Faculty of Engineering

Department of Electrical Engineering

High-Efficiency Pure Sine Wave Inverter

Prepared for
Dr. Natarajan Krishnamoorthy
Final Year Degree Project
Lakehead University
Thunder bay, Ontario, Canada

Prepared by

Mathais Mebratu (0648149), Imzan Khan (0646989)

mmebratu@lakeheadu.ca

ikhan5@lakeheadu.ca

Electrical Engineering

15 April 2017
Abstract
The purpose of the project was to design a high-power inverter to rival that of use

in the market in terms of cost and efficiency. The efficiency was the key driving

force in the project. The inverter consists of 3 stages: the boost stage, inverter stage,

and filter/load stage. The boost stage consists of an isolated DC-DC converter

which will take a low DC input supply and boost it to a regulated high DC output.

(controlled by PWM signals). The inverter takes the high DC bus from the boost

stage and inverts it to a chopped AC, which is filtered to output a pure sine wave.

Load testing and efficiency calculations are done on the inverter.

2
Acknowledgements
We would like to acknowledge all our professors that have instructed us during

our time at Lakehead University. It is due to their guidance that helped us in the

successful completion of the project and in our studies. In special recognition; we

would like to acknowledge Dr. Natarajan Krishnamoorthy for his support and

willingness to always help when needed. In addition, our classmates, whom of

which have helped us immensely in the completion of the degree project.

3
Table of Contents

Contents
Abstract .................................................................................................................................................... 2
Acknowledgements ......................................................................................................................... 3
Table of Contents ................................................................................................................................ 4
Table of Figures .................................................................................................................................... 7
List of Tables ........................................................................................................................................ 10
List of Equations ................................................................................................................................. 11
Acronyms ............................................................................................................................................. 14
1. Introduction .................................................................................................................................... 15
1.1 Benefit to Society ................................................................................................................ 15
1.2 Project Duties ......................................................................................................................... 16
1.3 Project Scheduling ............................................................................................................. 17
2. Background ................................................................................................................................... 19
2.1 DC-DC Converter ................................................................................................................ 20
2.2 Isolated DC-DC Converter ............................................................................................ 21
2.2.1 Push-Pull Operation ................................................................................................... 23
2.3 Inverter....................................................................................................................................... 23
2.3.1 Pulse Width Modulation ......................................................................................... 24
3. Design Specifications and Procedure ........................................................................... 29
3.1 System Specifications ...................................................................................................... 29
3.2 Push-Pull Converter Design Procedure ................................................................. 29
3.2.1 High-Frequency Transformer Design .............................................................. 36
3.2.2 Output Filter Inductor Design .............................................................................. 44
3.2.3 Snubber Design ........................................................................................................... 50
3.2.4 Converter losses .......................................................................................................... 51
3.3 Inverter Low Pass Filter Design .................................................................................... 54
4. Implementation of Design .................................................................................................... 56
4.1 DC-DC converter implementation ........................................................................... 56
4.1a High-frequency transformer implementation ........................................... 56

4
 4.2 Implementation of the three level SPWM ............................................................ 63
4.2.1 Internal Registers ......................................................................................................... 63
4.2.2 Fast PWM ......................................................................................................................... 66
4.2.3 Lookup Table................................................................................................................. 67
4.2.4 Arduino sPWM Implementation ........................................................................ 70
4.3 Inverter MOSFET Operation ........................................................................................... 73
4.3.1 MOSFET Gate Driver .................................................................................................. 73
4.3.2 MOSFET Consideration ............................................................................................ 76
4.3.3 Losses in MOSFET ......................................................................................................... 77
4.4 H-Bridge operation in Inverter .................................................................................... 78
4.5 Measurement and Protection Circuits ................................................................... 79
5. Results ................................................................................................................................................ 83
5.1 DC-DC converter testing ................................................................................................ 83
5.1.1 Unloaded Testing & Results .................................................................................. 83
5.1.2 Loaded Testing & Results ....................................................................................... 84
5.2 Inverter Testing & Results ................................................................................................ 89
5.2.2 Loaded Testing & Results ....................................................................................... 90
5.3 Full System Testing & Results......................................................................................... 93
5.4 Harmonic Content of sinusoidal output ................................................................ 97
6. Economic and Project Management Analysis ........................................................ 98
6.1 Bill of Materials....................................................................................................................... 98
7. Conclusion....................................................................................................................................101
7.1 Project Outcomes .............................................................................................................101
7.2 Future Work ...........................................................................................................................101
Bibliography ......................................................................................................................................103
Appendix ............................................................................................................................................108
A.1 PCB for DC-DC converter stage ..............................................................................108
A.2 PCB for Inverter stage ....................................................................................................109
A.3 Microcontroller Schematic ........................................................................................110
A.4 DC-DC Circuit Schematic ...........................................................................................111
A.5 Inverter Circuit Schematic..........................................................................................112
A.6 Lookup Table MATLAB Code.....................................................................................113
A.7 Microcontroller Code ....................................................................................................114

5
 A.8 American Wire Gauge Table ....................................................................................119
Index ......................................................................................................................................................120

6
Table of Figures
Figure 1: Block diagram of power inverter ...................................................... 19
Figure 2: Basic topology for Boost Converter .................................................. 20
Figure 3: Basic Schematic of Push-Pull circuit ................................................. 22
Figure 4: Basic topology for inverter circuit ..................................................... 23
Figure 5: PWM with Different Duty Cycles ........................................................ 24
Figure 6: Output amplitude varied based on PWM duty cycle ................... 25
Figure 7: 2-Level PWM Output ........................................................................... 26
Figure 8: 3-Level PWM Output ........................................................................... 27
Figure 9: Sine PWM Output ................................................................................. 28
Figure 10: E-core transformer ............................................................................. 38
Figure 11: Arrangement of transformer ............................................................ 38
Figure 12: Core window area calculation ....................................................... 39
Figure 13: Skin effect ........................................................................................... 42
Figure 14: Core loss curve comparison ............................................................ 44
Figure 15: Kool Mu core selector chart ............................................................ 46
Figure 16: Kool Mu toroidal core ....................................................................... 46
Figure 17: Per unit permeability vs. DC bias .................................................... 48
Figure 18: Core dimensions ................................................................................ 49
Figure 19: Snubber design .................................................................................. 50
Figure 20: Series resistor ....................................................................................... 50
Figure 21: Distribution of converter losses ........................................................ 53
Figure 22: Distribution of losses with 4 parallel MOSFETs ................................. 53
Figure 23: Secondary winding turns .................................................................. 56
Figure 24: Centre tap primary windings ........................................................... 57
Figure 25: Complete transformer winded 1..................................................... 57
Figure 26: Complete transformer winded 2..................................................... 58
Figure 27: Reduced air gap transformer .......................................................... 58
Figure 28: Primary center-tap transformer ....................................................... 59
7
Figure 29: Inductance measurement............................................................... 59
Figure 30: Primary inductance........................................................................... 59
Figure 31: Secondary inductance measurement .......................................... 60
Figure 32: Turns ratio test ..................................................................................... 61
Figure 33: Measuring leakage inductance 1 .................................................. 61
Figure 34: Primary inductance with secondary open ................................... 62
Figure 35: Measuring leakage inductance 2 .................................................. 62
Figure 36: Primary inductance with secondary short..................................... 62
Figure 37: Testing of high-frequency transformer ........................................... 63
Figure 38: Timer/Counter1 Control Register A Bit Setup ................................ 64
Figure 39: Timer/Counter1 Control Register B Bit Setup ................................. 64
Figure 40: Timer/Coumter1 Interrupt Mask Bit Setup ...................................... 64
Figure 41: Fast PWM Timing Diagram................................................................ 66
Figure 42: Lookup Table Code .......................................................................... 68
Figure 43: Lookup Table Output Plots ............................................................... 69
Figure 44: Sample Lookup Table outputs ......................................................... 70
Figure 45: Arduino Code Segment-Gate Inputs ............................................. 70
Figure 46: Inverter Prototype .............................................................................. 71
Figure 47: H-Bridge Input (From microcontroller’s pins 11&12) ..................... 72
Figure 48: Unfiltered Output Response of H-Bridge @120Vpp ...................... 72
Figure 49: H-Bridge Output after LC lowpass filter @ 353Vpp ....................... 73
Figure 50: H-bridge .............................................................................................. 74
Figure 51: MOSFET Driver- connections (left) and pinout(right) ................... 74
Figure 52: Driver Specifications .......................................................................... 75
Figure 53: H-Bridge Operation ........................................................................... 79
Figure 54: Voltage and Current transducers ................................................... 80
Figure 55: DC-DC converter testing arrangement ......................................... 83
Figure 56: Unloaded regulation of output DC voltage ................................. 84
Figure 57: Input and output results of 147.3 ohm load .................................. 84

8
Figure 58: Duty cycle of loaded DC-DC converter ....................................... 85
Figure 59: Drain current waveform ................................................................... 85
Figure 60: Secondary Output voltage waveform .......................................... 86
Figure 61: Passive attenuator ............................................................................. 86
Figure 62: Output DC voltage in loaded condition ....................................... 87
Figure 63: Efficiency Vs. Various loads.............................................................. 88
Figure 64: Inverter testing arrangement........................................................... 89
Figure 65: Resistive load testing of inverter ...................................................... 90
Figure 66: Fan as inductive load ....................................................................... 91
Figure 67: Rating of fan ....................................................................................... 91
Figure 68: Measurement of low speed............................................................. 92
Figure 69: The whole system put together ....................................................... 93
Figure 70: Input voltage and current into the inverter .................................. 94
Figure 71: Output voltage of boost stage ....................................................... 94
Figure 72: Output current of inverter ................................................................ 95
Figure 73: Output voltage waveform and measurement ............................ 95
Figure 74: Harmonic content of output waveform ........................................ 97
Figure 75: PCB for DC-DC converter .............................................................. 108
Figure 76: Inverter PCB ...................................................................................... 109
Figure 77: μController Module ......................................................................... 110
Figure 78: DC-DC circuit schematic ............................................................... 111
Figure 79: Inverter circuit schematic .............................................................. 112

9
List of Tables
Table 1: Scheduling breakdown ....................................................................... 18
Table 2: Power Range of converter types ....................................................... 21
Table 3: System Specifications........................................................................... 29
Table 4: Push-pull Converter Specifications .................................................... 30
Table 5: High-Frequency Transformer Design Parameters ............................ 36
Table 6: Output inductor design parameters ................................................. 45
Table 7: Power MOSFET ....................................................................................... 51
Table 8: Diode parameters ................................................................................ 52
Table 9: Wave Generation Bits .......................................................................... 65
Table 10: Compare Output Mode for Fast PWM ............................................ 67
Table 11: Clock Select Bit Description .............................................................. 67
Table 12: MOSFET Specifications ....................................................................... 78
Table 13: Protection thresholds ......................................................................... 81
Table 14: Protection thresholds ......................................................................... 82
Table 15: Results for loaded DC-DC converter ............................................... 87
Table 16: Complete results for loaded DC-DC converter ............................. 88
Table 17: Load test of inverter ........................................................................... 90
Table 18: Spot testing of resistive load ............................................................. 94
Table 19: Full load testing of the overall system ............................................. 96
Table 20: Bill of materials..................................................................................... 98
Table 21: Salvaged/already available parts ................................................ 100

10
List of Equations
Equation 1: Input to output voltage transfer function ................................... 29
Equation 2: Switching period ............................................................................. 31
Equation 3: Maximum duty cycle ..................................................................... 31
Equation 4: Maximum duty cycle for each phase ........................................ 31
Equation 5: Input power ..................................................................................... 31
Equation 6: Max. average input current ......................................................... 31
Equation 7: Max. Equivalent flat topped input current ................................. 32
Equation 8: Max. input RMS current ................................................................. 32
Equation 9: Max. MOSFET RMS current ............................................................. 32
Equation 10: Min. MOSFET breakdown voltage .............................................. 32
Equation 11: Transformer turns ratio.................................................................. 32
Equation 12: Min. duty cycle ............................................................................. 33
Equation 13: Nominal duty cycle ...................................................................... 33
Equation 14: Average output current .............................................................. 33
Equation 15: Secondary max. RMS current ..................................................... 33
Equation 16: Rectifier diode voltage................................................................ 33
Equation 17: Min. output filter inductor ............................................................ 34
Equation 18: Output filter inductor ................................................................... 34
Equation 19: Min. output current ...................................................................... 34
Equation 20: Max. output ripple ........................................................................ 34
Equation 21: Output filter capacitor value ..................................................... 35
Equation 22: Equivalent Series Resistance ....................................................... 35
Equation 23: RMS capacitor current ................................................................ 35
Equation 24: Ripple input voltage .................................................................... 35
Equation 25: Input capacitor............................................................................. 35
Equation 26: Apparent power ........................................................................... 37
Equation 27: Ke parameter ................................................................................ 37

11
Equation 28: Core geometry parameter ......................................................... 37
Equation 29: Kg relation to core ....................................................................... 37
Equation 30: Core window area ....................................................................... 39
Equation 31: Primary turns calculation ............................................................. 39
Equation 32: Bmax check .................................................................................. 39
Equation 33: Primary inductance value .......................................................... 40
Equation 34: Secondary turns calculation ...................................................... 40
Equation 35: Skin depth ...................................................................................... 40
Equation 36: Wire diameter ............................................................................... 40
Equation 37: Conductor section ....................................................................... 41
Equation 38: Wire diameter for AWG21 ........................................................... 41
Equation 39: Wire area for AWG21 ................................................................... 41
Equation 40: Wire resistance for AWG21 ......................................................... 41
Equation 41: Number of primary wires ............................................................. 42
Equation 42: Total area of primary side ........................................................... 42
Equation 43: Primary resistance ........................................................................ 42
Equation 44: Primary resistance value ............................................................. 43
Equation 45: Total area of secondary side...................................................... 43
Equation 46: Number of secondary wires........................................................ 43
Equation 47: Secondary resistance .................................................................. 43
Equation 48: Secondary resistance value ....................................................... 43
Equation 49: Total copper losses ....................................................................... 43
Equation 50: Transformer regulation ................................................................. 44
Equation 51: Inductor Peak current value ....................................................... 45
Equation 52: LI^2 product .................................................................................. 45
Equation 53: Minimum nominal inductance ................................................... 47
Equation 54: Number of turns ............................................................................ 47
Equation 55: DC bias........................................................................................... 47
Equation 56: Adjusted number of turns ............................................................ 48

12
Equation 57: Circumference of inner core ...................................................... 49
Equation 58: Width of AWG21 ........................................................................... 49
Equation 59: number of turns/layer .................................................................. 49
Equation 60: Total layers ..................................................................................... 49
Equation 61: Stored energy in a capacitor ..................................................... 50
Equation 62: Estimate of power dissipation ..................................................... 51
Equation 63: Capacitor in snubber .................................................................. 51
Equation 64: Dissipated power.......................................................................... 51
Equation 65: Power dissipated in Rs ................................................................. 51
Equation 66: Conduction loss in MOSFET ......................................................... 52
Equation 67: MOSFET gate loss .......................................................................... 52
Equation 68: Switching losses in MOSFET .......................................................... 52
Equation 69: Conduction losses in diode ........................................................ 52
Equation 70: Switching losses in diode ............................................................. 52
Equation 71: Filter cut-off frequency ................................................................ 54
Equation 72: Capacitive reactance ................................................................ 54
Equation 73: Filter capacitor value................................................................... 55
Equation 74: Resonant frequency of filter ....................................................... 55
Equation 75: Actual filter capacitor value ...................................................... 55
Equation 76: Measured turns ratio .................................................................... 60
Equation 77: Measured total inductance ....................................................... 62
Equation 78: Leakage inductance value........................................................ 63
Equation 79: Input compare register................................................................ 68
Equation 80: Sample number ............................................................................ 69
Equation 81: AC voltage reading ..................................................................... 81
Equation 82: Sample voltage calculation ...................................................... 81
Equation 83: AC voltage reading at the LCD................................................. 81

13
Acronyms
PWM – Pulse Width Modulation

SPWM – Sinusoidal PWM

THD – Total Harmonic Distortion

PSW – Pure Sine Wave

CCM – Continuous Conduction Mode

D – Duty Cycle

DCM – Discontinuous Conduction Mode

MOSFET – Metal-Oxide-Semiconductor Field-Effect Transistor

ICR – Input Compare Register

RMS – Root Mean Square

MGD – MOSFET Gate Driver

UPS – Uninterruptible Power Supply

PCB – Printed Circuit Board

EMF – Electromotive Force

ISR – Interrupt Service Routine

AWG – American Wire Gauge

LCR – Inductance, Capacitance and Resistance meter

14
1. Introduction
A power inverter is simply an electronic circuit that uses a DC voltage as an input

and converts that to an AC voltage (Output). For this process to be successful, the

DC input source must be stable and have enough current supply. There are several

types of inverters that give various types of output. These include square wave,

modified sine wave, pulsed sine wave and pure sine wave (PSW). They all have

different uses in the industry but the PSW is mostly used since they can power

many common electronic devices. The PSW inverter is considered in this project.

PSW can also be used in solar powered systems in which a low DC voltage input

supplied from the solar panels is inverted to high power AC output which can be

used to power AC appliances. Another use of PSW is in Uninterruptible power

supply (UPS) which are used to provide a constant flow of power in sensitive loads

that require as such, PSW find their use in such power supplies. The use of PSW

inverter also allows a user to have access to an AC power if they might be in an

area with only DC power available such as at work sites or in camping sites.

1.1 Benefit to Society

A high efficiency single-phase inverter at a low cost is not only a benefit for

industrial applications, but also to residential applications. Having a single-phase

inverter in conjunction with an existing solar powered system tied to the grid is

not new concept; however, in terms of efficiency and money it is not feasible.

Therefore, having a high efficiency, single-phase inverter at a comparatively low

cost in comparison to current market inverters; not only makes the renewable

energy more enticing money wise, but also promotes renewable energy.

15
1.2 Project Duties
Farrukh Gill – PCB design, ordering of parts, schematic drawing, inverter design.

Mathais Mebratu – Specification of parts, safety & protection devices, and DC-DC

converter design.

Imzan Khan – Programming of the Microcontroller, feedback and interfacing

circuity design, inverter design.

Simardeep Gill - DC-DC converter design, main converter design and

economic/project management analysis.

All group members will be involved in design and testing of the PWM

programming, DC-DC converter & Inverter circuits. Project management duties

will be rotated amongst the group; thus, all group members will have a good

understanding of every aspect of the project, as well as a leadership role in a

project based environment.

16
1.3 Project Scheduling

Task Name Start Finish Project Manager

Degree Project 2016/17 Mon 9/12/16 Mon 4/10/17 Imzan Khan

Weekly Meetings with Supervisor Mon 9/12/16 Sat 4/1/17

Develop Final Presentation Mon 9/12/16 Mon 9/12/16

Final Report Sat 11/5/16 Sat 4/8/17

Initiating & Research Mon 9/12/16 Fri 11/4/16

Topic Research & Selection Mon 9/12/16 Sat 10/29/16

Develop Project Proposal Sat 10/29/16 Fri 11/4/16

Prepare Preliminary Project Scope Mon 10/31/16 Mon 10/31/16

Develop System Block Diagram Mon 10/31/16 Mon 10/31/16

Identify Goals and Objectives Mon 10/31/16 Mon 10/31/16

Document Project Costs and Benefits Tue 11/1/16 Tue 11/1/16

Develop Strategies and Plans Tue 11/1/16 Wed 11/2/16

Distribute Roles & Responsibilities Wed 11/2/16 Fri 11/4/16

Programming & Implementation Process Fri 11/4/16 Sun 1/8/17 Mathais Mebratu

Software Setup Fri 11/4/16 Sun 1/8/17

Microcontroller Selection Mon 9/12/16 Mon 9/12/16

Acquire Programming Software Wed 11/2/16 Wed 11/2/16

Develop Arduino code Mon 12/19/16 Sun 1/8/17

Implementation Sat 11/5/16 Sat 12/3/16

DC-DC Converter build Sat 11/5/16 Thu 11/17/16

Testing of converter Sat 11/5/16 Tue 11/22/16

Optimization of Circuits (Software) Wed 11/23/16 Wed 11/30/16

PCB Design Wed 11/30/16 Sat 12/3/16

Execution Wed 11/30/16 Fri 3/24/17

Authorize Work Wed 11/30/16 Wed 11/30/16

Order Parts Wed 11/30/16 Fri 1/6/17

Circuit Development Wed 11/30/16 Thu 3/23/17

Construction of Circuits Wed 11/30/16 Mon 3/20/17

Manage Requirements Tue 3/21/17 Tue 3/21/17

Efficiency Test Wed 3/22/17 Wed 3/22/17

17
 Optimization of Circuitry (Hardware) Thu 3/23/17 Thu 3/23/17

Closing Fri 3/24/17 Mon 4/10/17

Enclosure of Circuits Fri 3/24/17 Fri 4/7/17

Close Project Fri 3/24/17 Mon 4/10/17

Assess Satisfaction Fri 3/24/17 Fri 3/24/17

Final Presentation Rehearsal Mon 3/27/17 Thu 4/6/17

Final Presentation Fri 4/7/17 Fri 4/7/17

Table 1: Scheduling breakdown

18
2. Background
In this section, background information on the various parts used in constructing

the PSW is discussed. A PSW inverter consists of an input power supply, a DC-

DC converter, a DC-AC inverter and an low pass filter at the output. An overall

operation of a PSW inverter is as:

1. Low-voltage DC is inputted, which is provided, for example, from a car

battery or solar panel

2. A DC-DC converter will step-up the low voltage DC to a higher DC voltage

3. Next, the Arduino microcontroller will generate the sPWM and PWM

signals to convert this high DC voltage to an high voltage AC output

An block overview of the operation:

Figure 1: Block diagram of power inverter

19
2.1 DC-DC Converter
The DC-DC converter is a specific type of switching converter that converts one

level of DC input voltage to another DC output voltage. A DC-DC converter that

converts a low voltage DC to a higher DC voltage is called a boost converter and

a buck converter is the opposite of this.

Figure 2: Basic topology for Boost Converter

From Fig. 2, the switch is usually a MOSFET or IGBT, as fast switching is required.

When a high frequency square wave is applied to the gate, the switch will

repeatedly turn ON and OFF. When the switch will be ON, the inductor will begin

to store energy in its magnetic field, and when the switch is OFF, the back EMF

from the inductor is produced with a reverse polarity than when the switch was

ON. Due to this, the Vs (supply voltage) and VL (back EMF) will be in series. The

diode is forward biased in this stage, and a capacitor at the load charges to (Vs +

VL) minus the voltage drop across the diode. Therefore, the output always sees a

steady output voltage of (Vs + VL), because during the ON stage the capacitor

charges, and during OFF stage it discharges through the load. At steady-state

operation, the output voltage is found as:

𝑉𝑖𝑛
𝑉𝑜𝑢𝑡 =
1 − 𝐷𝑢𝑡𝑦 𝐶𝑦𝑐𝑙𝑒
This equation is assuming continuous operation mode, which is also the mode

used for this project. Fig. 2 is a basic boost converter switching circuit, in this report

a high turns ratio (24 V to 180 VRMS) is required so a transformer isolated switching

circuit is preferred as opposed to an inductor or non-isolated DC-DC converter.

20
2.2 Isolated DC-DC Converter
In these circuits, the duty cycle of the square wave input is adjusted, which will in

turn allow control of the power transferred to the load.

Circuit Power Range

Fly-back 1 W – 100 W

Forward 1 W – 200 W

Push-Pull 200 W – 500 W

Half Bridge 200 W – 500 W

Full Bridge 500 W – 2000 W

Table 2: Power Range of converter types

In addition, in the transformer isolated converters, the load side is isolated from

the AC lines. There are several topologies of transformer isolated converters given

in Table 1. The half or full bridge configurations are more suitable for higher DC

input voltages, but the gate circuitry for the switches are more complex. In this

project, the push-pull topology is used. This topology is used due its efficient

operation at low input voltages and at higher power applications. The breakdown

voltage of the power transistors should be greater than twice the input DC voltage

and due to this reason push-pull converter are not suitable for high DC input

voltages. In contrast to half bridges, where the switches must withstand voltage

equal to input voltage. This topology also allows to have multiple outputs. An

advantage the push-pull has over half bridge is that neither switch requires an

isolated driver. This design also used the current mode PWM control like the non-

isolated converters. In push-pull converters, eight transistors are used on the

primary side with a center-tapped high frequency transformer.

21
 Figure 3: Basic Schematic of Push-Pull circuit

An H-Bridge is on the primary side of high-frequency transformer. These switches

are open and closed 180 degrees out of phase. These 4 MOSFETs correspond to

high side left, high side right, low side left and low side right. The selection of

these MOSFETs whether NPN or PNP depend on parameters such as the RDS (on)

resistance, which should be very low as to reduce power losses. Since the switches

conduct alternately to each other a bipolar output voltage is seen at the primary

transform thus a full wave rectifier bridge is at the secondary. In addition, note

that the freewheeling diode could also be added to the primary transformer to

control the voltage present on the secondary when the switches open. The duty

cycle of the PWM modulator must be less than 0.5 or 50% to avoid conduction of

the alternate switches at the same time. In this project, the PWM duty cyle is set

to 0.45 or 45%.

22
2.2.1 Push-Pull Operation
The basic operation is similar to the forward converter. When the high side left

and low side right switches are ON, the current will flow through the rectifier

diodes and charge the output inductor with a positive voltage across the load,

likewise during the ON state of the high side right and low side left switches the

voltage across the load is negative. The drivers for the switches doubles the

effective duty cycle; therefore, the operating frequency at the output filter is

double the switching frequency. The drivers should have an appropriate amount

of dead time between the alternating phases, as to ensure one switch is completely

off before another switch conducts.

2.3 Inverter
The Inversion process is explained and comprised of the following sections: the

conversion of DC to AC by the H-Bridge, which is controlled by a MOSFET driver.

The MOSFET driver is controlled by the SPWM and PWM signals of the Arduino

ATMEGA 2560 microcontroller.

Figure 4: Basic topology for inverter circuit

23
2.3.1 Pulse Width Modulation
A ‘level’ is based on the amplitude levels of the varying pulses, which also includes

the zero-crossing level.

2.3.1a Two-level PWM

A ‘level’ is based on the amplitude levels of the varying pulses, which also includes

the zero-crossing level.

The 2-level PWM is normally referred to as PWM (Pulse Width Modulation) and

refers to the concept of rapidly pulsing a digital signal between the positive (or

negative) peak value and zero. The duty cycle of a PWM waveform determines

how long the pulse remains ON or HIGH, or in other words the percentage in

which the waveform is at its peak value (Figure 5).

Figure 5: PWM with Different Duty Cycles (Source: “Arduino - PWM,” Arduino - PWM. [Online].

Available: https://www.arduino.cc/en/Tutorial/PWM)

The use of PWM in an inverter circuit is crucial as it provides a means of

decreasing the total harmonic distortion of the load current. In addition, using a

PWM signal to control the inputs of switches (Figure 4) allows for more efficiency

24
as you can turn the load fully OFF and fully ON; thus, reducing the amount of

power wasted compared to a linear switch controller.

Figure 6: Output amplitude varied based on PWM duty cycle (Source: “ermicroblog,”

Microcontrollers and Electronics Parts e-Shop. [Online]. Available: http://www.ermicro.com/blog/?p=706 )

PWM is comparable to an AM signal, as the duty cycle (analogous to a human’s

varying vocal level) of the PWM signal determines the amplitude of the output

signal. as when a voltage in between ON and OFF is required, this can be achieved

easily by modifying the duty cycle (Figure 5).

Whereas, the carrier frequency of an AM signal is analogous to the ON/OFF

frequency of a PWM and is known as the switching frequency, which is essential

to reproducing an efficient (low power loss) output signal, and will be discussed

in later sections with regards to switching losses.

This amplitude modulation idea can be further modified to achieve a Sinusoidal

waveform (Figure 6).

25
 Figure 7: 2-Level PWM Output (Source: “Delta modulation,” Wikipedia, 15-Mar-2017. [Online].

Available: https://en.wikipedia.org/wiki/Delta_modulation)

The output waveform in figure 7, illustrates the filtered response of a 2-Level PWM

with reference of a sinusoidal waveform. The output response is not desirable for

the specifications of the project, as it is would result in a higher harmonic

distortion; thus, reducing the efficiency of the inverter.

2.3.1b Three-level PWM

To improve the efficiency, the number of levels the PWM waveform contains is

increased, as the more levels a PWM waveform has, the better an approximation

of a pure sine wave can be achieved (after passing through a filter).

26
Figure 8: 3-Level PWM Output (Source: “Pulse-width modulation,” Wikipedia, 24-Mar-2017. [Online].

Available: https://en.wikipedia.org/wiki/Pulse-width_modulation)

Referenced to a sine wave, the pulsing waveforms are segmented into 3-levels:

the positive peak to 0, the zero-crossing level, and 0 to the negative peak. Whilst

another level was introduced, the output waveform is still not usable for a high

efficiency inverter system.

2.3.1c Sinusoidal PWM

Sinusoidal PWM (sPWM), is the varying and combination of pulse widths in such

that they are proportional to the amplitudes of a sinusoidal waveform at the

sample time set by the switching frequency.

27
Figure 9: Sine PWM Output (Source: Terbytes, “Terbytes/Arduino-Atmel-sPWM,” GitHub, 14-Mar-2016.
[Online]. Available: https://github.com/Terbytes/Arduino-Atmel-sPWM)

This method will be used in the project as its filtered output more clearly

resembles that of a sine wave, and will be discussed in further detail in section

4.2.

28
3. Design Specifications and Procedure
The design procedure used for this project are found in references [27] to [32].

These textbooks contain the equations to use for designing switching power

supplies.

3.1 System Specifications

The system specification is listed below:

Specification Value

Nominal Voltage Input 24 Vdc

Output Voltage 120 Vrms at 60 Hz

Output Power 500 W

Efficiency 80%

Switching Frequency 30 KHz (DC-DC); 100 KHz (DC-AC)

Table 3: System Specifications

The relationship between the input and output voltage is given by:

𝑁2
𝑉𝑜𝑢𝑡 = 2 ( ) 𝐷𝑉𝑖𝑛 (1)
𝑁1
Equation 1: Input to output voltage transfer function

3.2 Push-Pull Converter Design Procedure

As per the table 1, a proper gate driver is needed which can switch at the required

frequency and also maintain the duty cycle needed. The SG3525A Pulse Width

Modulator Control Circuit was chosen for this for several reasons. Firstly, it

29
provides a wide range of switching frequency (100 Hz to 400 kHz). Secondly it has

built in soft-start circuitry to ensure no fast build up of charge. Thirdly, deadtime

is simply arranged by connecting a single resistor between the CT and Discharge

pins. The driver will be used as a voltage mode control as the input voltage is not

a constrained source and will vary. Voltage mode is also advantageous in

applications where high power is needed such as in this project. Also, as stated in

section 2.2, in practice; leakage inductance from the high frequency transformer

causes significant over-voltages across the transistor and thus the breakdown

voltage of these devices should be higher than twice the input voltage. The

STP160N75F3 N-Channel Power MOSFET’s were selected for the DC-DC

converter design. These MOSFET’s provide an ultra low on-resistance, can handle

up to 75 Volts with 120 Amps at 25 degrees Celsius. These switches provide

extremely fast switching and can be operated normally up to 175 degrees Celsius.

From Table 3, the DC-DC converter specifications are listed. In the procedure,

calculations are based upon a power output of 700 Watts as a gross overload for

the converter.

Specification Value

Nominal Voltage Input (Vin) 24 V

Maximum Voltage Input (Vin,max) 28 V

Minimum Voltage Input (Vin,min) 20 V

Nominal Power Output (Pout) 500 W

Nominal Output Voltage (Vout) 180 V

Target Efficiency (η) 90 %

Switching Frequency (f) 30 kHz

Table 4: Push-pull Converter Specifications

30
The switching period is given by:

1 1
𝑇= = = 33.33 𝜇𝑠 (2)
𝑓 30 𝑘𝐻𝑧
Equation 2: Switching period

Maximum duty cycle:

∗ (3)
𝑡𝑜𝑛 = 0.5𝑇 = 16.67 𝜇𝑠
Equation 3: Maximum duty cycle

Maximum duty cycle of each phase:

∗
0.9𝑡𝑜𝑛
𝐷𝑚𝑎𝑥 = = 0.45 = 45% (4)
𝑇
Equation 4: Maximum duty cycle for each phase

Input power:

𝑃𝑜𝑢𝑡 700
𝑃𝑖𝑛 = = = 778 𝑊 (5)
𝜂 0.9
Equation 5: Input power

Maximum average input current:

𝑃𝑖𝑛 778
𝐼𝑖𝑛 = = = 38.9 𝐴 (6)
𝑉𝑖𝑛,𝑚𝑖𝑛 20
Equation 6: Max. average input current

31
Maximum equivalent flat topped input current:

𝐼𝑖𝑛 38.9
𝐼𝑝𝑓𝑡 = = = 43.22 𝐴 (7)
2𝐷𝑚𝑎𝑥 2 ∗ 0.45
Equation 7: Max. Equivalent flat topped input current

Maximum input RMS current:

𝐼𝑖𝑛,𝑅𝑀𝑆 = 𝐼𝑝𝑓𝑡 √2𝐷𝑚𝑎𝑥 = 43.22√0.9 = 41 𝐴 (8)

Equation 8: Max. input RMS current

Maximum MOSFET RMS current:

𝐼𝑚𝑜𝑠,𝑅𝑀𝑆 = 𝐼𝑝𝑓𝑡 √𝐷𝑚𝑎𝑥 = 29 𝐴 (9)

Equation 9: Max. MOSFET RMS current

Minimum MOSFET breakdown voltage:

𝑉𝐵𝑟𝑘,𝑀𝑜𝑠 = 1.3 ∗ 2 ∗ 𝑉𝑖𝑛.𝑚𝑎𝑥 = 72.8 𝑉 (10)

Equation 10: Min. MOSFET breakdown voltage

Transformer turns ratio:

𝑁2 𝑉𝑜𝑢𝑡
𝑁= = = 10 (11)
𝑁1 2𝑉𝑖𝑛,𝑚𝑖𝑛 𝐷𝑚𝑎𝑥
Equation 11: Transformer turns ratio

32
Minimum duty cycle value:

𝑉𝑜𝑢𝑡
𝐷𝑚𝑖𝑛 = = 0.32 (12)
2𝑁𝑉𝑖𝑛,max
Equation 12: Min. duty cycle

Duty cycle at nominal input voltage:

𝑉𝑜𝑢𝑡
𝐷𝑛𝑜𝑚𝑖𝑛𝑎𝑙 = = 0.38 (13)
2𝑁𝑉𝑖𝑛
Equation 13: Nominal duty cycle

Maximum average output current:

𝑃𝑜𝑢𝑡
𝐼𝑜𝑢𝑡 = = 3.9 𝐴 (14)
𝑉𝑜𝑢𝑡
Equation 14: Average output current

Secondary maximum RMS current:

𝐼𝑠𝑒𝑐,𝑅𝑀𝑆 = 𝐼𝑜𝑢𝑡 √𝐷𝑚𝑎𝑥 = 2.62 𝐴 (15)

Equation 15: Secondary max. RMS current

Rectifier diode voltage:

𝑉𝑑𝑖𝑜𝑑𝑒 = 𝑁𝑉𝑖𝑛,max = 280 𝑉 (16)

Equation 16: Rectifier diode voltage

33
Output filter inductor value:

Assuming a ripple current value 15% of output current (𝛥𝐼 = 0.15𝐼𝑜𝑢𝑡 = 0.585).

𝑁
(𝑁2 𝑉𝑖𝑛 − 𝑉𝑜𝑢𝑡 ) 𝑡𝑜𝑛,max 15𝜇
1
𝐿𝑚𝑖𝑛 ≥ = ((10)(24) − 180) = 1.5 𝑚𝐻 (17)
Δ𝐼 0.585
Equation 17: Min. output filter inductor

Based on equation 17, the output filter inductor is chosen as:

𝐿𝑓𝑖𝑙𝑡𝑒𝑟 = 2 𝑚𝐻 (18)

Equation 18: Output filter inductor

The converter is used in Continuous Conduction Mode (CCM). This means that

current in the output inductor does not drop to zero during switching cycles. CCM

gives higher efficiencies as compared to Discontinuous Conduction Mode (DCM).

This output filter inductor value will ensure a CCM operation for a minimum

output current of:

Δ𝐼 0.585
𝐼𝑜𝑢𝑡,min = = = 0.293 𝐴 (19)
2 2
Equation 19: Min. output current

Maximum output voltage ripple value:

Δ𝑉𝑜 = 0.1%𝑉𝑜𝑢𝑡 = 0.180 𝑉 (20)

Equation 20: Max. output ripple

34
Output filter capacitor value:

1 Δ𝐼𝐿 1 0.585
𝐶𝑚𝑖𝑛 = 𝑇𝑠 = ∗ ∗ 33.33𝜇 = 13.6 𝜇𝐹 (21)
8 Δ𝑉𝑜 8 0.180
Equation 21: Output filter capacitor value

The Equivalent Series Resistance (ESR) must be lower than:

Δ𝑉𝑜 0.180
𝐸𝑆𝑅𝑚𝑎𝑥 = = = 0.31 Ω (22)
Δ𝐼𝐿 0.585
Equation 22: Equivalent Series Resistance

RMS capacitor current:

2 2
𝐼𝐶,𝑟𝑚𝑠 = √𝐼𝐼𝑛,𝑟𝑚𝑠 − 𝐼𝑖𝑛 = √412 − 38.92 = 12.95 ≅ 13 𝐴 (23)

Equation 23: RMS capacitor current

Ripple input voltage:

Δ𝑉𝑖𝑛 = 0.1%𝑉𝑖𝑛,max = 0.028 𝑉 (24)

Equation 24: Ripple input voltage

Input capacitor:

Δ𝑇𝑜𝑛,max 15𝜇𝑠
𝐶𝑖𝑛 = 𝐼𝐶,𝑟𝑚𝑠 = 13 ∗ = 7 𝑚𝐹 (25)
Δ𝑉𝑖𝑛 0.028
Equation 25: Input capacitor

The high voltage conversion ratio needed from system specifications is achieved

by proper high-frequency transformer turns ratio design. The high-frequency

35
transformer must be designed in order to minimize the leakage inductance.

Furthermore, the winding of the transformer must be symmetrical to minimize the

imbalance of the primary inductance values. Special care of the PCB design is

needed as well to have proper balance of the inductances values. There is also the

challenge of saturating the transformer through difference of peak current values

in the transistors.

3.2.1 High-Frequency Transformer Design

Specification Value

Nominal Voltage Input (Vin) 24 V

Maximum Voltage Input (Vin,max) 28 V

Minimum Voltage Input (Vin,min) 20 V

Nominal Output Voltage (Vout) 180 V

RMS Input Current (Iin) 41 A

Output Current (Iout) 3.9 A

Switching Frequency (f) 30 kHz

Efficiency (η) 98 %

Regulation (α) 0.05 %

Maximum Operating Flux Density (Bm) 0.05 T

Window Utilization (Ku) 0.3

Duty Cycle (Dmax) 0.45

Temperature Rise (Tr) 30 degrees Celsius

Table 5: High-Frequency Transformer Design Parameters

36
Apparent power:

𝑃𝑜 1 1
𝑃𝑡 = + 𝑃𝑜 = ( + 1) 𝑉𝑜 𝐼𝑜 = ( + 1) ∗ 180 ∗ 4 = 1.455 𝑘𝑊 (26)
𝜂 𝜂 0.98
Equation 26: Apparent power

Electrical condition parameter calculation Ke:

𝐾𝑒 = 0.145𝐾𝑓2 𝑓 2 𝐵𝑚
2 (10−4 )
= 0.145(4)2 (30 𝑘𝐻𝑧)2 (0.05)2 (10−4 ) = 522 (27)

Equation 27: Ke parameter

Note Kf = 4 is the waveform coefficient for square waves.

Core geometry parameter:

𝑃𝑡 1.455 𝑘𝑊
𝐾𝑔 = = = 2.787 𝑐𝑚5 (28)
2𝐾𝑒 𝛼 2 ∗ 522 ∗ 0.05
100
Equation 28: Core geometry parameter

Kg is related to the transformer core through:

𝑊𝑎 𝐴2𝑐 𝐾𝑢 (5.7 ∗ 5.292 ∗ 0.3)

𝐾𝑔 = = = 3.57 (29)
𝑀𝐿𝑇 13.4
Equation 29: Kg relation to core

Where Wa is the core window area, Ac is the cross-sectional area and MLT is the

mean length per turn.

37
In this project, the E65/32/27 core with N27 ferrite was chosen due to its high

saturation flux density (320 mT) and a high AL of 7200 nH. These parameters allow

us to have the fewest turns while still getting the required inductance.

Figure 10: E-core transformer

Two of these cores are arranged as such:

Figure 11: Arrangement of transformer

38
The values used in equation (29) are found as:

Figure 12: Core window area calculation

𝑊𝑎 = 12.7 ∗ 45 = 571.5 𝑚𝑚2 = 5.715 𝑐𝑚2 (30)

Equation 30: Core window area

The cross-sectional area and the MLT is found from the data sheet.

From datasheet, Bs = 320 mT = 3200 Gauss so assuming Bmax = 1500 Gauss to ensure

no saturation.

The number of primary turns is found as:

𝑉𝑛𝑜𝑚𝑖𝑛𝑎𝑙 ∗ (108 ) 108

𝑁1 = = 24 ∗ ≅ 3 𝑡𝑢𝑟𝑛𝑠 (31)
(4 ∗ 𝑓𝑜 ∗ 𝐴𝑐 ∗ 𝐵𝑚𝑎𝑥 ) 4 ∗ 30 𝑘𝐻𝑧 ∗ 5.29 ∗ 1500
Equation 31: Primary turns calculation

Check Bmax with the N1 calculated by re-arranging equation (31):

24 ∗ 108
𝐵𝑚𝑎𝑥 = = 1260 𝐺𝑎𝑢𝑠𝑠 (32)
4 ∗ 30 𝑘𝐻𝑧 ∗ 5.29 ∗ 3
Equation 32: Bmax check

39
With the approximation for N1 = 3 turns, the Bmax is below the 1500 Gauss designed

for so the approximation is good.

The primary inductance value is:

𝐿𝑝 = 𝑁 2 𝐴𝐿 = 32 ∗ 7200 = 65 𝜇𝐻 (33)

Equation 33: Primary inductance value

The number of secondary turns is:

𝑁2 = 𝑁𝑁1 = 10 ∗ 3 = 30 𝑡𝑢𝑟𝑛𝑠 (34)

Equation 34: Secondary turns calculation

3.2.1a Wire selection for high-frequency transformer

In this section, the magnetic wire to be used for winding the transformer are found.

Current penetration depth (Skin depth):

6.62 6.62
𝛿= = = 0.0382 𝑐𝑚 ≅ 0.4 𝑚𝑚 (35)
√𝑓 √30 𝑘𝐻𝑧
Equation 35: Skin depth

The wire diameter:

𝑑 = 2𝛿 = 0.7644 𝑚𝑚 (36)
Equation 36: Wire diameter

40
Conductor section:

𝜋𝑑 2
𝐴𝑤 = = 0.4589 𝑚𝑚2 (37)
4
Equation 37: Conductor section

From the electrical lab, AWG21 is available and according to the American Wire

Gauge Conductor Size Table; this wire is able to withstand a maximum current of

1.2 A at a maximum frequency of 33 kHz for 100% skin depth. The table can be

found in appendix A.8.

The AWG21 wire has the following specifications:

𝑊𝑖𝑟𝑒 𝑑𝑖𝑎𝑚𝑒𝑡𝑒𝑟 = 𝑑𝐴𝑊𝐺21 = 0.7239 𝑚𝑚 (38)

Equation 38: Wire diameter for AWG21

𝑊𝑖𝑟𝑒 𝑎𝑟𝑒𝑎 = 𝐴𝑤 𝐴𝑊𝐺21 = 0.41 𝑚𝑚2 (39)

Equation 39: Wire area for AWG21

𝛺
𝑊𝑖𝑟𝑒 𝑟𝑒𝑠𝑖𝑠𝑡𝑎𝑛𝑐𝑒 = 𝑅𝐴𝑊𝐺21 = 41.984 (40)
𝑘𝑚
Equation 40: Wire resistance for AWG21

A current density, J, was chosen as 4.5 A/mm2 from the plot of current density

versus frequency. Since J is dependent on the frequency, its value drops

considerably at high frequency called the “Skin effect”. If reached at this point, the

high-frequency alternating current only flows through the top layer of the

conductor as pictured in Fig. 7. Due to these reasons, the current density was

chosen as such to avoid these issues.

41
Figure 13: Skin effect (Source: [Online]. Available: https://www.solo-labs.com/cabling-effects-selecting-right-
cable/. [Accessed: 05- Apr- 2017].

The AWG21 magnetic wires need to be bundled appropriately as to handle the

high input current of 38.9 A.

Number of primary wires:

𝐴𝑤𝑝 8.9
𝑆𝑛𝑝 = = ≅ 22 𝑏𝑢𝑛𝑑𝑙𝑒𝑠 (41)
𝐴𝑤𝐴𝑊𝐺21 0.41

Equation 41: Number of primary wires

Where,
𝐼𝑖𝑛 38.9 40
𝐴𝑤𝑝 = = ≅ = 8.9 𝑚𝑚2 (42)
𝐽 450 450
Equation 42: Total area of primary side

Note that the input current is rounded to 40 A as to over-compensate the

requirement.

The primary resistance is:

−5 𝛺
𝑅𝐴𝑊𝐺21 (4.1984 ∗ 10 ) 𝑚𝑚 𝜇𝛺
𝑟𝑝 = = = 1.91 (43)
𝑆𝑛𝑝 22 𝑚𝑚
Equation 43: Primary resistance

42
Value of resistance for the primary winding:

𝑅𝑝 = 𝑁1 ∗ 𝑀𝐿𝑇 ∗ 𝑟𝑝 = (3)(13.4)(1.91) = 767.82 𝜇𝛺 (44)

Equation 44: Primary resistance value

Total area of secondary side:

𝐼𝑜𝑢𝑡 4
𝐴𝑤𝑠 = = = 0.89 𝑚𝑚2 (45)
𝐽 4.5
Equation 45: Total area of secondary side

Number of secondary wires:

𝐴𝑤𝑠 0.89
𝑆𝑛𝑠 = = ≅ 3 𝑏𝑢𝑛𝑑𝑙𝑒𝑠 (46)
𝐴𝑤𝐴𝑊𝐺21 0.41

Equation 46: Number of secondary wires

The secondary resistance:

−5 𝛺
𝑅𝐴𝑊𝐺21 4.1984 ∗ 10 𝑚𝑚 = 14 𝜇𝛺
𝑟𝑠 = = (47)
𝑆𝑛𝑠 3 𝑚𝑚
Equation 47: Secondary resistance

Value of resistance for the secondary winding:

𝑅𝑠 = 𝑁2 ∗ 𝑀𝐿𝑇 ∗ 𝑟𝑠 = (30)(134)(14) = 0.0563 = 56.3 𝑚𝛺 (48)

Equation 48: Secondary resistance value

3.2.1b Losses in HF transformer

Total copper losses:

2
𝑃𝑐𝑢 = 𝑃𝑝 + 𝑃𝑠 = 𝑅𝑝 𝐼𝑖𝑛 + 𝑅𝑠 𝐼𝑠2 = (767.82 𝜇𝛺)(29 𝐴)2 + (56.3 𝑚𝛺)(4 𝐴)2 = 1.546 𝑊 (49)
Equation 49: Total copper losses

43
Transformer regulation:

𝑃𝑐𝑢 1.546
𝛼= 100 = ∗ 100 = 0.221 % (50)
𝑃𝑜𝑢𝑡 700
Equation 50: Transformer regulation

3.2.2 Output Filter Inductor Design

In this project, the Kool Mu toroidal core is chosen as the output inductor. The

reason for this is due to Kool Mu’s core having high saturation level (up to 10 500

Gauss) and core loss being significantly less in high frequency application. The

procedure used to design the inductor is found in “Magnetics Power Core

Catalog”. Two design parameters are needed to determine the core size and

number of turns. They are the inductance needed with DC bias and the DC current.

Table 5 lists the specification needed on the output filter inductor.

Figure 14: Core loss curve comparison

44
 Specification Value

Minimum inductance value (Lmin) 2 mH

Calculated inductance value (Lcalc) 1.538 mH

DC current (Io) 3.9 A

AC current (ΔI) 0.585 A

Output power (Po) 700 W

Ripple frequency (fr) 2*30 kHz = 60 kHz

Table 6: Output inductor design parameters

Peak current value across the inductor:

Δ𝐼 0.585
𝐼𝑝𝑘 = 𝐼𝑜 + = 3.9 + = 4.1925 𝐴 (51)
2 2
Equation 51: Inductor Peak current value

The product of LI2:

2
𝐿𝑐𝑎𝑙𝑐 𝐼𝑝𝑒𝑎𝑘 = (1.538)(4.1925)2 = 27.03 𝑚𝐻 ∗ 𝐴2 (52)

Equation 52: LI^2 product

45
 Figure 15: Kool Mu core selector chart

From the LI2 product in equation (52) and from Fig. 9, a Kool Mu core having a

permeability of 60μ is needed and the Kool Mu core part number is: 77716AC. The

core that lies above the diagonal permeability line is chosen.

Figure 16: Kool Mu toroidal core

46
For winding this core, the core size and dimension are needed. From the datasheet,

the specifications of this core are:

𝑀𝑎𝑥 𝑂𝐷 (𝐴) 𝑑𝑖𝑚𝑒𝑛𝑠𝑖𝑜𝑛 = 72.4 𝑚𝑚

𝑀𝑎𝑥 𝐻𝑇 (𝐶) 𝑑𝑖𝑚𝑒𝑛𝑠𝑖𝑜𝑛 = 40.6 𝑚𝑚

73 𝑛𝐻
𝐴𝐿 = ± 8%
𝑇𝑢𝑟𝑛𝑠 2
67.16 𝑛𝐻
𝐴𝐿𝑚𝑖𝑛 = (73)(0.92) = (53)
𝑇𝑢𝑟𝑛𝑠 2
Equation 53: Minimum nominal inductance

The number of turns is found as:

𝐿 2 𝑚𝐻
𝑁=√ =√ = 172.56 𝑡𝑢𝑟𝑛𝑠 (54)
𝐴𝐿 67.16 𝑛𝐻
𝑇𝑢𝑟𝑛𝑠 2

Equation 54: Number of turns

The resulting magnetic force (DC bias) in oersteds:

0.4𝜋𝑁𝐼 (0.4𝜋)(172.56)(3.9)
𝐻= = = 66.56 𝑜𝑒𝑟𝑠𝑡𝑒𝑑𝑠 (55)
𝐿𝑒 127
Equation 55: DC bias

Where, Le is the path length which is 127 mm from datasheet.

47
 Figure 17: Per unit permeability vs. DC bias

From Fig. 11 at 66.56 oersteds, there is a roll-off of 0.65 in per unit of initial

permeability (μpu). Therefore, the number of turns must be increased to take into

the account of reduction of initial permeability.

Adjusted number of turns:

172.56
𝑁= = 265.47 ≅ 266 𝑡𝑢𝑟𝑛𝑠 (56)
0.65
Equation 56: Adjusted number of turns

Due to availability of parts, the Kool Mu core chosen is the part number 77192A7

rather than the 77716A7. This core is slightly larger than the 77716AC and has a

higher nominal inductance (138 nH/Turns2).

48
3.2.2a Wire selection for output inductor

Figure 18: Core dimensions

The AWG21 magnetic wire is also selected for this section.

26
𝐶𝑖𝑟𝑐𝑢𝑚𝑓𝑒𝑟𝑒𝑛𝑐𝑒 = 2𝜋𝑅 = (2𝜋) ( ) = 81.68 𝑚𝑚 (57)
2
Equation 57: Circumference of inner core

𝑊𝑖𝑑𝑡ℎ 𝑜𝑓 𝐴𝑊𝐺21 = (0.8)(3) = 2.4 𝑚𝑚 (58)

Equation 58: Width of AWG21

𝑡𝑢𝑟𝑛𝑠 81.68
= = 34.03 (59)
𝑙𝑎𝑦𝑒𝑟 2.4
Equation 59: number of turns/layer

266
𝑇𝑜𝑡𝑎𝑙 𝑙𝑎𝑦𝑒𝑟𝑠 = = 7.82 ≅ 8 𝑙𝑎𝑦𝑒𝑟𝑠 (60)
24.03
Equation 60: Total layers

This core will have the inductance equal to or greater than the one needed when

biased with the specified DC current.

49
3.2.3 Snubber Design
The snubber is placed across the switches on the primary side in the DC-DC

converter. It is used to reduce the peak voltage at turn-off and to damp the ringing.

A quick snubber design procedure is outlined in this section. [32] contains this

procedure.

Figure 19: Snubber design (Source: http://www.cde.com/resources/technical-papers/design.pdf)

In this project: Io is the primary DC current and Eo is the source voltage.

𝐸𝑜 48
𝑅𝑠 = = = 1.23 ≅ 1.2 𝛺 (61)
𝐼𝑜 38.9
Figure 20: Series resistor

Due to availability of parts in lab, a 10 Ω power resistor was used for the snubber.

The initial voltage step due to the current flowing in 𝑅𝑠 is no greater than the

clamped output voltage. The power dissipated in 𝑅𝑠 is estimated from peak

energy stored in Cs:

𝐶𝑠 𝐸𝑜2
𝑈𝑝 = (62)
2
Equation 61: Stored energy in a capacitor

50
The average power dissipation at a switching frequency:

𝑃𝑑𝑖𝑠𝑠 = 𝐶𝑠 𝐸𝑜2 𝑓𝑠 (63)

Equation 62: Estimate of power dissipation

A good choice to make 𝐶𝑠 to be equal to twice the sum of the output capacitance

of the switch and estimated mounting capacitance. From STP160N75F3 datasheet,

the output capacitance is 1080 pF and the mounting capacitance is a typical value

estimated at 40 pF.

𝐶𝑆 = 2(1080 + 40) = 2.24 𝑛𝐹 (64)

Equation 63: Capacitor in snubber

𝑃𝑑𝑖𝑠𝑠 = (2.24 𝑛𝐹)(48)2 (30 𝑘𝐻𝑧) = 0.154 𝑊 (65)

Equation 64: Dissipated power

(2.24)(48)2
𝑈𝑝 = = 2.58 𝜇𝑊 (66)
2
Equation 65: Power dissipated in Rs

3.2.4 Converter losses

The losses for the power MOSFET and the output diodes is considered in this

section.

Device RDS(ON) tr+tf Vbr Id at 100 °C

STP160N75F3 4 mΩ 65 ns + 15 ns 75 V 120 A

Table 7: Power MOSFET

51
The diodes chosen for this project are the STTH8R06 “Ultrafast diode”. The reason

for this selection of diodes is due to its ultrafast switching and lower switching

losses.

Device VF at 175 °C trrMax VRRM IF at 100 °C IRM

STTH8R06 1.4 V 25 ns 600 V 8A 5.5 A

Table 8: Diode parameters

For the conduction and switching losses, it is assumed that circuit operates in

worst case condition.

2
𝑃𝑐𝑜𝑛𝑑 = 1.6𝑅𝑑𝑠,𝑂𝑁 𝐼𝑀𝑜𝑠,𝑅𝑀𝑆 = (1.6)(0.004)(41)2 = 10.76 𝑊 (61)

Equation 66: Conduction loss in MOSFET

𝑃𝑔𝑎𝑡𝑒 = 𝑄𝑔 𝑉𝑔𝑠 𝑓 = (85 𝑛𝐶)(20 𝑉)(30 𝑘𝐻𝑧) = 0.051 𝑊 (62)

Equation 67: MOSFET gate loss

𝑉𝑏𝑟 𝐼𝑑 (𝑡𝑟 + 𝑡𝑓 ) (75)(120)(65𝑛 + 15𝑛)

𝑃𝑆𝑊,(𝑂𝑁+𝑂𝐹𝐹) = = = 10.8 𝑊 (63)
2𝑇 (2)(33.33𝜇)
Equation 68: Switching losses in MOSFET

3.9
𝑃𝑐𝑜𝑛𝑑,𝐷𝑖𝑜𝑑𝑒 = 𝑉𝐹 𝐼𝑠𝑒𝑐,𝑅𝑀𝑆 = (1.4) ( ) = 3.86 𝑊 (64)
√2
Equation 69: Conduction losses in diode

From diode datasheet, VRRM = 600 V however VRRM = 350 V is considered.

𝑉𝑅𝑅𝑀 𝐼𝑅𝑀 𝑡𝑟𝑟𝑀𝑎𝑥 ((350)(5.5)(25)(30 𝑘𝐻𝑧))

𝑃𝑑𝑖𝑜𝑑𝑒,𝑆𝑊 = 𝑓= = 0.72 𝑊 (65)
2 2
Equation 70: Switching losses in diode
52
 CONVERTER LOSSES
Diode
Diode
Conduction Switching
Losses: 14.73% Losses: 2.75%

Mosfet
Switching Mosfet
Losses: Conduction
41.23% Losses: 41.27%

Figure 21: Distribution of converter losses

To reduce the conduction losses, 4 power MOSFET are in parallel connection. The

conduction power loss is simply divided by four.

Diode
CONVERTER LOSSES
Switching Mosfet
Losses: 4% Conduction
Losses: 15%

Diode Mosfet
Conduction Switching
Losses: 21.3% Losses:
59.72%

Figure 22: Distribution of losses with 4 parallel MOSFETs

53
3.3 Inverter Low Pass Filter Design
The low-pass filter design is needed to remove the high frequency sPWM

switching leaving a smooth 60 Hz sine wave. The filter should not attenuate the

60 Hz fundamental frequency.

As a rule of thumb, the cut-off frequency must be 10 times away from the

fundamental to avoid attenuation. But to have more ideal filter, the cut-off

frequency must be logarithmically equally far away from the switching and

fundamental frequency.

𝐹𝑐 = √𝑓0 𝑓𝑠𝑃𝑊𝑀 = √(60)(20000) = 1095 𝐻𝑧 (66)

Equation 71: Filter cut-off frequency

The capacitor should be a suitable value in order not to add reactive current

through the inductor thereby immaturely saturating the inductor. Therefore, the

capacitor was chosen based upon an additional 250 mA reactive current on top of

the load current flowing through the inductor in order to provide the filtering

action.

Now, 𝑉 = 120 𝑉 and 𝐼𝑐 = 250 𝑚𝐴 so the following can be found.

𝑉 120
𝑋𝑐 = = = 480 𝛺 (67)
𝐼𝑐 0.25
Equation 72: Capacitive reactance

54
Then the capacitance value is found using 𝑋𝑐 = (𝜔𝑐)−1:

1 1
𝐶≤ = = 5.526 ≅ 5.6 𝜇𝐹 (68)
2𝜋𝑓0 𝑋𝑐 (2𝜋)(60)(480)
Equation 73: Filter capacitor value

Therefore, the capacitor must be equal or less than 5.6 μF. The inductor is pre-

determined to be 100 mH from already available parts. So the inductive reactance

is found as:
1
𝐹𝑐 = (69)
2𝜋√𝐿𝐶
Equation 74: Resonant frequency of filter

Solving for C:
1 1
𝐶= = = 211.2 𝑛𝐹 (70)
4𝜋 2 𝐹𝑐2 𝐿 4𝜋 2 (1095)2 (0.1)
Equation 75: Actual filter capacitor value

Capacitor chosen as a standard value of 220 nF.

55
4. Implementation of Design
The implementation of the PSW is done through two stages: implementation and

testing of DC-DC converter and implementation and testing of the inverter.

4.1 DC-DC converter implementation

Before the entire topology of the push-pull converter could be implemented on a

test board and the PCB, the high-frequency transformer and the output inductor

need to be winded and tested.

4.1a High-frequency transformer implementation

The winding of the secondary side(HV) includes 22 bundles of 3 turns around the

bobbin. Secondary side was winded first to accommodate the thick wires for the

centre tap primary windings.

Figure 23: Secondary winding turns

56
Similarly, the primary is winded along with centre tap wires terminated at the

bobbin. Fig. 22 shows the arrangement of the centre tap windings.

Figure 24: Centre tap primary windings

The finished primary and secondary windings with the core included are shown

Fig 23 and 24.

Figure 25: Complete transformer winded 1

57
 Figure 26: Complete transformer winded 2

In order to reduce the air gaps, a zip tie is used through the bobbin and clasping

the two e-core as much as possible.

Figure 27: Reduced air gap transformer

Now measurements will be conducted using an LCR meter to confirm the primary

inductance, secondary inductance, turns ratio and leakage inductance.

58
Figure 28: Primary center-tap transformer

Figure 29: Inductance measurement

Figure 30: Primary inductance

59
Note that this value is taken from the positive primary wire to the center tap which

is common to both primaries. From equation (33), the calculated primary

inductance is 65 μH and this closely matches the measured one at 67.1 μH.

Similarly, for the second primary; a similar value to 67.1 μH which confirms the

polarity dot connection from Fig. 19. The secondary inductance was measured and

shown in Fig. 20.

Figure 31: Secondary inductance measurement

The turns ratio:

𝑁𝑠 7.417 𝑚𝐻
= √𝐼𝑛𝑑𝑢𝑐𝑡𝑎𝑛𝑐𝑒 𝑟𝑎𝑡𝑖𝑜 = √ = 10.5 (66)
𝑁𝑝 67.1 𝜇𝐻

Equation 76: Measured turns ratio

Compared to equation (11), the measured turns ratio is similar to calculated turns

ratio. To confirm the above calculations a voltage signal was injected into the

60
primary winding and a reading was taken from the secondary. Fig. 29 illustrates

the procedure.

Figure 32: Turns ratio test

Leakage inductance is result due to the imperfect magnetic linking from one

winding to another. This leakage inductance is shown as an inductor in parallel

with the primary. To measure leakage inductance, the secondary is left open and

LCR is measures the inductance of the primary winding. The value measured is

the primary inductance plus the leakage inductance.

Figure 33: Measuring leakage inductance 1

61
 Figure 34: Primary inductance with secondary open

𝐿𝑡𝑜𝑡𝑎𝑙 = 𝐿𝑝 + 𝐿𝑙𝑒𝑎𝑘𝑎𝑔𝑒 = 280.19 𝜇𝐻 (67)

Equation 77: Measured total inductance

Now, the secondary side is shorted and inductance is measured again from the

primary. Due to the short, only the leakage inductance should be measured.

Figure 35: Measuring leakage inductance 2

Figure 36: Primary inductance with secondary short

62
Therefore:
𝐿𝑙𝑒𝑎𝑘𝑎𝑔𝑒 = 1.5 𝜇𝐻 (68)

Equation 78: Leakage inductance value

Figure 37: Testing of high-frequency transformer

From Fig. 33, the output is regulated at 180 V from a 24.2 V input.

4.2 Implementation of the three level SPWM

4.2.1 Internal Registers
The 3-level PWM signal is produced using internal registers of the Arduino

microcontroller (ATMEGA 2560). The registers are setup to produce a PWM

waveform suitable for controlling the power MOSFET’s of the inverter.

63
The internal registers are manipulated as follows:

Figure 38: Timer/Counter1 Control Register A Bit Setup

In Control Register A, TCCR1A is used to enable the wave generate mode

(WGM) of the Arduino. For 0b10100010: WGM11, COM1B1, COM1A1 are

enabled.

Figure 39: Timer/Counter1 Control Register B Bit Setup

In Control Register B, TCCR1B is used to enable the wave generation mode

(WGM). For 0b00011001: CS10, WGM12, WGM13 are enabled.

Figure 40: Timer/Coumter1 Interrupt Mask Bit Setup

64
In Interrupt Mask Register, 0b00000001 enables TOIE1

Bit 0 – TOIEn: Timer/Countern, Overflow Interrupt Enable

Interrupt Service Routine (ISR) defines what code is to be run continuously and

sets the timer to be used by that segment of code.

When this bit is written to one, and the I-flag in the Status Register is set (interrupts

globally enabled), the Timer/Countern Overflow interrupt is enabled.

The above-mentioned registers are used in conjunction to enable Mode 14 of the

Waveform Generation Mode Bit table, shown below. In this mode, Fast PWM

mode of the Arduino is initialized.

Table 9: Wave Generation Bits (Source of Figures and Tables: ATMEL, “Atmel ATmega640/V-1280/V -
1281/V-2560/V-2561/V,” 8-bit Atmel Microcontroller with 16/ 32/64KB In-System Programmable Flash, Feb-2014.

[Online]. Available: http://www.atmel.com/Images/Atmel-2549-8-bit-AVR-Microcontroller-ATmega640-1280-1281-

2560-2561_datasheet.pdf)

65
4.2.2 Fast PWM
The fast PWM mode provides a high frequency PWM waveform generation

option. The fast PWM differs from the other PWM options by its single-slope

operation. The counter counts from BOTTOM to TOP then restarts from BOTTOM.

The TOP value is set by the Input Compare Register (ICRn), which will be

explained in detail in the subsequent sections.

Due to the single- slope operation, the operating frequency of the fast PWM mode

can be twice as high as the phase correct and phase and frequency correct PWM

modes that use dual-slope operation. This high frequency makes the fast PWM

mode well suited for power regulation, rectification, and DAC applications. High

frequency allows physically small sized external components (coils, capacitors),

hence reduces total system cost.

Figure 41: Fast PWM Timing Diagram

66
 Table 10: Compare Output Mode for Fast PWM

The table above shows how Fast PWM utilizes Compare Output Mode;

COM1A1 & COM1B1 are both set to 1 and COM1A0 & COM1B0 are set to 0,

when configuring TCCR1A, which means the output compare registers are in

non-inverting mode.

Table 11: Clock Select Bit Description

The table above shows Clock Select Bit Description, CS10 is initialized while

initializing TCCR1B and sets our clock to 16MHz with no pre-scaling.

4.2.3 Lookup Table

The value of ICR1 is decided per the Arduino’s internal clock and the carrier

frequency.

67
 𝐼𝑛𝑡𝑒𝑟𝑛𝑎𝑙 𝐶𝑙𝑜𝑐𝑘 𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 16𝑀𝐻𝑧
𝐼𝐶𝑅1 = = = 800 (69)
𝐷𝑒𝑠𝑖𝑟𝑒𝑑 𝑆𝑤𝑖𝑡𝑐ℎ𝑖𝑛𝑔 𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 20𝑘𝐻𝑧

Equation 79: Input compare register

In the project, a switching frequency of 20kHz set for MOSFET’s, due to

suggestions in most Power Electronic textbooks, and based on the components

designed in the DC-DC portion. This value (800) is compared with the output

compare register values (OCR1A & OCR1B) set by the lookup table generator

in MATLAB. The lookup table is generated using the following code:

Figure 42: Lookup Table Code

In section 2.3.1a, the PWM waveform was described as being like an AM

waveform, this idea will be revisited and compared with the MATLAB code in

Figure 40.

The desired output frequency of the inverter is 60Hz. It is designed to be able to

power common electrical and electronic devices. The 60Hz will be the modulating

signal of the sPWM. This will be the signal amplitude of the duty cycle of the pulse

widths will be based upon.

68
In the development of the sPWM waveform, each full waveform will be comprised

of a certain number of pulse widths, which is calculated via MATLAB based on

the desired switching frequency and the required output frequency.

𝐷𝑒𝑠𝑖𝑟𝑒𝑑 𝑆𝑤𝑖𝑡𝑐ℎ𝑖𝑛𝑔 𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 20𝑘𝐻𝑧

𝑆𝑎𝑚𝑝𝑙𝑒 = = ≅ 332 (70)
𝑂𝑢𝑡𝑝𝑢𝑡 𝑆𝑖𝑔𝑛𝑎𝑙 𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 60𝐻𝑧

Equation 80: Sample number

The 60Hz sine wave will be sampled at 20kHz, which resulted in approximately

332 points (referred to as ‘samples’ in the code). The Arduino cannot produce a

negative voltage; thus, only the amplitudes of the positive half of the waveform is

required, and as such, the sample values is divided by two.

To sample the signal at the correct times, the output signal period is divided by

the number of samples, and is also divided by two; this value is referred to as

‘StopTime’ in the code, and will set the run time of program.

Figure 43: Lookup Table Output Plots

Lastly, the lookup table values (Figure 41) are generated via the sine wave function

69
in MATLAB. Then this is plotted with each value multiplied by the ICR value to

normalize to meet the compare register requirements of the Arduino

microcontroller.

For example: If the output compare register value is outputs 403 then the duty

cycle of that pulse would be 0.503 (≃50%), and occurs at approximately 1.4ms (≃

1.38)

Figure 44: Sample Lookup Table outputs: Amplitude(Left) and Time (Right)

As stated prior, the Arduino is incapable of outputting negative values (negative

voltages); thus, two positive waveforms are produced with the 2nd half being

identical to the first, but 180 degrees out of phase. Therefore, two output registers

are required, acting as the HIGH inputs to MOSFET drivers

4.2.4 Arduino sPWM Implementation

Figure 45: Arduino Code Segment-Gate Inputs

70
The lookup table values are used in the Arduino code (full code attached to the

Appendix A.5) as the OCRn values (Figure 39).

Note: The above code is ran continuously using global interrupts at the internal

clock frequency.

The application of the sPWM and PWM pins in the circuit will be discussed later
in the report.

4.2.4a SPWM Waveforms Test

Figure 46: Inverter Prototype

A prototype of the inverter was built as a means of testing the SPWM inputs on

the driver in conjunction with the driver and H-Bridge. The drain voltage was

tested at approximately 120Vpp and 150Vpp and was passed through the LC

filter at the output of the inverter.

71
Figure 47: H-Bridge Input (From microcontroller’s pins 11&12)

Figure 48: Unfiltered Output Response of H-Bridge @120Vpp

72
 Figure 49: H-Bridge Output after LC lowpass filter @ 353Vpp

4.3 Inverter MOSFET Operation

4.3.1 MOSFET Gate Driver
The output from Arduino pin is 5V at 40mA. This source doesn’t have enough

drive to run the H-Bridge on its own. Therefore, the use of MOSFET gate drivers

(MGD) is essential to the proper operation of the circuit. To choose the MGD, the

following specifications were considered.

The H-Bridge had to work to at least peak voltage of 180V, which comes from the

boost stage of the inverter. The gate driver supply voltage had to be at least 10V

or higher to operate N-Channel MOSFET IRFP460. Also, to make sure that either

𝑄1 & 𝑄2 or 𝑄3 & 𝑄4 (Fig. 48) are not turned on at the same time, there had to be built

in protection.

73
 Figure 50: H-bridge

To meet the above criteria, the IR2010 part was selected from Digi-Key® as the

MGD.

Figure 51: MOSFET Driver- connections (left) and pinout(right)

74
 Figure 52: Driver Specifications (Source of Figures 49-50: I. T. AG, “Products,” Infineon Technologies.
[Online]. Available: http://www.infineon.com/cms/en/product/power/gate-driver-ics/level-shift-gate-drivers/high-and-

low-side-drivers/IR2010/productType.html?productType=5546d462533600a401533d22608a563b.)

The capacitors between 𝑉𝑑𝑑 − 𝑉𝑠𝑠 , 𝑉𝑐𝑐 − 𝐶𝑂𝑀 are decoupling capacitors. Decoupling

capacitors in practice are used to compensate for the inductance of the supply

lines. The capacitor between 𝑉𝑏 & 𝑉𝑠 is the bootstrap capacitor, the only voltage

seen by the bootstrap capacitor is 𝑉𝑐𝑐 − 𝐷𝑖𝑜𝑑𝑒 𝐷𝑟𝑜𝑝. The gate charge for the high-

side MOSFET is provided by the bootstrap capacitor which is charged by the 15 V

supply through the bootstrap diode during the time when the device is off. Since

the capacitor is charged from a low voltage source the power consumed to drive

the gate is small. There are two ways of calculating bootstrap capacitance, using

the formula or using the graph. Both methods are explained in the application note

AN-978 from International Rectifier. Here the bootstrap diode blocks the voltage

seen at the 𝑉𝑠 terminal of IR2010 from 𝑉𝑐𝑐 . The bootstrap diode is between 𝑉𝑏 & 𝑉𝑐𝑐

of the IR2010.

The IR2010 is a high power, high voltage, and high-speed power MOSFET and

IGBT driver with independent high and low side referenced output channels.

Logic inputs are compatible with standard CMOS or LSTTL output, down to 3.0V

75
logic. Project requirement for 𝑉𝑂𝐹𝐹𝑆𝐸𝑇 (𝑚𝑎𝑥) 𝑖𝑠 180𝑉, this MOSFET driver can go

up to 200V.

Switching time had to be as low as possible, while maintaining the specifications.

𝑡𝑜𝑛 and 𝑡𝑜𝑓𝑓 for this driver is 95ns and 65ns respectively. The gate driver supply

range is from 10 to 20V. The MOSFET used is an N-Channel MOSFET (IRFP460)

which has a 𝑉𝐺𝑆 of ±20V. A gate driver is a power amplifier that accepts a low

power input from the microcontroller (Arduino) and produces a high current

drive input for the gate of high power transistor, such as the power MOSFET used

in this project. The max current on the Arduino output pin is 40mA, which is not

enough to run the MOSFET; therefore, the 𝐼𝑜 of the MOSFET driver is essential to

the working of the inverter efficiently. Input capacitance of the MOSFET (IRFP460)

is 1300pF; microcontrollers are usually designed to drive loads of less than a 100pF.

The maximum 𝐼𝑜 rating of the MGD is 3A; this current is used to charge the gate

capacitor of the MOSFET, the higher the current the faster the gate capacitor

charges reducing switching losses. The cross-conduction prevention logic

prevents the MOSFETs from turning on at the same time, specifically 𝑄1 & 𝑄2 or

𝑄3 & 𝑄4 (see Fig. 48).

4.3.2 MOSFET Consideration

MOSFETs are the preferred power transistors over power BJTs because of their

fast switching times and their stability over wide operating conditions. Second,

because of the high input impedance of the device, the drive circuitry can be of

low power, and compact and simple. Two or more MOSFETS can be connected in

parallel with ease to supply high-power loads. This is possible because of the

positive temperature coefficient of the device; when the MOSFET conducting the

76
higher current heats up it is forced to share its current with the other parallel

MOSFETS.

4.3.3 Losses in MOSFET

The on-resistance (𝑅𝐷𝑆_𝑂𝑁 ) of a Power MOSFET is a very important parameter

because it determines how much current the device can carry for low to medium

switching frequencies. For example, a switching frequency of 20kHz used in this

project. Paralleling the MOSFETS also reduces the 𝑅𝐷𝑆_𝑂𝑁 of the devices there by

reducing conduction losses. During continuous conduction at constant drain

current, the MOSFET conduction loss is found simply by calculating 𝐼𝐷2 𝑅𝐷𝑆_𝑂𝑁 .

The other source of power loss is through switching losses. As the MOSFET

switches on and off, its intrinsic parasitic capacitance (gate capacitance) stores and

then dissipates energy during each switching transition. The losses are

proportional to the switching frequency and the values of the parasitic

capacitances. As the physical size of the MOSFET increases, its capacitance also

increases; thus, increasing MOSFET size also increases switching loss. In

general, a higher switching frequency and a higher input voltage require a lower

𝑄𝐺 , therefore the lower the 𝑄𝐺 the lower the switching losses.

To build the H-Bridge per the project requirements the appropriate MOSFET had

to be selected. Based on the criteria mentioned above, the least amount of losses

had to be taken in to account. In addition, the voltage of 170V on the H-Bridge and

the current requirement of 4.2A had to be met.

The IRFP460 exceeded the project requirements, as it has the following

characteristics;

77
 Table 12: MOSFET Specifications (Source: alldatasheet.com, “IRFP460 Datasheet(PDF) - International
Rectifier,” IRFP460 Datasheet(PDF) - International Rectifier. [Online]. Available:

http://www.alldatasheet.com/datasheet-pdf/pdf/68529/IRF/IRFP460.html. [Accessed: 05-Apr-2017])

Therefore, the IRFP460 is used in conjunction with IR2010 MGD’s to develop the

H-Bridge circuit.

4.4 H-Bridge operation in Inverter

The sPWM from the Arduino is outputted on to Pins 11 and 12 in conjunction with

the alternating digital high and digital low signals on Pins 9 and 10. In the project,

two MGDs are used: one to control the switching of MOSFET 𝑆1 and 𝑆2 (Fig. 51)

the other MGD controls MOSFET 𝑆3 and 𝑆4 . Pin 11 and 12 are connected to 𝑀𝐺𝐷1

and 𝑀𝐺𝐷2 high input respectively. Also, Pin 9 and 10 are connected to 𝑀𝐺𝐷1 and

𝑀𝐺𝐷2 low input respectively. To operate the H-Bridge 𝑆1 and 𝑆3 are supplied

with sPWM from each of the MGD’s 𝐻𝑜 Pin. 𝑆2 and 𝑆4 are used to control the

output of the H-Bridge. When 𝑆2 is high the output flows through the load from

𝑆3 and 𝑆2 for one cycle, and for the other half it flows through 𝑆1 to 𝑆2 completing

one cycle. This output from the H-Bridge is also an sPWM but much higher

magnitude, resembles a sine wave after lowpass LC filter.

78
 Figure 53: H-Bridge Operation (Source: “multiple motor h bridge,” Electrical Engineering Stack
Exchange. [Online]. Available: https://electronics.stackexchange.com/questions/207319/multiple-motor-h-bridge.)

4.5 Measurement and Protection Circuits

When dealing with high powered applications proper caution must be taken when

performing measurements. In the full code of the Arduino, sensor values were

read via the analog inputs from HV DC bus voltage, HV AC output voltage and

AC current of the load and temperature of boost stage transistors. An AC voltage

and current transducer was used to measure the AC quantities while a simple

voltage divider circuit was used to measure the HV DC bus voltage. The

transducers are shown in Fig. 52 and Fig 53. The voltage transducer is rated as

0-150 VAC with an output DC current of 0-1 mA and should be terminated with a

load resistor of 4.99 kΩ (1%) to produce 0-5 VDC that can be inputted in to the

microcontrollers analog pins. Similarly, the current transducer is rated as 0-5 A AC

and outputs 0-1 mA DC current and should also be terminated with a load resistor

of 4.99 kΩ (1%).

79
To test the noise performance and accuracy of the transducers, a test was

performed injecting a rated voltage of 120 VAC and 5 AC to terminals 2 and 4 with

10 kΩ terminating resistor at terminals 1 and 2. See figure below.

Figure 54: Voltage and Current transducers with

10 kΩ terminating resistors between 1 and 2

Prior to reading the values directly at the inputs of the microcontroller they must

be first stepped down to a safe level acceptable by the Arduino (0-5V). The analog

values read from the pins vary from [0,1023] due to their 10-bit resolution; thus,

before any conversion takes place in the code, the analog input is multiplied by

1023 to normalize the value back to the [0,5] VDC range. However, for the DC Bus

regulation voltage, the [0,170] VDC is stepped down to [0,2.5] VDC as a means of

protecting the Arduino. Next, a specific conversion factor (CF) is applied to each

analog input value (as show in Table 13) depending the measurement being

performed.

80
 Measurement Analog Conversion factor(CF) Threshold Digital

Pin Pin

AC voltage 1 150𝑉𝑟𝑚𝑠 𝑉𝑟𝑚𝑠 < 106 Vrms 21

𝐶𝐹 = = 30
5𝑉𝐷𝐶 𝑉𝐷𝐶
Overcurrent 3 5𝐴 𝐴 <4A 26
𝐶𝐹 = =1
5𝑉 𝑉
DC Bus Regulation 5 180𝑉𝑑𝑐 150Vdc< x< 180Vdc 21
𝐶𝐹 = = 72
2.5𝑉𝐷𝐶

Table 13: Protection thresholds

Example Output:

AC Voltage = 60 VRMS
𝐴𝐶 𝑉𝑜𝑙𝑡𝑎𝑔𝑒 1023
𝑃𝑖𝑛1 𝑅𝑒𝑎𝑑𝑖𝑛𝑔 = ( ∗ 5) ∗ (71)
150𝑉𝑟𝑚𝑠 5𝑉𝑑𝑐
Equation 81: AC voltage reading

60𝑉𝑟𝑚𝑠 1023
𝑃𝑖𝑛1 𝑅𝑒𝑎𝑑𝑖𝑛𝑔 = ( ∗ 5) ∗ ≅ 409.2 (72)
150𝑉𝑟𝑚𝑠 5𝑉𝑑𝑐
Equation 82: Sample voltage calculation

5𝑉𝑑𝑐 5𝑉𝑑𝑐
𝐴𝐶 𝑉𝑜𝑙𝑡𝑎𝑔𝑒 = 𝐶𝐹 ∗ 𝑃𝑖𝑛1 𝑅𝑒𝑎𝑑𝑖𝑛𝑔 ∗ = 30 ∗ 409.2 ∗ = 60𝑉𝑟𝑚𝑠 (73)
1023 1023
Equation 83: AC voltage reading at the LCD

In addition, these voltage and current values are output to an LCD, along with

power calculations (calculated based on the voltage and current readings and the

temperature readings on the heat sink).

81
When a certain threshold is passed, a digital output is sent to the relay input of the

inverter circuit or the shutdown pin of the DC-DC converter to disconnect the

output from the board.

Measurement: Threshold:

Overcurrent < 4A

DC Bus Regulation 150VDC < x < 180VDC

AC voltage < 106 V

Table 14: Protection thresholds

82
5. Results
Several tests were conducted on the inverter and boost stage. In this section,

unloaded and loaded results are discussed along with important parameters such

as efficiency, input/output power and waveforms for voltage and current.

5.1 DC-DC converter testing

Figure 55: DC-DC converter testing arrangement

5.1.1 Unloaded Testing & Results

With an unload test on the DC-DC boost converter, the results were as expected.

Perfect regulation of the DC output voltage at 180V.

83
 Figure 56: Unloaded regulation of output DC voltage

5.1.2 Loaded Testing & Results

Now the DC-DC boost converter is loaded with various power resistor loads.

Firstly, a 147.3 Ω load is placed at the output. The load is obtained by two 75Ω

power resistor in series rated at 250 W.

Figure 57: Input and output results of 147.3 ohm load

84
Figure 58: Duty cycle of loaded DC-DC converter

Figure 59: Drain current waveform

85
 Figure 60: Secondary Output voltage waveform

Note for Fig. 33, the waveform is attenuated as there is not enough time-space

divisions to display the actual waveform. The passive attenuator consists of “L”

shape configuration as shown in Fig. 54.

IO1 R1 IO2

1MΩ
INPUT R2 OUTPUT
1MΩ
IO4

IO3

Figure 61: Passive attenuator

In the design, 𝑅1 = 𝑅2 . To get the actual waveform, the output waveform in Fig.

53 should be doubled.

86
 Figure 62: Output DC voltage in loaded condition

From Fig. 50 to Fig. 55, the following results are extracted for 147.3 Ω loaded

condition.

Load (Ω) Vin (V) Iin (A) Vout (V) Iout (A) Pin (W) Pout (W) D (%) η (%)

147.30 24.10 11.90 182.79 1.245 286.79 227.57 40.90 79.35

Table 15: Results for loaded DC-DC converter

87
Similarly, for other loads, the complete results are tabulated in table 14.

Load Vin Iin Vout Iout Pin Pout D (%) η (%)

(Ω) (V) (A) (V) (A) (W) (W)

1840.00 24.00 0.93 182.10 0.10 22.32 18.03 29.00 80.77

340.00 24.00 4.30 182.50 0.50 103.20 90.34 38.00 87.54

248.00 24.10 6.66 183.00 0.74 160.51 134.69 39.50 83.91

147.30 24.10 11.90 182.79 1.25 286.79 227.57 40.00 79.35

104.40 24.00 16.80 183.01 1.74 403.20 317.71 42.00 78.80

75.00 23.30 25.08 185.00 2.50 584.36 462.50 44.00 79.15

Table 16: Complete results for loaded DC-DC converter

Efficiency Vs. Various loads

90.00

88.00

86.00

84.00
Efficiency

82.00

80.00

78.00

76.00

74.00
1840.00 340.00 248.00 147.30 104.40 75.00
Load value in ohms

Figure 63: Efficiency Vs. Various loads

88
As seen from Fig. 56, the efficiency is maximum at 340Ω load at 87%. However, at

that level, the output power is not satisfactory (90.34 W). Therefore, the full load

is considered at a 75Ω load with 462.50 Watts of output power while achieving an

output current of 2.5 A. The efficiency suffers at 75Ω load at 79.15% but

considering a design target efficiency of 90%, this is not too bad. Due to the high

frequency noise, environment, losses from switches and transformer is the reason

for the drop in efficiency.

5.2 Inverter Testing & Results

Figure 64: Inverter testing arrangement

89
5.2.2 Loaded Testing & Results
In the loaded testing of the inverter, both resistive and inductive loads were

considered. In this stage, due to limitations of the power supplies up to 126 V input

DC supply was available with current limit at 20 A. Therefore, the testing is based

upon this rating however in section 5.2.3, the full system (boost stage + inverter

stager + filter/load stage) is tested at the proper input DC input voltage from the

boost stage.

Firstly, various power resistor loads are tested on the inverter. The following were
measured.

Figure 65: Resistive load testing of inverter

Load Iin Vout(rms) Iout Pin Pout

Vin (V) η (%)
(Ω) (A) (V) (A) (W) (W)

1840.00 126.90 0.08 94.00 0.02 10.53 1.79 16.96

1370.00 127.00 0.10 93.90 0.07 12.45 6.48 52.06

200.00 127.00 0.38 90.20 0.45 48.26 40.50 83.92

75.00 126.90 0.98 91.00 1.21 124.36 110.11 88.54

Table 17: Load test of inverter

90
 Figure 66: Fan as inductive load

Figure 67: Rating of fan

The inductive load will have two types of power consumption. One is the real

power due to the resistive component and a reactive power due to the inductive

component. The power factor of the fan is assumed to be 0.9 so 𝑃𝑜𝑢𝑡 = 𝑉𝐼𝑐𝑜𝑠(𝜃).

The results are tabulated for the fan for the varying speeds.

91
 Figure 68: Measurement of low speed

Load Vin Iin (A) Vout(rms) Iout Pin (W) Pout (W) η (%)

(V) (V) (A)

Low 126.90 0.13 93.50 0.11 16.12 9.59 59.52

Speed

Med 126.90 0.14 93.50 0.12 17.13 10.18 59.44

Speed

High 126.90 0.15 91.63 0.14 19.54 11.79 60.34

Speed

92
5.3 Full System Testing & Results
In this section, the power inverter is tested with all stages joined together. In the

final design, an adjustable feedback resistor is added to fine tune the high output

DC bus from the boost stage. As a result of this, the output RMS voltage can be set

to 120 VRMS.

Figure 69: The whole system put together

In the full system testing, only resistive loads were considered. Note that the boost

stage provides a regulated 170 V input DC supply into the inverter.

93
Firstly, a 104.4Ω is used at the load and the following are measured:

Load Vin Iin Vout(rms) Iout Frequency Pin Pout D η

(Ω) (V) (A) (V) (A) (Hz) (W) (W) (%) (%)

104.4 24 7.3 115 1.076 60 175.2 123.74 11.3 70.6

Table 18: Spot testing of resistive load

Now, the inverter is tested at full-load of 50Ω. The full video of this demonstration

is emailed to Dr. Natarjan Krishnamoorthy.

Figure 70: Input voltage and current into the inverter

Figure 71: Output voltage of boost stage

94
 Figure 72: Output current of inverter

Figure 73: Output voltage waveform and measurement

95
From Fig. 60 to Fig. 63:

Load Vin Iin Vout(rms) Iout Frequency Pin Pout D η

(Ω) (V) (A) (V) (A) (Hz) (W) (W) (%) (%)

50 23.90 16.30 119.30 2.28 60 389.57 272.36 45 70

Table 19: Full load testing of the overall system

Therefore, as a conclusion to the inverter, it is found that the inverter can provide

70% efficiency at full load. The output is a pure sinusoidal wave with 120 Vrms.

96
5.4 Harmonic Content of sinusoidal output

Since the inverter is a pure sine wave inverter measurements were taken to see the

harmonic content of the sinusoidal output. The Fig. 74 show a substantial third

harmonic present at the output. This is largely due to the presence of intentional

dead time during switching of half bridge MOSFETs from on to off to prevent

simultaneous conduction.

Figure 74: Harmonic content of output waveform

To solve the above issue, a higher order lowpass LC filter needs to be added at the

output to supress the third harmonic. Due to time constraint this function wasn’t

implemented on this project.

97
6. Economic and Project Management Analysis
6.1 Bill of Materials

Part # Part Name Description Qty Units Supplier Unit Cost

Cost

ATmega2560 Arduino Microcontroller 1 Each Arduino $ $ 35.00

MEGA2560 35.00

E65-N27 Transformer Use for high- 1 Each Digi-Key $ 7.97 $ 15.94

Ferrite Core frequency

transformer in

Push-Pull

495-5498-ND Bobbin Coil Bobbin for 1 Each Digi-Key $ 6.70 $ 6.70

transformer

IR2010PBF- IC Gate High/Low side 4 Each Digi-Key $ 6.02 $ 24.08

ND Driver drivers for

inverter

SG3525ANGO IC Regulator Push-pull gate 3 Each Digi-Key $ 1.55 $ 4.65

S-ND Control drivers

BER169-ND Thermal Pad Cooling 20 Each Digi-Key $ 0.13 $ 2.56

MOSFETs

STP160N75F3 N-Channel 75v, 120A, TO- 4 Each Digi-Key $ 6.51 $ 26.04

MOSFET 220 package

STTH8R06 Rectifier For full-wave 6 Each Digi-Key $ 1.99 $ 11.94

Diodes rectification

Total 40 $ 126.91

Table 20: Bill of materials

98
In addition to the parts listed in Table 14, there are salvaged and already owned

parts tabulated in Table 15.

Part # Part Name Description Qty Units Supplier Unit Cost

Cost

77192A7 Kool Mu Toroidal Output 1 Each Magnetics N/A N/A

Core inductor

(SALVAGED)

N/A Various resistors N/A N/A N/A N/A N/A N/A

and capacitors

N/A Regular copper N/A N/A N/A N/A N/A N/A

wires: stranded

and unstranded

N/A Connector probes Use with scope N/A N/A N/A N/A N/A

N/A Solid enameled N/A N/A N/A N/A N/A N/A

Copper Magnet

Wire (From

electrical lab)

N/A Choke inductors Filtering sine 3 N/A N/A N/A N/A

(SALVAGED) wave

N/A Designed printed Placing 2 N/A N/A N/A N/A

circuit board electronic

(PCB) components

N/A Capacitor bank Input/output 2 N/A N/A N/A N/A

N/A Various N/A N/A N/A N/A N/A N/A

measuring

equipment

N/A Voltage power N/A N/A N/A N/A N/A N/A

supplies

N/A Breadboards N/A N/A N/A N/A N/A N/A

99
N/A Soldering N/A N/A N/A N/A N/A N/A

equipment

N/A Various IC gate Testing N/A N/A N/A N/A N/A

drivers, purposes

MOSFET's and

diodes

(SALVAGED)

N/A Voltage/current Testing N/A N/A N/A N/A N/A

transducer purposes

N/A Relay Safety N/A N/A N/A N/A N/A

purposes

N/A Total 8 N/A

Table 21: Salvaged/already available parts

With the purchase of this inverter, to estimate whether it was justified or now

requires an projection of profits that it will generate during its period of service.

However, in design projects such as this one, the inverter was built upon

principles, engineering design calculations and judgment so its design is time

invariant assuming little technological advances. The engineering design would

not change significantly change over a large time period.

100
7. Conclusion
The goal of this project was to design a high efficiency power inverter. The design

and implementation was very challenging. Due to this, lots of concepts were

learned and understood. Numerous learning outcomes achieved through the

complete project design.

7.1 Project Outcomes

The design of the power inverter was done through three stages: boost stage,

inverter stage and load stage. The project met the expectations in most cases.

However, in certain loads the efficiency was not as expected (around 70%). The

inverter manufactured in the project was built at around $127, with components

recycled from inactive electronic devices. The overall efficiency of the inverter was

a 70% (at full load), while it was not the desired output (>82%), it is able to provide

a regulated output to an AC load at a lower power. All of us learned a lot about

project management and engineering design.

7.2 Future Work

Some of the design modifications and improvements that can be implemented

into the project are listed below:

 Addition of a better AC load regulation and protection scheme using PID

control

 Implement a 5-level sPWM for better THD performance at the output

 Include temperature protection for all semiconductor switches

 Improve the overall efficiency of the system by using exotic converter

types such as resonant converters and synchronous rectification at the

boost stage

101
 Use better MOSFETs with much lower RDS_ON to reduce conduction losses

at the H-bridge

 Add non energy dissipative snubber circuit to protect power

semiconductor switches

 Perform detailed analysis of harmonic distortion at the output of the

inverter

 Design an enclosure to house the components and PCB

102
Bibliography

[1] “Arduino - PWM,” Arduino - PWM. [Online]. Available:

https://www.arduino.cc/en/Tutorial/PWM. [Accessed: 23-Dec-2016].

[2] “Arduino - SecretsOfArduinoPWM,” Arduino - SecretsOfArduinoPWM.

[Online]. Available: https://www.arduino.cc/en/Tutorial/SecretsOfArduinoPWM.

[Accessed: 23-Dec-2016].

[3] ATMEL, “Atmel ATmega640/V-1280/V -1281/V-2560/V-2561/V,” 8-bit Atmel

Microcontroller with 16/ 32/64KB In-System Programmable Flash, Feb-2014.

[Online]. Available: http://www.atmel.com/Images/Atmel-2549-8-bit-AVR-

Microcontroller-ATmega640-1280-1281-2560-2561_datasheet.pdf. [Accessed: 15-

Oct-2016].

[4] “Calculating power loss in switching MOSFETs | EE Times,” EETimes.

[Online]. Available: http://www.eetimes.com/document.asp?doc_id=1278970.

[Accessed: 4-Mar-2017].

[5] “Delta modulation,” Wikipedia, 15-Mar-2017. [Online]. Available:

https://en.wikipedia.org/wiki/Delta_modulation. [Accessed: 5-Mar-2017].

[6] “Description of Pulse Width Modulation blog post,” Description of Pulse

Width Modulation blog post | Acroname. [Online]. Available:

https://acroname.com/blog/description-pulse-width-modulation-blog-post.

[Accessed: 01-Apr-2017].

103
[7] “E-Learning Power Electronics,” E-Learning Power Electronics. [Online].

Available: http://www.powere.dynamictopway.com/i10.htm. [Accessed: 10-Mar-

2017].

[8] M. J. Fisher, Power Electronics. Boston, MA: PWS Kent, 1991.

[9] J. Honda and J. Cerezo, “Class D Audio Amplifier Design,” Class D Audio,

08-Oct-2003. [Online]. Available: http://www.irf.com/product-

info/audio/classdtutorial.pdf. [Accessed: 12-Feb-2017].

[10] International Rectifier, “HV Floating MOS-Gate Driver ICs,” Application

Note. [Online]. Available:

http://web.mit.edu/6.131/www/document/float_drive.pdf. [Accessed: 2-Feb-

2017].

[11] R. Locher, “AN-558 Introduction to Power MOSFETs and Their

Applications,” Introduction to Power MOSFETs and Their Application. [Online].

Available: http://www.ti.com/lit/an/snva008/snva008.pdf. [Accessed: 5-Jan-2017].

[12] N. Mohan, T. M. Undeland, and W. P. Robbins, Power electronics:

converters, applications, and design. New Delhi, India: Wiley India, 2008.

[13] V. R. Moorthi, Power electronics: devices, circuits and industrial

applications. New Delhi: Oxford Univ. Press, 2010.

104
[14] “Power MOSFET - Vishay.” [Online]. Available:

http://www.bing.com/cr?IG=3D6922ACB2F740359BEC7D. [Accessed: 5-Jan-2017].

[15] Terbytes, “Terbytes/Arduino-Atmel-sPWM,” GitHub, 14-Mar-2016. [Online].

Available: https://github.com/Terbytes/Arduino-Atmel-sPWM. [Accessed: 13-

Jan-2017].

[16] I. T. AG, “Products,” Infineon Technologies. [Online]. Available:

http://www.infineon.com/cms/en/product/power/gate-driver-ics/level-shift-gate-

drivers/high-and-low-side-

drivers/IR2010/productType.html?productType=5546d462533600a401533d22608a

563b. [Accessed: 1-Feb-2017].

[17] alldatasheet.com, “IRFP460 Datasheet(PDF) - International Rectifier,”

IRFP460 Datasheet(PDF) - International Rectifier. [Online]. Available:

http://www.alldatasheet.com/datasheet-pdf/pdf/68529/IRF/IRFP460.html.

[Accessed: 1-Feb-2017].

[18] “B66388A2000T1,” EPCOS (TDK) | Magnetics - Transformer, Inductor

Components | DigiKey. [Online]. Available: https://www.digikey.ca/product-

detail/en/epcos-tdk/B66388A2000T1/495-5498-ND/3914878. [Accessed: 27-Oct-

2016].

[19] “FAN7382,” FAN7382 PDF Datasheet 625V. [Online]. Available:

https://www.fairchildsemi.com/products/power-management/mosfet-igbt-gate-

drivers/half-bridge-drivers/FAN7382.html. [Accessed: 10-Feb-2017].

105
[20] “MIC4421A,” Microchip Technology Inc. [Online]. Available:

http://www.microchip.com/wwwproducts/en/MIC4421A. [Accessed: 06-Apr-

2017].

[21] “SG3525ANG,” ON Semiconductor | Integrated Circuits (ICs) | DigiKey.

[Online]. Available: https://www.digikey.ca/product-detail/en/on-

semiconductor/SG3525ANG/SG3525ANGOS-ND/920304. [Accessed: 27-Oct-

2016].

[22] “STP160N75F3,” STP160N75F3 - N-channel 75 V, 3.5 mOhm typ., 120 A

STripFET(TM) Power MOSFET in TO-220 package - STMicroelectronics. [Online].

Available: http://www.st.com/en/power-transistors/stp160n75f3.html. [Accessed:

15-Sep-2016].

[23] “STTH8R06,” STTH8R06 - TURBO 2 Ultrafast High Voltage Rectifier -

STMicroelectronics. [Online]. Available: http://www.st.com/en/diodes-and-

rectifiers/stth8r06.html. [Accessed: 02-Nov-2016].

[24] “Transformer Cores, E, E65/32/27, N27, 147 mm, 535 mm,” Newark element14.

[Online]. Available:

http://www.newark.com/epcos/b66387g0000x127/transformer-cores-e-e65-32-

27/dp/07X1867. [Accessed: 15-Jan-2017].

[25]"Application note - AN2794 - 1 kW dual stage DC-AC converter", ST, 2017.

[26] 2017. [Online]. Available: https://www.solo-labs.com/cabling-effects-

selecting-right-cable/. [Accessed: 15- Apr- 2017].

106
[27] P. Sen, Modern power electronics, 1st ed. New Delhi: S. Chand & Co., 2004.

[28] R. Mack, S. Maniktala, M. Brown, N. Kularatna and Marty Brown., Power

Sources and Supplies: World Class Designs (Newnes world class designs series), 1st ed.

Newnes, 2008.

[29] C. Basso, Switch-mode power supplies, 1st ed. New York: McGraw-Hill, 2008.

[30] A. Pressman, K. Billings and T. Morey, Switching power supply design, 1st ed.

New York: McGraw-Hill, 2009.

[31] N. Mohan, W. Robbins and T. Undeland, Power electronics, 1st ed. Hoboken,

NJ: Wiley, 2007.

[32] M. Rashid, Power electronics, 1st ed. Boston u.a: Pearson, 2014.

[33] 2017. [Online]. Available: http://www.cde.com/resources/technical-

papers/design.pdf. [Accessed: 16- Apr- 2017].

[34]"Application note - AN2794 - 1 kW dual stage DC-AC converter", ST, 2017.

107
Appendix
A.1 PCB for DC-DC converter stage

Figure 75: PCB for DC-DC converter

108
A.2 PCB for Inverter stage

Figure 76: Inverter PCB

109
A.3 Microcontroller Schematic

Figure 77: μController Module

110
A.4 DC-DC Circuit Schematic

Figure 78: DC-DC circuit schematic

111
A.5 Inverter Circuit Schematic

Figure 79: Inverter circuit schematic

112
A.6 Lookup Table MATLAB Code
%% Matlab program written to generate sPWM amplitude values for
%% different switching frequencies
%%
%%
%%
%%
%%
%%
%% Time specifications:
1 clk=16000000; %Arduino Internal Clock
2 swFreq=20000; %Carrier Frequency of PWM
3 Fsig = 60; %Modulating/Output Frequency of PWM
4 samples = (swFreq/Fsig)/2; % samples for one half sine wave
5 StopTime = (1/Fsig)/2; % Stops at 180 degrees
6 dt = StopTime/samples; % seconds per sample
7 t = (dt:dt:StopTime)'; % time for 1st half cycle
8 t2 = (dt+dt:dt:2*StopTime); % time for 2nd half cycle
9 A=clk/swFreq; %ICRn (TOP) Value
10
11 %% Sine wave: Lookup Table Values
12 x = A*sin(2*pi*Fsig*t);
13
14 %Used to insert the zero values into each lookup table for the times
15 %when the MOSFET needs to be OFF
16 ZPadding=zeros(1,samples);
17 y1=transpose(round(x));
18 ypos=horzcat(y1,ZPadding);
19 yneg=horzcat(ZPadding,y1);
20
21 %Used to output the lookup table values to easily copy over
22 % to the Arduino
23 csvwrite('Positive Cycle.txt',ypos);
24 csvwrite('Negative Cycle.txt',yneg);
25 subplot(2,1,1)
26 plot(t2,ypos)
27 xlabel('time (in seconds)');
28 ylabel('OCR Value');
29 title('Positive Signal versus Time');
30 subplot(2,1,2)
31 plot(t2,yneg)
32 xlabel('time (in seconds)');
33 ylabel('OCR Value');
34 title('Negative Signal versus Time');

113
A.7 Microcontroller Code
1 #include <avr/io.h>
2 #include <avr/interrupt.h>
3 #include <LiquidCrystal.h>
4 #include <OneWire.h>
5 #include <DallasTemperature.h>
6 #include <LiquidCrystal.h>
7 int lookUp1[] = {15, 30, 45, 60, 75, 90, 105, 120, 135, 150, 165,
179, 194, 209, 223, 238, 252, 266, 280, 294, 308, 322, 336, 350,
363, 377, 390, 403, 416, 429, 441, 454, 466, 478, 490, 502, 514,
525, 537, 548, 559, 569, 580, 590, 600, 610, 620, 629, 638, 647,
656, 664, 673, 681, 689, 696, 703, 711, 717, 724, 730, 736, 742,
747, 753, 758, 762, 767, 771, 775, 778, 782, 785, 788, 790, 792,
794, 796, 797, 798, 799, 800, 800, 800, 800, 799, 798, 797, 795,
794, 792, 789, 787, 784, 781, 777, 774, 770, 765, 761, 756, 751,
746, 740, 734, 728, 722, 715, 708, 701, 694, 686, 678, 670, 662,
653, 644, 635, 626, 616, 607, 597, 587, 576, 566, 555, 544, 533,
521, 510, 498, 486, 474, 462, 450, 437, 424, 412, 399, 385, 372,
359, 345, 332, 318, 304, 290, 276, 262, 247, 233, 218, 204, 189,
175, 160, 145, 130, 115, 100, 85, 70, 55, 40, 25, 10, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0};
8
9 int lookUp2[] = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 15, 30, 45, 60,
75, 90, 105, 120, 135, 150, 165, 179, 194, 209, 223, 238, 252, 266,
280, 294, 308, 322, 336, 350, 363, 377, 390, 403, 416, 429, 441,
454, 466, 478, 490, 502, 514, 525, 537, 548, 559, 569, 580, 590,
600, 610, 620, 629, 638, 647, 656, 664, 673, 681, 689, 696, 703,
711, 717, 724, 730, 736, 742, 747, 753, 758, 762, 767, 771, 775,
778, 782, 785, 788, 790, 792, 794, 796, 797, 798, 799, 800, 800,
800, 800, 799, 798, 797, 795, 794, 792, 789, 787, 784, 781, 777,
774, 770, 765, 761, 756, 751, 746, 740, 734, 728, 722, 715, 708,
701, 694, 686, 678, 670, 662, 653, 644, 635, 626, 616, 607, 597,
587, 576, 566, 555, 544, 533, 521, 510, 498, 486, 474, 462, 450,
437, 424, 412, 399, 385, 372, 359, 345, 332, 318, 304, 290, 276,
262, 247, 233, 218, 204, 189, 175, 160, 145, 130, 115, 100, 85, 70,
55, 40, 25, 10};
10 //PWM Output Pins

114
11 int spwm1 = 11;
12 int spwm2 = 12;
13 int pin9 = 9;
14 int pin10 = 10;
15
16 //Analog Input Pins
17 int Voltage = 1;
18 int Current = 3;
19 int Bus = 5;
20
21 //Digital Output Pins
22 int relayControl = 21;
23 int Temperature = 24;
24 int OverCurrent = 26;
25 int Error = 30;
26 int flag = 0;
27
28 //LCD Pins
29 LiquidCrystal lcd(8, 2, 4, 5, 6, 7);
30
31 //(rs, enable, d4, d5, d6, d7)
32 // Data wire for the temperature sensor on pin 3
33 #define ONE_WIRE_BUS 24
34
35 //Setup a oneWire instance to communicate with any OneWire devices
36 OneWire oneWire(ONE_WIRE_BUS);
37
38 //Pass our oneWire reference to Dallas Temperature. (Onewire and
39 //dallas temperature libraries had to be downloaded for the DS18b20
40 //temperature sensor)
41 DallasTemperature sensors( & oneWire);
42
43 void setup() {
44 Serial.begin(9600);
45
46 // Register initialization
47 TCCR1A = 0b10100010;
48 TCCR1B = 0b00011001;
49 TIMSK1 = 0b00000001;
50
51 //Input Compare Register, TOP value, Arduino Clock/Desired
52 //Switching frequency
53 ICR1 = 800;
54
55 sei(); // Enable global interrupts.
56
57 pinMode(spwm1, OUTPUT);
58 pinMode(spwm2, OUTPUT);
59 pinMode(pin9, OUTPUT);
60 pinMode(pin10, OUTPUT);
61
62 pinMode(relayControl, OUTPUT);

115
 63 pinMode(OverCurrent, OUTPUT);
64 pinMode(Error, OUTPUT);
65
66 //LCD Setup
67
68 while (flag == 0) {
69 lcd.begin(16, 2);
70 sensors.begin(); // starts the temperature sensor reading
71 sensors.requestTemperatures();
72 // Send the command to get temperature readings
73 // set cursor position to start of first line on the LCD
74
75 // read the input on analog pin 1:
76 int voltageSensor = analogRead(Voltage);
77
78 // Convert the analog reading (which goes from 0 - 1023)
79 // to a voltage (0 - 5V):
80 float fivevolt = voltageSensor * (5.0 / 1023.0);
81
82 // Linearly Interpolate [0,5]V to [0,150]V
83 float voltageOutput = 30 * voltageSensor * (5.0 / 1023.0);
84 lcd.setCursor(0, 0);
85 lcd.print("Voltage: ");
86 lcd.print(voltageOutput);
87 lcd.print("V");
88
89 // read the input on analog pin 3:
90 int currentSensor = analogRead(Current);
91
92 // Convert the analog reading (which goes from 0 - 1023)
93 // to a voltage (0 - 5V):
94 float currentOutput = currentSensor * (5.0 / 1023.0);
95
96 lcd.setCursor(0, 1);
97 lcd.print("Current: ");
98 lcd.print(currentOutput);
99 lcd.print("A");
100
101 delay(1000);
102 lcd.clear();
103
104 // read the input on analog pin 5:
105 int busSensor = analogRead(Bus);
106
107 // Convert the analog reading (which goes from 0 - 1023)
108 // to a voltage (0 - 5V):
109 float busVoltage = busSensor * (5.0 / 1023.0);
110 // Linearly Interpolate [0,2.5]V to [0,180]V
111 float busOutput = busVoltage * 72;
112 lcd.setCursor(0, 0);
113 lcd.print("Bus Volt: ");
114 lcd.print(busOutput);

116
115 lcd.print("V");
116
117 if (busOutput >= 180 || busOutput <= 150.0) {
118 digitalWrite(relayControl, HIGH);
119 // If busoutput > 180V then turn on the relay
120 // to shutdown the Inverter
121 digitalWrite(Error, HIGH);
122 flag = 1;
123 }
124 if (currentOutput >= 4) {
125 // If the current is > 4A then pin26 set to HIGH
126 // to the over current protection circuit
127 digitalWrite(OverCurrent, HIGH);
128 digitalWrite(Error, HIGH);
129 flag = 2;
130 }
131
132 // Temperature reading
133 lcd.setCursor(0, 1);
134 lcd.print(" Temp: ");
135 lcd.print(sensors.getTempCByIndex(0));
136 // You can have more than one DS18B20 on the same bus;
137 // 0 refers to the first IC on the wire
138 lcd.print("C");
139
140 delay(1000);
141 lcd.clear();
142
143 float Power = voltageOutput * currentOutput;
144 lcd.print(" Power: ");
145 lcd.print(Power);
146 lcd.print("W");
147 delay(1000);
148 lcd.clear();
149 }
150 if (flag == 1) {
151 lcd.setCursor(0, 0);
152 lcd.print("Error");
153 lcd.setCursor(0, 1);
154 lcd.print("Check Bus O/P");
155 }
156 if (flag == 2) {
157 lcd.setCursor(0, 0);
158 lcd.print("Error");
159 lcd.setCursor(0, 1);
160 lcd.print("Check Current");
161 }
162 }//end of void setup
163
164
165
166

117
167 void loop() {}
168 ISR(TIMER1_OVF_vect) {
169 static int num;
170 OCR1A = lookUp1[num];
180 if (num < 166) {
190 digitalWrite(pin9, LOW);
191 digitalWrite(pin10, HIGH);
192 } else if (num > 166) {
193 digitalWrite(pin9, HIGH);
194 digitalWrite(pin10, LOW);
195 }
196
197 OCR1B = lookUp2[num];
198 if (++num >= 332) {
199 num = 0; // Reset num
200 }
201 }

118
A.8 American Wire Gauge Table

Table 22: American Wire Gauge (AWG)

119
Index
core window area, 39
3
cross-conduction, 78
3-level, 66 current density, 43
Current penetration depth, 42
A
current transducer, 81, 101
adjustable feedback resistor, 95
alternate phases, 25 D

amplitude, 25, 26, 27 DC Bus Regulation, 83, 84

Arduino, 20, 25, 26, 29, 66, 67, 70, DC-DC converter, 20, 21, 22, 32,
71, 72, 73, 76, 78, 80, 99, 103, 58, 76, 108
105, 113 Decoupling, 77
Arduino microcontroller, 20, 66 digital high and digital low, 80
attenuation, 56 drain current, 79
driver, 23, 25, 31, 74, 76, 78
B
dual-slope operation, 68
bipolar output voltage, 24
duty cycle, 23, 24, 26, 27, 31, 33,
boost converter, 21
34, 35
bootstrap capacitance, 78
bootstrap diode, 77 E

breakdown, 23, 32, 34 efficiency, 23, 26, 28, 29

Equivalent Series Resistance (ESR),
C
36
center tap wires terminated, 59
center-tapped, 23 F

compare registers, 69 fast PWM mode, 68

conduction, 24, 54, 55, 79 fast switching, 21, 32, 79
Continuous Conduction Mode, 36 filtered response, 28
Control Register, 66 filtering action, 56
Converter losses, 53 flat topped, 34

120
forward converter, 24 K
freewheeling diode, 24 Kool Mu toroidal core, 46, 48
frequency, 24, 25, 27, 31, 37, 42,
43, 47, 58, 65, 68, 70, 73, 79 L

frequency correct, 68 LCR meter, 61

full load, 91, 98, 102 leakage inductance, 32, 37, 61,
fundamental frequency, 56 63, 64
linear switch controller, 26
G
Logic, 78
Gate capacitor, 78
low side, 24, 78
gate driver, 78
M
H
magnetic field, 21
half bridge, 23 MATLAB, 70
H-Bridge, 24, 25, 74, 75, 76, 80, 81 Maximum input RMS current, 34
high frequency, 21, 23, 32, 43, 46, Maximum Operating Flux Density,
68 38
high power, 16, 32, 78 modulator, 24
high side, 24 MOSFET, 21, 25, 32, 34, 53, 54, 55,
high-power loads, 79 66, 70, 76, 77, 78, 79, 80, 105
mounting capacitance, 53
I

implementation, 58 N
inductive loads, 92 N-Channel, 32, 76, 78
initial, 50
internal registers, 66 O

Interrupt Service Routine (ISR), 67 oersteds, 49, 50

intrinsic, 79 output diodes, 53
Inversion, 25 Output filter capacitor, 36
Output Power, 31

121
output signal, 27 S
Overcurrent, 83, 84 saturation, 39, 41, 46
over-voltages, 32 Secondary maximum RMS
current, 35
P
shorted, 64
parallel connection, 55
sine wave, 16, 29, 30, 81
parasitic capacitances, 79
Sinusoidal waveform, 27
Passive, 9, 88
Skin effect, 43, 44
PCB, 37, 58, 108, 109
snubber, 13, 52, 53
peak current values, 38
soft-start circuitry, 31
permeability, 48, 50
SPWM, 25, 29, 68, 74, 80
phase correct, 68
steady-state operation, 21
polarity dot, 62
switching converter, 21
power dissipated, 52
switching losses, 27, 53, 54, 79
power factor, 93
switching period, 33
power inverter, 16, 20
prototype, 74 T
pulsing waveforms, 29 the Input Compare Register
pure sine wave, 16, 28 (ICRn), 68
push-pull topology, 23 Total copper losses, 46
PWM control, 23 total harmonic distortion, 26
transformer, 22, 23, 24, 32, 37, 38,
R
39, 40, 42, 46, 58, 59, 60, 61, 65
RDS (on) resistance, 24
transformer isolated, 22, 23
reactive current, 56
Transformer turns ratio, 34
rectifier diodes, 25
Regulation, 38 U
ripple current, 35 Uninterruptible power supply, 16
unload test, 85

122
 V Wire selection, 42, 51

Voltage mode, 32 Z
voltage transducer, 81
zero-crossing level, 25, 26, 29
W

waveform coefficient, 39

123