Zero Voltage Switching Resonant Converter Using 0.6μm ...

Zero Voltage Switching Resonant Converter Using 0.6μm Technology

BY

Ashish Mishra, B.E.

A Thesis

In

ELECTRICAL AND COMPUTER ENGINEERING

Submitted to the Graduate Faculty Of Texas Tech University in

Partial Fulfillment of the requirements for

the degree of

MASTER OF SCIENCES

Approved

Dr. Stephen Bayne

Chair of Committee

Dr. Changzhi Li

Dr. Mark Sheridan

Dean of the Graduate School

May, 2016

Texas Tech University, Ashish Mishra, May 2016

ii

ACKNOWLEDGEMENTS

I would like to express my thanks to my committee chair Dr. Stephen Bayne, who has

given me an opportunity to do research under him. He has the attitude and substance of

a genius. Without his guidance and help this thesis would not have been possible.

Secondly, I would like to thank my committee member Dr. Changzhi Li for his support and

concern during my research. He helped me to overcome the hurdles during my project.

I would like to thank my parents and sister for being a constant support throughout my

life and giving me the opportunity to study in the US by making a lot of sacrifices and

suffering lots of odds. Their blessing and love has made me what I am today.


iii

TABLE OF CONTENTS

ACKNOWLEDGEMENTS ii

ABSTRACT v

LIST OF TABLES vi

LIST OF FIGURES vii

1. INTRODUCTION 1

1.1 Introduction to Power Electronics 1

1.2 Buck Converter 3

1.2.1 On time event 4

1.2.2 Off time event 4

1.3 Buck Converter modes of operation 5

2. CADENCE AND C5 PROCESS 8

2.1 Cadence 8

2.1.1 Cadence User Interface 8

2.2 C5/AMI06 process 10

3. ASTABLE MULTIVIBRATOR 13

3.1 Multivibrator 13

3.2 Astable Multivibrator 14

3.2.1 Operation 15

3.2.1.1 Stage 1 15

3.2.1.1 Stage 2 15

4. OPERATIONAL AMPLIFIER 19

4.1 Introduction to operational amplifier 19


iv

4.2 OPAMP performance parameters 19

4.3 OPAMP using CMOS 21

4.3.1 Single Stage OPAMP 21 4.3.2 Two stage OPAMPs 23

4.4 Compensation Circuit 24

4.5 Comparator 27

5. RESONANT CONVERTER 31

5.1 Introduction 31

5.2 Resonant Switch 32

5.3 ZVS-MRC Buck converter 33

5.4 Operation of ZVS-MRC Buck converter 34

5.5 Design: Converter schematic and output 35

6. COMPLETE CIRCUIT OF ASTABLE MULTIVIBRATOR 44 AND BUCK CONVERTER 6.1 Results 44

7. FUTURE WORK: WIRELESS POWER TRANSFER 52

7.1 Introduction 52

7.2 Proposed model 53

8. CONCLUSION 55

REFERENCES 57


v

ABSTRACT

There has been a rapid advancement in the technology for the development of high

performance DC-DC converters. Resonant switching buck converter is one such power

conversion topology and is designed to achieve maximum power density and the ability

to regulate the output voltage for different input voltages. This thesis discusses the design

of a zero voltage switching buck converter operating at 1MHz. The design is implemented

on C5 process.

The design accommodated input voltage in the range of 5V-8V DC and gives a constant

output voltage of 3V. The output power level is 300mW.The simulation was performed

on Cadence platform to verify the operating conditions of the proposed converter. Finally

this resonant converter was compared to the conventional converter to verify the

efficiency increase and switching losses decrease.


vi

LIST OF TABLES

1 C5 Design rules 11

2 Component values used in Astable Multivibrator 17

3 Width and Length ratios of Compensation OPAMP 26

4 Comparator Circuit Width and Length ratios of MOSFETs 28

5 ZVS-MRC Buck Converter components 37

6 ZVS-MRC Components Width and Length 38

7 Results 51


vii

LIST OF FIGURES

1 Basic block diagram of a power electronic system 1

2 DC-DC converter types 2

3 On and Off events of buck converter 3

4 Buck Converter waveforms in CCM 5

5 Buck Converter waveforms in DCM 6

6 Common Interface window 9

7 Library Manager Window 9

8 Design Window 10

9 Text Window 10

10 Layer map of C5 process 11

11 Components fabricated using C5 process. 12

12 Multivibrator Types 13

13 Astable Multivibrator Circuit 14

14 Schematic of Astable Multivibrator 16

15 Output wave form of Astable Multivibrator 17

16 Astable Multivibrator with inductor across both the outputs 18

17 Typical OPAMP Frequency response 20

18 Single Stage OPAMP 22


viii

19 Two Stage Operational Amplifier 23

20 Compensation Circuit 25

21 Compensation Circuit in Cadence 26

22 Bode Plot of Compensation Circuit 27

23 Comparator Circuit 28

24 Comparator Circuit in Cadence 29

25 Output of comparator 30

26 Zero voltage resonant switch 32

27 Buck Converter 33

28 ZVS-MRC 34

29 Schematic of Power Stage 36

30 Resonant circuit Schematic 37

31 ZVS-MRC Buck 38

32 ZVS-MRC Buck in Cadence 40

33 Block Diagram 41

34 Schematic of ZVS-MRC Buck 41

35 Output Voltage at 5, 6, and 8V input voltages 42

36 Gate Voltage of MOSFET and Voltage across MOSFET 42

37 Complete System in Cadence 45


ix

38 Output of Astable Multivibrator 46

39 System Schematic after placing Pins 47

40 System symbol 48

41 Output of the complete system 48

42 Output from -20°C to 100°C temperature range 49

43 System Layout 50


1

Chapter 1: Introduction

1.1 Introduction to Power Electronics

Power Electronics is defined as the process and control of power by providing voltage and

load to suit the load. It is the technology behind the switch mode power supplies,

converters, inverters, motor drives and motor soft starters. It comprises of power

semiconductor devices such as MOSFETs, BJTs, IGBTs active components like resistors and

passive ones like inductors and capacitors. [1, 2]

Figure 1 shows the block diagram of a power conversion system. The power conversion

systems are classified into four categories depending on the input and output power:

a. ac to ac (ac to ac converters)

b. ac to dc (rectifiers)

c. dc to dc ( dc to dc converters)

d. dc to ac (inverters)

Figure 1. Basic block diagram of a power electronic system [1]

This thesis is focused on the design of dc-dc converter. Prior to the development of power

semiconductor devices and related technologies, the dc voltage was converted to ac


2

voltage with the use of vibrator and this ac voltage was either stepped up or stepped

down as per the requirement. Finally this ac voltage was converted back to dc using

rectifier. This procedure was expensive and efficiency was very less. With the advent of

power semiconductor and integrated circuit the conversion of voltages has become cheap

and efficiency as also increased.

DC-DC converters are mostly used in portable electronic devices like mobile phones,

laptops etc., switch mode power supplies and dc motor drives. Portable devices require

batteries as a primary source of power. DC-DC converters have the ability to change the

voltage from a low to high hence they can save the space and money for achieving high

voltage requirement. Switch mode power supplies are able to regulate an unregulated

power supplies. Figure 2 shows different classification of dc-dc converters. [3]

The DC- DC converters are basically divided into two broad categories:

a. Non isolated (no transformer)

b. Isolated (transformer)

Figure 2. DC-DC converter types [3]

The DC-DC converter use switches to change the voltage level from input to output. By

controlling the switch on and off duration, the average output voltage is controlled and


3

level shifted. This paper focusses on Buck converter, which is a non-isolated type DC-DC

converter.

1.2 Buck Converter

Buck converter is a voltage step-down and current step up converter. Before the buck

converter was in use, the DC voltage step down operation was done with the help of linear

regulators. The linear regulators were very inefficient as it used to dissipate excess energy

in the form of heat and did not perform current step-up. The buck converter on the other

hand are very efficient which makes them suitable for various tasks like stepping down

main voltage of a computer to very low voltage as required by chips. [4]

Buck converter is basically a square wave generator followed by LC filter. The operation

can be divided into two categories:

i. On event

ii. Off event

Figure 3. On and Off events of buck converter [5]


4

1.2.1 On time event

Figure 3 (i) shows the on time, configuration of buck converter. When the switch is closed

the current in the inductor L rises. On applying voltage to the inductor L affected by the

resistor in series rLf gives an exponential current. The shape is affected by the following

equation

𝐼𝐿(𝑡) =𝑉𝐿

𝑟𝐿𝑓(1 − 𝑒−𝑡

𝑟𝐿𝑓𝐿⁄ ) 1.1

Here VL is the current applied across the inductor L and rLf is the inductor series resistor.

Since series resistor rLf is kept very small a new equation is obtained in the vicinity of zero.

𝐼𝐿(𝑡) =𝑉𝐿

𝑟𝐿𝑓(1 − 1 + −𝑡

𝑟𝐿𝑓𝐿⁄ ) =

𝑉𝐿

𝐿𝑡 1.2

Looking at the buck configuration during the on time, one terminal of the inductor sees

Vin and the other is connected to Vout. On applying equation 1.2 to the situation gives the

current reaching the peak value is dependent on the on time of the switch denoted as ton.

𝐼𝑝𝑒𝑎𝑘 = 𝐼𝑣𝑎𝑙𝑙𝑒𝑦 +𝑉𝑖𝑛−𝑉𝑜𝑢𝑡

𝐿𝑡𝑜𝑛 1.3

1.2.2 Off-time event

Figure 3 (ii) shows the off time, configuration of buck converter. The current reaches a

value as set by on time of the switch. The PWM signal is low, so the switch is in the off

state. To counter the collapsing magnetic field, the inductor reverses the voltage polarity

across it. The diode gets forward biased to complete the path of the inductor current. If


5

the voltage drop across the diode is neglected, the one terminal of the inductor is at

ground and the other is at Vout. The valley current in the inductor is given by the equation

1.4. [5]

𝐼𝑣𝑎𝑙𝑙𝑒𝑦 = 𝐼𝑝𝑒𝑎𝑘 −𝑉𝑜𝑢𝑡

𝐿𝑡𝑜𝑓𝑓 1.4

1.3 Buck Converter modes of operation

Buck Converter operates in two modes:

I. CCM

II. DCM

CCM (Continuous Conduction Mode)

Figure 4. Buck Converter waveforms in CCM [6]


6

Figure 4 represents waveforms of a buck converter when it operates in CCM. The current

flows continuously through the inductor [iL>0]. When the switch is turned on, it conducts

the inductor current and the diode becomes reversed biased. Voltage across the inductor

becomes equal to Vin-Vout. When the switch is turned off, the energy stored in the inductor

causes the inductor current IL to flow and the diode becomes forward biased and the path

gets completed. Voltage across the inductor is equal to –Vout. [7]

𝑇𝑟𝑎𝑛𝑠𝑓𝑒𝑟 𝐸𝑞𝑢𝑎𝑡𝑖𝑜𝑛 𝑀 =𝑉𝑜𝑢𝑡

𝑉𝑖𝑛= 𝐷 1.5

Here D= Duty Cycle(𝑇𝑜𝑛

𝑇𝑆𝑤⁄ )

Ton =On time of switch

TSW=total time of switching

DCM (Discontinuous Conduction Mode)

Figure 5. Buck Converter waveforms in DCM [8]


7

In some cases, the energy requirement for the load is very small. The current through the

inductor falls to for some time during a period. In comparison to CCM, here the inductor

gets completely discharged at the end of the period. Figure 5 shows the waveforms of a

buck converter when it is operating in DCM mode.

𝑇𝑟𝑎𝑛𝑠𝑓𝑒𝑟 𝐸𝑞𝑢𝑎𝑡𝑖𝑜𝑛 = 𝑀 =𝑉𝑜𝑢𝑡

𝑉𝑖𝑛=

2

1+√1+8𝐿

𝑅𝑇𝑠𝑤𝐷2

1.6


8

Chapter 2: Cadence and C5 process

2.1 Cadence

Today’s chips have billions of transistors in them, they are even smaller than wavelength

of light. The design of such chips of such complexity while keeping the performance, low

power and easily available in the market is possible only with advanced electronic design

automation (EDA) tools. EDA software and hardware makes it possible to design

individual transistors to the development software before hardware is built. [9]

The Cadence tool kit has several program for different types of application like layout,

schematic, verification and simulation. The open architecture makes it possible to

integrate tools from other sources or its own. The Design Framework II (DFW) is used to

integrate all these tools.

The DFW is the most remarkable feature of Cadence. It provides a common user interface

and a common data base for the tools. It makes it possible to switch between different

applications without converting the data base.

2.1.1. Cadence User Interface

The cadence user interface is graphic and windows, forms and menus.

The main components are as


9

i. Command Interface Window (CIW): This window controls the environment.

Different tools can be accessed through this window. It also serves as log

window for many application. Figure 6 shows a typical CIW. [10]

Figure 6. Common Interface window

ii. Library Manager: It gives overview of different libraries and different contents

within it. Figure 7 shows a typical Library Manager window. [10]

Figure 7. Library Manager Window

iii. Design Window: This window shows the current design. Many different design

windows can be on at the same time which use different or same tools. Figure

8 shows a typical design window. [10]


10

Figure 8. Design Window

iv. Text window: Text windows shown in Figure 9 can be either log, report or the

editor.[10]

Figure 9. Text Window [11]

2.2 C5/AMI06 process

C5 is a non-silicided CMOS process. It has three metal layers, two poly layers and a high

resistance layers. Contacts are stacked. This process is usually used for low voltage

applications. This process supports following design rules as stated in Table 1:


11

Table 1. C5 Design rules [13]

Design Rules Lambda(μm) Feature size(μm)

C5F/N Rules n/a 0.60

SCMOS_SUBM 0.30 0.60

SCMOS 0.35 0.60(after sizing)

(i) (ii)

Figure 10. Layer map of C5 process (i) Common layers (ii) optional technology features

[12]

Figure 10 shows the layers that are present in C5 technology. Some of the components

that can be fabricated with C5 process are NMOS, PMOS, resistors and capacitors.

Resistors are fabricated using high resistor and electrode (poly2) layer and capacitors

using poly1-poly2 layers.


12

Figure 11. Components fabricated using C5 process. [12]


13

Chapter 3: Astable Multivibrator

3.1 Multivibrator

Figure 12. Multivibrator Types

This is a type of electronic circuit used to implement various multistate systems like

oscillator, timers and flip-flops. It has two amplifying devices such as transistors and are

cross coupled by resistors or capacitors. [15]

Figure 12 shows that Multivibrator circuits are classified into three types:

i. Astable Multivibrator: This Multivibrator does not have a stable state and

continuously switches from one state to another. It is also known as relaxation

oscillator.[15]

ii. Monostable Multivibrator: These Multivibrator has one stable and unstable

state. Trigger pulse causes the stable state to go from stable condition to

unstable one, after a time it again returns back to stable state. This circuit is

also called one shot.[15]

iii. Bistable Multivibrator: This circuit has two stable states. External trigger

causes it to switch from one state to another. This is also called as flip flops.[15]


14

Multivibrator circuits are used widely in the applications which require square waves or

timed intervals.

3.2 Astable Multivibrator

Figure 13. Astable Multivibrator Circuit

Monostable and Bistable Multivibrator require external trigger to operate. But the

Astable Multivibrator has an inbuilt trigger mechanism which continuously switches

between the two unstable states.

The Astable Multivibrator is a type of cross coupled transistor switching circuit which has

no stable output. The circuit switches between two states with the maximum transition

rate due to accelerating positive feedback. It is implemented using coupling capacitors,

which transfers voltage changes because voltage cannot change across a capacitor


15

instantly. In each state one transistor is turned on while the other is off. Figure 13 is a

schematic of an Astable Multivibrator circuit.

3.2.1 Operation

The operation of Astable Multivibrator can be divided into two stages:

3.2.1.1 Stage 1(M1 is off, M2 is on)

In the beginning the capacitor C2 is charged to the voltage V1 and the polarity of the

capacitor is as shown in figure 13. The MOSFET M2 is on and it connects the left side

(positive plate) of the capacitor C2 to the ground. Right side of the capacitor is connected

to the gate of M1. Since the right side is connected to ground as a result the left plate is

at voltage –V1. This keeps M1 to be firmly in off state.

Now C2 begins to discharge via resistor R3. As a result the voltage at the right plate rises

from –V1 towards V1. Since the transistor M1 is reversed biased as a result all the current

from R3 goes into C2. C1 which is fully discharged starts to charge up quickly due to low

value resistance R2 and M2 forward biased gate-source junction. Hence C1 restores its

charge and prepares for stage 2, where it acts as time setting capacitor.

3.2.1.2 Stage 2 (M1 is on, M2 is off)

Now the capacitor C1 is fully charged and the right plate of C1 is at positive potential. M1

is on and it connects right plate to ground. Since right plate if C1 is connected to ground


16

as a result the left plate if at –V1 volts which is applied to the gate of M2 to keep it firmly

off.

C1 begins to discharge through resistor R4. As a result the voltage in the left plate of C1

rises from –V1 to V1. In the meantime the capacitor C2 begins to charge through low

value resistor R1 and M1 forward biased gate source junction. C2 restores it charge and

prepares for next stage. Schematic of an Astable Multivibrator circuit is shown in Figure

14 and Table 2 shows the values of the elements used in the design of the circuit.

Figure 14. Schematic of Astable Multivibrator

𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 (𝑓) ≈1

1.38 𝑅 𝐶 3.1


17

Table 2. Component values used in Astable Multivibrator

Figure 15. Output wave form of Astable Multivibrator


18

Figure 16 Astable Multivibrator with inductor across both the outputs

Figure 15 shows the output of the Astable Multivibrator and figure 16 shows the

schematic of the Astable Multivibrator with an inductor between both the output

terminals.


19

Chapter 4: Operational Amplifier

4.1 Introduction to Operational Amplifiers

“An Operational Amplifier (OPAMP) is a DC-coupled high gain electronic voltage amplifier

with a differential input and usually, a single ended output.” [28]

In the current design the OPAMPs are used in the system. OPAMPS are used widely in

various types of applications such as Phased Locked Loops (PLLs), Oscillators, A/D and D/A

Converters, Pulse Width modulators, and in RF applications as Mixers, Transmitters and

Receivers. OPAMPs were originally designed for analog computers to perform

mathematical operations for linear, non-linear and frequency dependent circuits. Due to

wide range of applications the OPAMPs are widely used in Analog circuits. OPAMPs have

little dependence on temperature and manufacturing variations. [27, 28]

4.2. OPAMP performance parameters

OPAMPs are designed to meet the requirements of the application where they are used.

There is a tradeoff between the stability and the slew rate of the OPAMP. For being on

the safe side resulting due to fabrication errors the length and width of the MOSFETs used

in the OPAMPs are not kept selected as the least value.

Parameters for OPAMPS:

1. Open Loop Gain: The function of the OPAMP is the amplification of the input

signal. For a good OPAMP the gain should be high. Ideal OPAMPs have infinite gain


20

𝐺 =𝑉𝑜𝑢𝑡

𝑉𝑖𝑛, but for the real OPAMPs the gain range in the range of 100,000 to 1

million. Figure 17 shows the frequency response of an OPAMP. [29]

Figure 17. Typical OPAMP Frequency response [30]

2. Large Signal Voltage Gain: Large Signal Voltage Gain is usually used in preference

to open voltage gain. This voltage gain is used when the amplifier produces large

voltage usually in the ratio 𝑉

𝑚𝑉 . This covers voltage gain over a large range,

depending on the design, and maximum and minimum temperature. [30]

3. Closed Loop Voltage Gain: In the circuits the high gain of the amplifier is reduced

to operating range by providing negative feedback to it. This method makes the

level response to extend from 0Hz to large frequencies up to MHz. This ultimately

helps in reducing noise and distortion in the circuit. The blue colored line in figure

17 shows the response of the OPAMP with negative feedback. [30]


21

4. Gain Bandwidth Product: The closed loop gain and small signal bandwidth in

OPAM are very closely related. So the term Gain Bandwidth Product is used to

describe these two terms. In the above figure the Gain Bandwidth Product is equal

to 1. 4 𝑀𝐻𝑍 (10𝑋 140𝐾ℎ𝑍). [30]

5. Slew Rate: Slew rate of an OPAMP is used to describe how fast the OPAMP can

change the output voltage in response to the change in input voltage. Higher the

slew rate, faster the OPAM can change the output voltage, and more easily it can

produce large frequency signal.

𝑆𝑙𝑒𝑤 𝑅𝑎𝑡𝑒 (𝑉

𝜇𝑆) = 2𝜋𝑓𝑉𝑝𝑘 =

𝐼𝑐𝑚

𝐶𝑜𝑢𝑡 4.1

4.3 OPAMP using CMOS

OPAMPS are divided in to two types:

1. Single Stage OPAMP

2. Two Stage OPAMP

4.3.1 Single Stage OPAMP

Single stage OPAMPs as shown in figure 18 have low gain but high bandwidth. These

OPAMPs can amplify higher frequencies and are fast but gain is low. For single ended

circuit the mirror pole is not good. This pole decreases the phase margin of the OPAMP.

In some cases the phase margin comes close to 0°.


22

Figure 18. Single Stage OPAMP [29]

Double Ended Single Stage OPAMP

Gain is given as 𝐴𝑣 = 𝑔𝑚1. (𝑟01‖𝑟03) 4.2

Here 𝑟01, 𝑟03 are the internal resistance of the MOSFET M1 and M3

Transfer function is given as 𝐴(𝑠) = 𝑔𝑚1. (𝑟01‖𝑟03)

(1+𝑠

𝑝1)(1+

𝑠

𝑝2)) 4.3

Amplifier has two poles, one at the input and the other at the output. Bandwidth of the

amplifier is dependent on the Load capacitance. The dominating pole p1 is given as :

𝑝1 =1

(𝑟01‖𝑟03)𝐶𝐿 4.4

The second pole is present at higher frequency.

Single Ended Single Stage OPAMP

Gain is given as 𝐴𝑣 = 𝑔𝑚1. (𝑟01‖𝑟03) 4.5


23

Transfer function is given as 𝐴(𝑠) = 𝑔𝑚1. (𝑟01‖𝑟03)(1+

𝑠

𝑧1)

(1+𝑠

𝑝1)(1+

𝑠

𝑝2)) 4.6

This circuit has a dominating pole, a mirror pole and a zero about two times higher than

the dominating pole, hence it is ignored.

Dominating pole is given as 𝑝1 =1

(𝑟01‖𝑟03)𝐶𝐿 4.7

Since the other pole and zero is located at very high frequency hen they can be ignored.

Thus the transfer function can be approximated to 𝐴(𝑠) =𝑔𝑚1.(𝑟01‖𝑟03)

1+𝑠

𝑝1

4.8

Due to the limited gain the two stage amplifiers are used.

4.3.2 Two Stage OPAMPs

Figure 19. Two Stage Operational Amplifier [32]

Two stage amplifier as shown in figure 19 are used to enhance the gain of the OPAMP as

single stage fails to do so. There is a tradeoff since higher number of stages the number


24

of parasitic poles increases. Hence, it is more difficult to make the OPAMP stable and have

a broader unity gain bandwidth. [31]

Stage 1

Gain 𝐴𝑣1 = 𝑔𝑚1 4.9

Output Resistance 𝑅𝑜𝑢𝑡1 = 𝑟02‖𝑟04 4.10

Stage 2

Gain 𝐴𝑣2 = 𝑔𝑚6 4.11

Output Resistance 𝑅𝑜𝑢𝑡2 = 𝑟05‖𝑟06 4.12

Transfer equation 𝐴(𝑠) = −(𝐴𝑣1𝑅𝑜𝑢𝑡1). −(𝐴𝑣2𝑅𝑜𝑢𝑡2) = 𝑔𝑚1(𝑟02‖𝑟04). 𝑔𝑚6𝑟05‖𝑟06 4.13

4.4 Compensation network

Two stage OPAMP was used in the design of Compensation circuit. The compensation

network used is shown in figure 20 and the dimensions of the MOSFETs in table 3.


25

Figure 20. Compensation Circuit

The two stage OPAMP was selected due increase the stability of the system by adjusting

the value of the miller capacitance. Gain of the OPAMP is given as

𝐴𝑣 = 𝑔𝑚11(𝑟06‖𝑟11). 𝑔𝑚4𝑟04‖𝑟08 = 6.4 𝑘𝑑𝑏

The compensation was of type I as the resonant elements around the witch were in series.

So the transfer equation of the resonant part was not possible to calculate. So type I

compensation circuit was selected. Miller capacitance was taken as 2pF to make the

OPAMP stable. Width of the MOSFETs were selected by the trial and error method to

select the smallest value, so that the circuit provides desired results.


26

Table 3. Width and Length ratios of Compensation OPAMP

Figure 21. Compensation Circuit in Cadence


27

Figure 21 shows the schematic of compensation circuit in Cadence and the frequency

response of the circuit is shown in figure 22.

Figure 22. Bode Plot of Compensation Circuit

The phase of the system id 90° and the cross over frequency is 2.5 MHz.

4.5 Comparator

Comparator with slew rate more than 17 𝑉𝜇𝑠⁄ was designed using AMI06 technology in

Cadence. The schematic of the comparator and the dimensions of the MOSFETs are

shown in figure 23 and table 4 respectively. Figure 24 shows the schematic of the

converter in Cadence and the figure 25 shows the output of the comparator.


28

Figure 23. Comparator Circuit

Table 4. Comparator Circuit Width and Length ratios of MOSFETs


29

Figure 24. Comparator Circuit in Cadence

The miller capacitance of the camparator was taken as 2pF. This high value of the miller

capacitance decreases the slew rate of the OPAMP, but it makes the circuit stable. So to

achieve the desired slew rate the current through the common-mode transistoe i.e N6

has to be high to compensate the negative effect of the miller capacitance on the circuit.

So the mumbar of parallel components was increased to increase the commom-mode

current to achieve high slew rate. In the comparator the same trial and error method was

used to select the least width of the MOSFET to achieve the desired results. The length of

the MOSFET is laken as 1μm as 600nm is the least length that can be selected for C5

prosess, but to be on the safe side during fabrication 1μm is preffered.


30

Figure 25. Output of comparator


31

Chapter 5: Resonant Converter

5.1 Introduction

The demand for portable electronic equipment is rising. This is evident from the

replacement of landline telephones with cordless phones and eventually with the mobile

phones. Size of the portable electronic equipments is also decreasing from millimeter

scale to nanoscale so that more passive devices can be integrated to increase the power

density, reduce the manufacturing cost and to make these devices more portable and

convenient. The passive component’s size is inversely related to the frequency of

operation. Therefore, to decrease the size of the device components, higher operating

frequency is selected. Conventional DC-DC converters can provide high efficiency if they

are operated at lower frequency particularly in the scale of KHz, but going up in the

switching frequency decreases the efficiency of these converters as the switching losses

play a significant role at higher frequency. [16]

To reduce switching losses so as to increase the efficiency of the converters the resonant

switching method is used. Resonant switching or soft switching is done either at zero

voltage or at zero current. As a result the losses due to switching is reduced by a very

large amount. Zero Voltage Switching Multi Resonant Buck converter (ZVS-MRC Buck

converter) is one of the advanced ZVS method widely used. To overcome the drawbacks

of a conventional Buck converter, additional elements such as resonant inductor, diode

and capacitors are connected across the switch.


32

Different methods have been proposed in the past to reduce the conduction losses to

achieve higher efficiency [17-22]. Even though the conduction loses are reduced, the

voltage stress on the MOSFET increases ultimately deteriorating the performance of the

converter. Another method to solve this issue was the use of series LLC resonant

converter (SRC) but these converters suffers from high temperature stress and output

voltage increases at light loads, so external dummy loads are connected to it at no load

condition. [23].

5.2 Resonant Switch

For ZVS, resonant capacitor Cs is connected in parallel to the switch to achieve switching

at zero voltage .For the unidirectional switch, the voltage in capacitor Cs oscillates in both

positive and negative half cycles, a diode is connected to the source of the MOSFET. So,

resonant switch operates in full wave mode. If the switch is not unidirectional it operates

in half wave mode. Figure 26 shows the arrangement of the components for the switch

to operate in half wave and full wave mode respectively. [24, 25]

Figure 26. Zero voltage resonant switch


33

5.3 ZVS-MRC Buck converter

In steady state a conventional buck converter is shown in figure 27(b) can be assumed

as a constant current source Ii which supplies power to a constant voltage load V0 by

adjusting the duty cycle of the switch S0 as in figure 27 (b). When the switch S0 is replaced

by S0-Lr-Cs and Cd in parallel to diode D, we get a voltage mode multi resonant buck

converter as shown in Figure 28. The operation of this circuit is dependent on the values

of L, Cs and Cd. The capacitor Cd increases the number of resonance in the circuit to three.

The equations needed to calculate the resonant components ate in the design section.

Figure 27. Buck Converter (a) Basic circuit structure (b) Steady State Equivalent Circuit

[24]


34

(i) (ii)

Figure 28. ZVS-MRC (i) Multi resonant zero voltage switch, (ii) ZVS-MRC Buck Converter [26]

5.4 Operation of ZVS-MRC Buck Converter

The operation cycle of ZVS MRC is divided into four stages

A. Stage 1: [t0,t1]

At this stage the switch conducts and the current through resonant inductor is less than

output current 𝐼𝑜. The difference of this current 𝐼𝑜 − 𝐼𝐿 flows through the diode 𝐷 [26,

27].

B. Stage 2:[t1,t2]

Resonant Inductor current now reaches till output current 𝐼0, diode 𝐷 is reversed biased

and resonance of 𝐿𝑟 and 𝐶𝑑 begins. The capacitor voltage increases and which results in

zero voltage turn-off of the diode [26, 27].

C. Stage 3:[t2,t3]

Stage 3 ends with the discharge of the capacitor 𝐶𝑑 and diode 𝐷 being forward biased.

At this stage the resonant capacitor and inductor form the series resonant circuit [26].


35

Current through the inductor crosses zero and capacitor voltage reaches peak value and

again starts to decrease and reaches zero.

D. Stage 4:[t3,t4]

The stage starts when the diode is forward biased vD is equal to zero. The resonant

inductor current 𝐼𝑙 increases linearly and flows through the switch. So the switch is on at

zero voltage and the zero current increases till 𝐼0 [26, 27]. The switching cycle ends with

voltage across Cs is equal to zero and switch being turned on

5.5 Design: Converter Schematic and Output

Resonant Buck Converter is designed for following requirements:

Input Voltage Range Vd : 5V − 8V

Output Voltage Vout: 3V

Load Resistance Rl: 10 Ω

Switching frequency fsw = 1MHz

Output Voltage ripple ΔVout = 0.02Vp−p

Solution:

Assumption:

Output current ripple ΔIl is assumed to be from 30% of the load current

Buck converter is designed at worst case scenario i.e. at lowest duty cycle

I. Calculation of Inductor and Capacitance values of Power Stage:

Iout =Vout

Rl 5.1


36

Switching time for each cycle Tsw =1

fsw 5.2

Duty cycle D =Vout

Vin 5.3

Calculation of output filter components of Buck converter

𝐿𝑓 =(𝑉𝑖𝑛−𝑉𝑜𝑢𝑡)𝐷𝑇

𝑠𝑤

𝛥𝐼𝑙 5.4

𝐶𝑓 =𝛥𝐼𝑙𝑇𝑠𝑤

8𝛥𝑉𝑜𝑢𝑡 5.5

Figure 29 shows the schematic of the power stage.

Figure 29. Schematic of Power Stage

II. Resonant Circuit Calculations

Zn = √𝐿

𝐶𝑠 (Characteristic Impedance) 5.6

ZnIout > Vd 5.7

Zn >Vd

Iout 5.8

r ≤ D 5.9

Where r = Rl

Zn 5.10

resonant frequency fr =3(1+π)fsw

4π(1−Dmin) 5.11

𝐷𝑚𝑖𝑛𝑖𝑠 𝑡ℎ𝑒 𝑚𝑖𝑛𝑖𝑚𝑢𝑚 𝑑𝑢𝑡𝑦 𝑐𝑦𝑐𝑙𝑒 𝑓𝑜𝑟 𝑡ℎ𝑒 𝑑𝑒𝑠𝑖𝑔𝑛.

Calculation of resonant inductor and capacitors value of the resonant switch

L =Zn

2πfr 5.12


37

L < Lf (Condition should be satisfied) 5.13

Cs =1

2πfrZn 5.14

𝐶𝑑𝐶𝑠

⁄ > 1 5.15

The schematic of the resonant circuit is given in the figure 30 and table 5 shows the

values of the components that are required to design ZVS-MRC.

Figure 30. Resonant circuit Schematic

Table 5. ZVS-MRC Buck Converter components

Component Value Industry

Specification

Inductance Lf 20.83μH 22 μH

Inductance L 3.01μH 3 μH

Capacitance Cf 562.5nF 560nF

Capacitance Cs 3.36nF 3.6nF

Capacitor Cd 10.8nF 11nF

Slew Rate of

Camparator > 17V/μs


38

Figure 31 shows the complete ZVSMRC with all the components values. Figure 32 shows

the ZVS-MRC in Cadence. The dimensions of the MOSFETs for ZVS-MRC is given in table

6. The block diagram of the complete circuit with fed back is shown in figure 33 and

figure 34 shows the complete circuit in Cadence.

Figure 31. ZVS-MRC Buck

Table 6. ZVS-MRC Components Width and Length

Component Width Length Number of parallel

components

N0 50μ 1μ 100

N1 50μ 1μ 100

N2 50μ 1μ 300

The MOSFET which is used as switch was operating to give desired results when the width

was selected as 1300μm. But to make the converter size small this width was reduced to


39

50μm and number of parallel components were increased to 300. The width of 1.3mm

would increase the size of the converter to mm range. But with the use of the parallel

components the size of the whole converter is still in the μm range. This is verified by the

layout of the complete system. The diodes were also giving desired results at 1200μm

width. This increased width of the components helps in better current and voltage

handling. The above width and length were selected to use common centroid method for

the layout design to further decrease the area of the chip.


40

Figure 32. ZVS-MRC Buck in Cadence


41

Figure 33. Block Diagram

Figure 34. Schematic of ZVS-MRC Buck


42

Figure 35. Output Voltage at 5, 6, and 8V input voltages

Figure 36. Gate Voltage of MOSFET and Voltage across MOSFET

The output of the converter is shown in figure 35 at three different input voltages. Figure

36 shows the gate voltage of the MOSFET used as switch and the voltage across the

switch. It can be seen that when the switch is turned on the voltage across the switch is

below zero and the switch is turned on at zero voltage. The circuit goes into resonance


43

during turn off and the voltage stress on the switch is more than that of conventional

converter due to the resonant elements across the switch.


44

Chapter 6: Complete Circuit of Astable Multivibrator and Resonant Buck Converter

6.1 Results

The figure 37 given below shows the complete circuit for the ZVS-MRC Buck converter

with on chip signal generator. The signal is generated from the astable Multivibrator

shown in figure 38 is given to the comparator for the generation of PWM signal. And this

PWM signal is given to the resonant switch. The frequency of the signal is 1MHz.


45

Figure 37. Complete System in Cadence


46

Figure 38. Output of Astable Multivibrator

Some of the components for the converter are very large in values so they cannot be built

on chip. So these components have to be off chip. So, to do so it requires input/output /

input-output ports to do so. Figure 39 shows the ports that are needed to implement it

and the figure 40 gives the proposed chip.


47

Figure 39. System Schematic after placing Pins


48

Figure 40. System symbol

Figure 41. Output of the complete system

The curve in the deep blue color colour is the output obtained when the input voltage

was 5V. Yellow curve was obtained then the input voltage was 6V and the pink was

obtained when the input voltage was 8V.


49

Figure 42. Output from -20°C to 100°C temperature range

The output of the converter if shown in figure 41 after the implementation of the on chip

signal generator. The figure 42 shows the output of the converter at the worst case when

the converter is operated at the lowest duty cycle, here the operating temperature was

done in the range of -20°C to 100°C with a step size of 20°C. The colour graphs obtained

in figure is for the different temperatures i.e., -20°C, 0°C, 20°C, 40°C, 60°C, 80°C and

100°C. From the above two graphs it can be said that the output voltage is constant at

all the different conditions. The layout of the complete chip is shown in figure 43. The

dimension of the chip is 983.4μm X 879.15μm.


50

Figure 43. System Layout


51

Table 7. Results

Conventional Buck converter was designed to operate at 1MHz frequency in AMI06

technology to compare the results. As it can be seen from the table 7 that for the resonant

switching converter there is an increase in the stress on the Switch, but at high

frequencies the switching losses are the more dominant than any other types of losses.

So the reduction in the switching losses leads to increase in the efficiency of the converter.


52

Chapter 7: Future Work: Wireless Power Transfer

7.1 Introduction

There is research going on in the field of wireless power transfer so as to replace the wire,

to power handheld, consumer electronic devices. In the design of wireless power transfer

the concept of coupling of inductor is taken. The power transfer of dc power takes place

from primary to secondary side takes place. The design takes into consideration of 100V

into the primary side and transfers it to the secondary side. The output voltage is passed

through LC filter to remove ripple. Previous literatures focused on the use of power

transfer by AC voltage sources [33-38]. But the portable equipments work mostly on the

DC voltages hence it required an additional rectifier circuit to convert AC voltage into DC

voltage, hence increases the cost of production.

Wireless power transfer is made by combining the Astable Multivibrator, MOSFET and an

inductor. Coupling factor is taken as 0.1 for as distance of 75cm between the coils as

shown in figure 44.

Figure 44. Relation between coupling factor and distance of separation of coils [39]


53

7.2 Proposed Model

Astable Multivibrator is operated at 8.13 KHz frequency and the output of the Astable

Multivibrator is given to the gate of the MOSFET which works as a switch. Inductor is used

to transfer the power to the secondary side using the principle of mutual coupling of

inductor.

(i)

(ii)

Figure 45. (i) Wireless power transfer circuit (ii) Output voltage of Astable Multivibrator and output voltage on secondary side


54

Figure 45 shows the schematic of Astable Multivibrator, output of the Multivibrator

given to the gate of the switch and the output voltage of 7V seen across the load

resistance as seen across load resistor


55

Chapter 8: Conclusion

The DC-DC converter was designed using Zero Voltage Multi Resonant technique. A

conventional and the ZVS-MRC converters were designed and their efficiency were

compared. The switching losses for ZVS scheme was reduced but the voltage stress on

the MOSFET was increased but since the switching was taking place at higher frequencies

hence the voltage stress on the switch has negligible negative effect on the overall

performance of the system. The switching losses are decreased in ZVS scheme so the

efficiency of the system increases as switching losses are dominant at high frequencies.

The Astable Multivibrator was modified from BJTs to MOSFETs as switching device

because the MOSFET can operate at higher frequencies. The Astable Multivibrator was

designed to produce triangular wave ≈1MHz to be fed into the comparator for PWM

generation.

The comparator designed was able to detect the change at its input terminals of 30 𝑉𝜇𝑠⁄

and produce PWM waves accordingly.

The wireless power transfer model proposed is designed to transfer DC voltage of 7V for

a distance of 75cm. The range can be further enhanced by use of repeaters between

transmitting coil and receiving coil.


56

The future work is to improve this design and achieve efficiency ≈95% and reduce the

voltage stress on the switch. The future work also includes to reduce the size of the

converter further by implementing it on 45nm technology.


57

References

[1]. Power Electronics and Drives (Version 3-2003) by Dr. Zainal Salam, UTM-JB

[2]. https://en.wikibooks.org/wiki/Power_Electronics

[3]. https://en.wikipedia.org/wiki/DC-to-DC_converter

[4]. https://en.wikipedia.org/wiki/Buck_converter

[5]. Switch Mode Power Supplies by Christophe P. Basso

[6]. http://www.radio-electronics.com/info/power-management/switching-mode-

power-supply/smps-buck-switching-regulator-concept-05.gif

[7]. Power Electronics – Converters, Applications and Design by Ned Mohan, Tore M.

Undeland and Williams P. Robbinson

[8]. http://i.stack.imgur.com/ctDSU.png

[9]. http://www.cadence.com/cadence/Pages/default.aspx

[10] http://www.eit.lth.se/cadsys/chapter1.pdf

[11] http://www.eda.ncsu.edu/wiki/Tutorial:CDK_Layout_Tutorial

[12]. http://www.eda.ncsu.edu/wiki/MOSIS_Layers#Layers_that_correspond_to_op

tional_technology_features:

[13]. https://www.mosis.com/vendors/view/on-semiconductor/c5

[14]. AMI C5N Process Design Rules by Z. Tao and M. Keramat, UNIVERSITY OF

CONNECTICUT

[15]. https://en.wikipedia.org/wiki/Multivibrator#Multivibrator_frequency

https://en.wikibooks.org/wiki/Power_Electronics

https://en.wikipedia.org/wiki/DC-to-DC_converter

https://en.wikipedia.org/wiki/Buck_converter

http://www.radio-electronics.com/info/power-management/switching-mode-power-supply/smps-buck-switching-regulator-concept-05.gif

http://www.radio-electronics.com/info/power-management/switching-mode-power-supply/smps-buck-switching-regulator-concept-05.gif

http://i.stack.imgur.com/ctDSU.png

http://www.cadence.com/cadence/Pages/default.aspx

http://www.eit.lth.se/cadsys/chapter1.pdf

http://www.eda.ncsu.edu/wiki/Tutorial:CDK_Layout_Tutorial

http://www.eda.ncsu.edu/wiki/MOSIS_Layers#Layers_that_correspond_to_optional_technology_features

http://www.eda.ncsu.edu/wiki/MOSIS_Layers#Layers_that_correspond_to_optional_technology_features

https://www.mosis.com/vendors/view/on-semiconductor/c5

https://en.wikipedia.org/wiki/Multivibrator#Multivibrator_frequency


58

[16]. Liu, Kwang-Hwa; Lee, F.C.Y., "Zero-voltage switching technique in DC/DC

converters," Power Electronics, IEEE Transactions on , vol.5, no.3, pp.293,304, Jul

1990

[17]. Ting-Ting Song; Huai Wang; Chung, H.S.-H.; Tapuhi, S.; Ioinovici, A., "A High-

Voltage ZVZCS DC--DC Converter With Low Voltage Stress," Power Electronics,

IEEE Transactions on , vol.23, no.6, pp.2630,2647, Nov. 2008

[18]. Xinke Wu; Xiaogao Xie; Chen Zhao; Zhaoming Qian; Zhao, Rongxiang, "Low

Voltage and Current Stress ZVZCS Full Bridge DC–DC Converter Using Center

Tapped Rectifier Reset," Industrial Electronics, IEEE Transactions on , vol.55, no.3,

pp.1470,1477, March 2008

[19]. Jung-Goo Cho; Ju-Won Baek; Chang-Yong Jeong; Geun-Hie Rim, "Novel zero-

voltage and zero-current-switching full-bridge PWM converter using a simple

auxiliary circuit," Industry Applications, IEEE Transactions on , vol.35, no.1,

pp.15,20, Jan/Feb 1999

[20]. Eung-Ho Kim; Bong-Hwan Kwon, "Zero-Voltage- and Zero-Current-Switching

Full-Bridge Converter With Secondary Resonance," Industrial Electronics, IEEE

Transactions on , vol.57, no.3, pp.1017,1025, March 2010

[21]. Fuxin Liu; Jiajia Yan; Xinbo Ruan, "Zero-Voltage and Zero-Current-Switching

PWM Combined Three-Level DC/DC Converter," Industrial Electronics, IEEE

Transactions on , vol.57, no.5, pp.1644,1654, May 2010


59

[22]. Woo-Jin Lee; Chong-Eun Kim; Sang-Kyoo Han; Gun-Woo Moon, "A New High

Efficiency Phase Shifted Full Bridge Converter for Sustaining Power Module of

Plasma Display Panel," Power Electronics Specialists Conference, 2005. PESC '05.

IEEE 36th , vol., no., pp.2630,2634, 16-16 June 2005

[23]. Kyung-Hwa Park; Kang-Hyun Yi, "A Wide Load Range ZVS Dual Half-Bridge

Converters for Large-Sized PDP Sustaining Power Modules," Display Technology,

Journal of , vol.11, no.1, pp.86,96, Jan. 2015

[24]. Rashid, Muhammad H... Power Electronics Handbook Devices, Circuits and

Applications = Edited by Muhammad H. Rashid, Amsterdam: Elsevier, 2011. Print.

[25]. Hui, S. Y.(Ron), and Henry S.h. Chung. "Resonant and Soft-switching Converters."

Power Electronics Handbook (2007): 405-49. Web.

[26]. Tabisz, W.A.; Lee, F.C., "Zero-voltage-switching multi-resonant technique-a

novel approach to improve performance of high frequency quasi-resonant

converters," Power Electronics Specialists Conference, 1988. PESC '88 Record.,

19th Annual IEEE , vol., no., pp.9,17 vol.1, 11-14 April 1988

[27]. Patni, Purvi. Integrated Power Supply. Thesis. Texas Tech University, 2012. N.p.:

n.p., n.d. Web. 9 June 2015. <http://repositories.tdl.org/ttu-

ir/handle/2346/47507>.

[28]. https://en.wikipedia.org/wiki/Operational_amplifier

[29]. "Design of Analog CMOS Integrated Circuits 1st Edition." Design of Analog CMOS

Integrated Circuits: Behzad Razavi: 9780072380323: Amazon.com: Books. N.p.,

https://en.wikipedia.org/wiki/Operational_amplifier


60

n.d. Web. 15 Feb. 2016. <http://www.amazon.com/Design-Analog-CMOS-

Integrated-Circuits/dp/0072380322>.

[30]. http://www.learnabout-electronics.org/Amplifiers/amplifiers64.php

[31]. http://www.edaboard.com/thread330450.html

[32]. http://www.spiceopus.si/imgopt/opt006-01.gif

[33]. Hoang, N.K.; Sang-Gug Lee, "A method of regulating wireless power transfer

based on the analysis of power communication," in Emerging Technologies:

Wireless Power (WoW), 2015 IEEE PELS Workshop on , vol., no., pp.1-3, 5-6 June

2015

[34]. Awai, I.; Ishizaki, T., "Superiority of BPF theory for design of coupled resonator

WPT systems," in Microwave Conference Proceedings (APMC), 2011 Asia-Pacific ,

vol., no., pp.1889-1892, 5-8 Dec. 2011

[35]. Bodrov, Alexey, and Seung-Ki Sul. "Analysis of Wireless Power Transfer by

Coupled Mode Theory (CMT) and Practical Considerations to Increase Power

Transfer Efficiency." Wireless Power Transfer - Principles and Engineering

Explorations (2012): n. pag. Web.

[36]. Matias, R.; Cunha, B.; Martins, R., "Modeling inductive coupling for Wireless

Power Transfer to integrated circuits," in Wireless Power Transfer (WPT), 2013

IEEE , vol., no., pp.198-201, 15-16 May 2013

[37]. Kato, Masaki, Takehiro Imura, and Yoichi Hori. "Study on Maximize Efficiency

by Secondary Side Control Using DC-DC Converter in Wireless Power Transfer via

http://www.learnabout-electronics.org/Amplifiers/amplifiers64.php

http://www.edaboard.com/thread330450.html

http://www.spiceopus.si/imgopt/opt006-01.gif


61

Magnetic Resonant Coupling." 2013 World Electric Vehicle Symposium and

Exhibition (EVS27) (2013): n. pag. Web.

[38]. Tsunekawa, K., "A feasibility study of wireless power transmission system by

using two independent coupled electric fields," in Microwave Workshop Series

on Innovative Wireless Power Transmission: Technologies, Systems, and

Applications (IMWS), 2011 IEEE MTT-S International , vol., no., pp.141-144, 12-13

May 2011

[39]. Kurs, A., A. Karalis, R. Moffatt, J. D. Joannopoulos, P. Fisher, and M. Soljacic.

"Wireless Power Transfer via Strongly Coupled Magnetic Resonances." Science

317.5834 (2007): 83-86. Web.

Zero Voltage Switching Resonant Converter Using 0.6μm ...

Documents

Transcript of Zero Voltage Switching Resonant Converter Using 0.6μm ...