run into a few more roadblocks that will need to be addressed. Regardless, we
are all confident with the final design and know that with the resources available
and our own engineering skills that we can overcome any issues presented in the
implementation process. The following list outlines all of the functions and
features that our final product will offer:

1. AC and DC voltage measurement from 0-1000V

2. Measurement values displayed on an LCD
a. Real-time voltage
b. Maximum voltage
c. Minimum voltage
d. Average voltage
e. RMS voltage of AC signals
 f. Peak voltage of AC signals
g. Frequency of AC signals (to a certain limit)
3. Connectability with a smartphone application
a. Read all aforementioned values
b. Plot AC signals versus time and frequency
c. Export data to be shared with associates
4. Guaranteed user and product safety through strategically selected
hardware and test leads

In conclusion, throughout the process of this project, including the research,

testing, and troubleshooting, our group learned a lot of valuable information.
When it comes to the electrical hardware and circuitry, we had to apply all of the
knowledge that we have learned throughout our years of electrical and computer
engineering studies. Technologies such as varistors and thermistors used for
input protection were topics that we had to research and master on our own as
they were not covered in our classes. Existing knowledge of resistors, capacitors,
diodes, transistors, op amps, and many others were put to the test throughout
the design process. When it comes to the application and microcontroller, we had
to use all of the various techniques learned to optimize our design. Though we
were all tasked with the research and development of individual sections for the
final product, it was a group effort that required the input and insight from each
member. Throughout the stress and heartache, we acknowledged and overcame
all of the obstacles in one way or another and came to a final design which fully
covers all of the engineering knowledge we have acquired thus far. We are all
fully confident that by the end of next semester, we will have a working product
that will work as described above.
Appendix A: Works Cited
1. Varistor and the Metal Oxide Varistor Tutorial. (2016). Retrieved December 01,
2016, from http://www.electronics-tutorials.ws/resistor/varistor.html

2. How to test a fuse the easy way with multimeter. (n.d.). Retrieved December
01, 2016, from http://www.electronicrepairguide.com/how-to-test-a-fuse.html

3. ​ABCs of multimeter safety: Multimeter safety and you​ [PDF]. (n.d.). Everett,
WA: Fluke Corporation.

4. ​Analog and Digital Meters​ [PDF]. (n.d.). Toyota Motor Sales.

5. Teach ICT - GCSE ICT - advantages of bluetooth. (n.d.). Retrieved November

27, 2016, from
http://www.teach-ict.com/gcse_new/communication/wi-fi_bluetooth/miniweb/pg3.
htm

6. Brian, M., Wilson, T. V., & Johnson, B. (2001, April 30). How WiFi Works.
Retrieved November 22, 2016, from
http://computer.howstuffworks.com/wireless-network3.htm

7. Escobar, E. (2015, July 15). How Does Wi-Fi Work? Retrieved November 22,
2016, from https://www.scientificamerican.com/article/how-does-wi-fi-work/

8. ZigBee Overview - What is ZigBee? (n.d.). Retrieved November 22, 2016,

from http://www.telegesis.com/about-us/zigbee-overview/

9. Faulkner, C. (2015, November 17). What is NFC? Everything you need to

know. Retrieved November 20, 2016, from
http://www.techradar.com/news/phone-and-communications/what-is-nfc-and-why
-is-it-in-your-phone-948410

10. Rouse, M. (2007, May). What is Near Field Communication (NFC)? -

Definition from WhatIs.com. Retrieved November 20, 2016, from
http://searchmobilecomputing.techtarget.com/definition/Near-Field-Communication

11. Fitzpatrick, J. (2013, July 16). The Difference Between WEP, WPA, and
WPA2 Wi-Fi Passwords. Retrieved November 20, 2016, from
http://www.howtogeek.com/167783/htg-explains-the-difference-between-wep-wp
a-and-wpa2-wireless-encryption-and-why-it-matters/
12. RF Wireless Technology. (n.d.). Retrieved November 19, 2016, from
http://www.mouser.com/applications/rf-wireless-technology/

13. ​LED Drive Control Special Circuit TM1637​ [PDF]. (n.d.). Title Micro
Electronics.
http://olimex.cl/website_MCI/static/documents/Datasheet_TM1637.pdf

14. Mary, R. (n.d.). Wireless Communication and types. Retrieved November 18,
2016, from
http://www.engineersgarage.com/articles/wireless_communication?page=6

15. Girish, M. (n.d.). Arduino Bluetooth Interfacing - Simplest tutorial for

bluetooth basics. Retrieved November 18, 2016, from
https://diyhacking.com/arduino-bluetooth-basics/

16. Mobile OS market share 2016 | Statista. (n.d.). Retrieved November 18,
2016, from
https://www.statista.com/statistics/266136/global-market-share-held-by-smartpho
ne-operating-systems/

17. Make a Digital Voltmeter Using an Arduino. (2015, June 04). Retrieved
November 18, 2016, from
http://www.allaboutcircuits.com/projects/make-a-digital-voltmeter-using-the-ardui
no/

18. Hareendran, T. K. (n.d.). Arduino. Retrieved November 18, 2016, from

http://www.electroschematics.com/9351/arduino-digital-voltmeter/

19. Up and running with material design. (n.d.). Retrieved November 15, 2016,
from https://developer.android.com/design/index.html

20. Working Principle of Voltmeter and Types of Voltmeter. (n.d.). Retrieved

November 14, 2016, from Working Principle of Voltmeter and Types of Voltmeter.
(n.d.). Retrieved November 14, 2016, from
http://www.electrical4u.com/working-principle-of-voltmeter-and-types-of-voltmeter

21. Bohon, C. (2016). Apple's Swift programming language: The smart person's
guide - TechRepublic. Retrieved November 19, 2016, from
http://www.techrepublic.com/article/apples-swift-programming-language-the-smar
t-persons-guide/

22. Raza, A. (n.d.). Android vs. iOS: Which Should I Learn First? Retrieved
November 18, 2016, from
https://www.upwork.com/hiring/mobile/android-vs-ios-which-to-learn-first/